2
“The Inner Circle Guide to Outbound & Call Blending”
© ContactBabel 2016
Please note that all information is believed correct at the time of publication, but ContactBabel does
not accept responsibility for any action arising from errors or omissions within the report, links to
external websites or other third-party content.
3
CONTENTS
Contents ............................................................................................................................................... 3
List of Tables ......................................................................................................................................... 6
Introduction .......................................................................................................................................... 9
Background to Outbound ................................................................................................................... 11
The Role of Outbound in Businesses Today .................................................................................... 12
Inbound & Outbound Activity ..................................................................................................... 12
The Nature of Outbound Activity ................................................................................................ 16
Outbound: Vertical Markets ....................................................................................................... 18
Outbound: Contact Centre Size................................................................................................... 20
Drivers of Outbound Automation ....................................................................................................... 21
Efficiency & Productivity ................................................................................................................. 21
End-user question #1: ................................................................................................................. 22
Apart from increasing the number of outbound calls being made, what benefits can automated
outbound solutions provide? ...................................................................................................... 22
Flexibility ......................................................................................................................................... 23
Customer Experience ...................................................................................................................... 23
Agent Engagement.......................................................................................................................... 24
HR Factors in Outbound ...................................................................................................................... 25
Salaries and Bonuses for Outbound Agents .................................................................................... 25
Agent Attrition ................................................................................................................................ 27
Agent Absence ................................................................................................................................ 28
Agent Recruitment Costs ................................................................................................................ 29
Outbound Automation: Implementation and Usage .......................................................................... 30
Outbound Dialling Practices ............................................................................................................ 30
Use of Diallers ............................................................................................................................. 30
Dialler modes .............................................................................................................................. 31
4
Outbound Outcomes .................................................................................................................. 35
Agent Utilisation Rates ................................................................................................................ 39
Dialling activity: time of day ........................................................................................................ 41
Dialling activity: maximum number of attempts ......................................................................... 45
Ring time and connection time ................................................................................................... 47
Dialling activity: the role of mobile ............................................................................................. 48
End-user question #2: ................................................................................................................. 49
We’re interested in increasing productivity, but concerned about agent burnout and increased
attrition. Do you have any pointers on best practice? ................................................................ 49
Answer Machine Detection ......................................................................................................... 50
Use of CLI / CLID .......................................................................................................................... 53
The Use of IVM (Interactive Voice Messaging)............................................................................ 54
Use of broadcast messaging ....................................................................................................... 55
Compliance with Regulations .......................................................................................................... 56
Dialling activity: TPS .................................................................................................................... 58
End-user question #3: ................................................................................................................. 59
What impact does current and future legislation have on outbound, and how do your solutions
address this? ............................................................................................................................... 59
Complying with Regulations ........................................................................................................... 60
Call Blending ................................................................................................................................... 62
Multimedia Blending ................................................................................................................... 64
End-user question #4: ................................................................................................................. 65
What benefits are there from call blending? What effects can we expect to see on productivity
and agent engagement? ............................................................................................................. 65
Multichannel Outbound Strategies and Techniques ....................................................................... 66
End-user question #5: ................................................................................................................. 68
Are outbound and call blending solutions separate from other multichannel interaction, or has
there been any move to integrate these channels together? ..................................................... 68
5
The Role of Scripting ....................................................................................................................... 69
Dynamic Scripting ....................................................................................................................... 69
End-user question #6: ................................................................................................................. 74
Have you noticed any changes in how scripting is being used? How do businesses tend to use
this? ............................................................................................................................................ 74
Call-Back & Call-Me requests .......................................................................................................... 75
Outbound in the Cloud ................................................................................................................... 85
End-user question #7: ................................................................................................................. 87
Has there been much uptake of cloud-based solutions? Is this mainly down to cost, or are there
other benefits as well? ................................................................................................................ 87
The Future of Outbound ..................................................................................................................... 88
End-user question #8: ................................................................................................................. 90
Are you seeing a change in the nature of outbound calling? How are your solutions developing
to address this and what can we expect to see in terms of future functionality from outbound
solutions? .................................................................................................................................... 90
About Rostrvm Solutions .................................................................................................................... 91
About ContactBabel ............................................................................................................................ 92
6
LIST OF TABLES
Figure 1: Inbound and outbound activity 2002-2015 .......................................................................... 12
Figure 2: Inbound and outbound activity by contact centre size, end-2015 (UK) ............................... 13
Figure 3: Outbound agent positions (equivalent), by contact centre size, 2015 (UK) ......................... 13
Figure 4: Outbound agent positions (equivalent), by vertical market, 2015 (UK) ............................... 15
Figure 5: Outbound activity (UK) ........................................................................................................ 16
Figure 6: Use of outbound calling, by vertical market ........................................................................ 18
Figure 7: Outbound activity by vertical market ................................................................................... 19
Figure 8: Use of outbound calling, by contact centre size................................................................... 20
Figure 9: Outbound activity by contact centre size ............................................................................. 20
Figure 10: Salaries by contact centre activity type (UK) ...................................................................... 25
Figure 11: Annualised bonuses paid to outbound agents ................................................................... 26
Figure 12: Annualised basic salaries & bonuses paid to outbound agents (mean/median/1st & 3rd
quartiles) ............................................................................................................................................. 26
Figure 13: Bonuses paid to outbound agents as a proportion of basic salary (mean/median/1st & 3rd
quartiles) ............................................................................................................................................. 26
Figure 14: Mean and median agent attrition by contact centre activity type (UK) ............................. 27
Figure 15: Historical mean agent attrition by contact centre activity type (with 2018 projection) –
(UK) ..................................................................................................................................................... 28
Figure 16: Short-term absence by contact centre activity type (UK) .................................................. 28
Figure 17: Recruitment costs by contact centre activity ..................................................................... 29
Figure 18: Use of automated outbound diallers, by contact centre size (UK) ..................................... 30
Figure 19: Use of dialling modes, by contact centre size (outbound operations only) ....................... 32
Figure 20: Use of dialling modes, by outbound activity (outbound operations only) ......................... 33
Figure 21: Proportion of calls answered, by outbound activity type .................................................. 35
Figure 22: Proportion of calls answered, by size ................................................................................. 35
Figure 23: Of the calls that were answered, how many were by a consumer? (by outbound activity
type).................................................................................................................................................... 36
7
Figure 24: Of the calls that were answered, how many were by a consumer? (by contact centre size)
............................................................................................................................................................ 36
Figure 25: Of the calls that were answered by a consumer, how many were hung up by the consumer
when an agent was available? (by outbound activity type) ................................................................ 37
Figure 26: Of the calls that were answered by a consumer, how many had a recorded message is
played as no agent was available (by outbound activity type) ............................................................ 37
Figure 27: Of the calls that were answered by an answerphone, where a recorded message is played,
by outbound activity type ................................................................................................................... 38
Figure 28: Average outbound agent utilisation rates .......................................................................... 39
Figure 29: Cumulative opening and closing hours, Mon-Fri ................................................................ 41
Figure 30: Cumulative opening and closing hours, Sat ....................................................................... 42
Figure 31: Cumulative opening and closing hours, Sun....................................................................... 43
Figure 32: Call attempts by activity type (debt collection) .................................................................. 45
Figure 33: Call attempts by activity type (warm calling – sales calls to existing customers) ............... 46
Figure 34: Call attempts by activity type (cold calling – sales calls to new prospects) ........................ 46
Figure 35: Minimum mean and median ring time before call termination, by outbound activity type
............................................................................................................................................................ 47
Figure 36: % of outbound dialling carried out to mobile phones ........................................................ 49
Figure 37: Estimation of AMD accuracy .............................................................................................. 51
Figure 38: Use of answer machine detection (AMD), by contact centre size (UK) – only those using
automated outbound solutions .......................................................................................................... 52
Figure 39: Methods of checking TPS (n=91) ........................................................................................ 58
Figure 40: Use of call blending by contact centre size ........................................................................ 62
Figure 41: Average speed to answer, by call blending environment type........................................... 63
Figure 42: Use of automated outbound communication .................................................................... 66
Figure 43: Use and type of scripting, by vertical market ..................................................................... 70
Figure 44: Use and type of scripting, by contact centre size ............................................................... 71
Figure 45: The effectiveness of scripting............................................................................................. 72
Figure 46: Inhibitors to scripting ......................................................................................................... 73
8
Figure 47: Reasons given for dislike of contact centre queuing .......................................................... 75
Figure 48: Types of telephony call-back offered to customers (only from respondents offering
telephony call-back) ............................................................................................................................ 77
Figure 49: Use of website and queue call-back options and queue position announcements, by
contact centre size .............................................................................................................................. 78
Figure 50: Triggers for offering call-back functionality ....................................................................... 79
Figure 51: Typical queue time before call-back is offered .................................................................. 80
Figure 52: Proportion of customers in telephony queue accepting offer of call-back ........................ 81
Figure 53: Effectiveness of telephony call-back functionality ............................................................. 82
Figure 54: Effectiveness of website call-back functionality ................................................................. 83
Figure 55: Dialler deployment model, by contact centre size ............................................................. 86
9
INTRODUCTION
“The Inner Circle Guide to Outbound and Call Blending” is the 8th in the Inner Circle series of
ContactBabel reports. Other subjects include Cloud-based Contact Centres, Multichannel, Self-
Service, Interaction Analytics and PCI DSS Compliance, and can be downloaded free of charge from
here.
The Inner Circle Guides are a series of analyst reports investigating key customer contact solutions.
The Guides aim to give a detailed and definitive view of the reality of the implementing and using
these technologies, an appraisal of the vendors and products available and a view on what the
future holds.
The Inner Circle Guides are free of charge to readers. Research and analysis costs are borne by
sponsors - solution providers in the specific area of study - whose advertisements, case studies and
thought leadership pieces are included within these Guides.
Solutions providers have not had influence over editorial content or analyst opinion, and readers can
be assured of objectivity throughout. Any vendor views are clearly marked as such within the report.
As well as explaining these solutions to the readers, we have also asked the potential users of these
solutions whether they have any questions or comments to put directly to solution providers, and
we have selected eight of the most popular to ask. These branded Q&A elements are distributed
throughout the report and give interesting insight into real-life issues.
NB: statistics within this report refer to the UK industry, unless stated otherwise. There is a version
of this report available for download with equivalent US statistics.
“Small” contact centres are defined in the report as having 50 or fewer agent positions; “Medium”
51-200 agent positions; and “Large” 200+ agent positions.
Inbound Response
From simple call queuing to multi-skilled, multi-channel contact handling – deliver great service where and when it’s needed. Powerful routing means that contacts are handled gracefully and delivered to the right person at the right time.
Call us on 01483 494 690Email us on [email protected] rostrvm.com
Outbound Contact
Hit your targets and keep promises to customers with Ofcom compliant blended predictive, progressive and preview dialling. Delight customers with proactive service across multiple channels.
Precision contact
In Control
Real-time and historical management information delivered to your connected devices as you need it. Compliant voice recording, system configuration, adds, moves and changes are all in your control from wherever you are.
Agents Anywhere
Say and do the right things with agents anywhere – in the office, at home, on the move. . . scripting, note taking and call outcome capture supported by CallGuide for an effective, compliant and measured customer experience.
11
BACKGROUND TO OUTBOUND
The traditional outbound call was simply about selling more products to new and existing customers.
However, legislation and customer pressure impacted on cold calling, and the past years have seen
an increasing proportion of outbound calling being made to existing customers, either to deliver
customer care or to inform them proactively about events and circumstances which affect them, as
well as providing call-back functionality. Debt collection also still forms a key part of the outbound
industry.
Outbound calling is fundamentally different from inbound, and - facing significant and growing
cultural and legislative issues - must be managed sensitively:
the nature of outbound is intrusive and usually driven by the needs of the business rather
than the customer (except in cases of call-back requests and for proactive outbound service)
this means that customers are more likely to be defensive and wary of the purpose of the
call. Trust needs to be built very quickly in order to overcome this negative start point:
having the right information about the customer to hand will improve the experience for
both agent and customer
outbound work can be very hard on agents: few people actively welcome most outbound
calls, and persistent refusal, lack of interest and rudeness can be very wearing for agents,
especially if productivity-enhancing technology such as diallers are being used over-
aggressively. Management should consider ways of alleviating agent stress, through sensible
scheduling and call blending, judicious use of technology, focused training and improving
working environments, amongst other ways
especially where the technology exists to do so, it can be tempting to treat outbound calling
campaigns as an exercise in maximising call volumes and (theoretically) revenues. However,
this can result in brand damage and high staff attrition rates through over-pressured and
exhausted agents delivering poorer quality interactions
there has been a tendency to use offshore contact centres for low-value outbound sales
campaigns which would otherwise be unprofitable to run. However, the same high
standards of training and support are needed by offshore agents to do their job properly:
too many businesses simply put the agents on a dialler with an inflexible script in front of
them and then wonder why their customers and prospects become negative towards their
brand
tough legislation has emerged which is reducing the amount of cold calling which businesses
can do. Cold calling is illegal in Germany, and the Do-Not-Call register in the US and the
Telephone Preference Scheme (TPS) in the UK allow customers to opt out of receiving any
sales calls at all in theory.
12
THE ROLE OF OUTBOUND IN BUSINESSES TODAY
INBOUND & OUTBOUND ACTIVITY
Since 2002, the reported outbound telephony activity increased from 29.8% to 33.7% in 2004, and
back to 32.9% in 2005, a high-water mark in the importance of outbound in the UK contact centre
industry.
In 2006, the figure fell to 29.1%, which was a decrease of 3.5% in actual numbers of outbound calls,
the first time that outbound calling from the UK has actually fallen, most likely as a result of a
reduction in cold calling from UK operations, driven by cost and legislation. Since then, the
proportion of work that was outbound declined quite consistently year on year until 2011.
Since 2011, there has been a slight rise in the importance of outbound and a stable point seems to
have been reached.
Figure 1: Inbound and outbound activity 2002-2015
70.2%66.7% 66.3% 67.1%
70.9% 71.0% 71.9%74.6%
79.3% 79.7% 79.3%76.9% 78.3% 77.7%
29.8%33.3% 33.7% 32.9%
29.1% 29.0% 28.1%25.4%
20.7% 20.3% 20.7%23.1% 21.7% 22.3%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Inbound & Outbound Activity 2002-2015
Inbound
Outbound
13
Figure 2: Inbound and outbound activity by contact centre size, end-2015 (UK)
Agent positions
% of inbound activity % of outbound activity
10-25 agent positions 73.5% 26.5%
26-50 agent positions 75.0% 25.0%
51-100 agent positions 81.0% 19.0%
101-250 agent positions 80.1% 19.9%
251-500 agent positions 77.3% 22.7%
501-1,000 agent positions 74.2% 25.8%
1,000+ agent positions 79.5% 20.5%
Mean average
77.7% 22.3%
NB: the term “outbound agent positions (equivalent)” describes the number of exclusively-
outbound, full-time agents required. In practice, not all outbound work is done by outbound-only
agents, hence the need for “outbound agent equivalents”.
Figure 3: Outbound agent positions (equivalent), by contact centre size, 2015 (UK)
Size band (agent positions)
Outbound agent positions (equivalent)
10-25 agent positions 9,010
26-50 agent positions 20,313
51-100 agent positions 18,483
101-250 agent positions 35,223
251-500 agent positions 26,105
501-1,000 agent positions 37,926
1,000+ agent positions 21,320
Total
168,380
14
In past years, the finance vertical market had the largest number of outbound agents, involved in
debt collection, persuading customers to change financial products (e.g. credit cards) and
increasingly, cross-selling and up-selling to existing customers. Businesses are aware that one of the
key moves towards increased profitability is to get customers purchasing multiple products, e.g. a
personal loan, a current account, a credit card and insurance from the same provider. However, the
sector is seeing an increased use of cross-selling on inbound calls, as well as direct mail and web-
based marketing, rather than an increase in outbound telephony, with offshore playing some role in
outbound as well. The sub-sector involved in calling about mis-sold products such as payment
protection insurance (PPI) is in part based offshore, and is not included in these figures.
The outsourcing and telemarketing sector is a key part of the UK’s outbound activity and is the
largest exponent of outbound calling, with campaign-based outbound for sectors such as finance,
telecoms and utilities still very important. It also does a significant amount of client satisfaction
checking and market research, all of which are outbound activities.
The services sector carries out a high proportion of outbound activity, especially around sales, as is
the case with the communications sector.
The retail & distribution sector uses outbound as a sales tool, calling existing customers, as well as
providing information about deliveries. However, the proportion of activity that is outbound is
below the industry average.
There is a low level of outbound calling in public services, as many of these operations are non-sales,
reactive helpdesk environments, which answer the public’s queries. As such, proactive outbound
campaigns are rarely needed.
Printing & publishing has a higher level of outbound activity than most other vertical markets,
caused by the large number of calls made by newspapers, magazines and directories in order to sell
advertising space.
15
Figure 4: Outbound agent positions (equivalent), by vertical market, 2015 (UK)
Vertical market
Proportion of outbound agent equivalents
Outbound agent positions
(equivalent)
Outsourcing and Telemarketing 39% 37,050
Finance 21% 27,930
Communications 25% 26,250
Services 28% 20,720
Retail and Distribution 15% 14,850
IT 19% 9,310
Manufacturing 30% 8,400
Public Services 9% 5,760
Transport and Travel 14% 4,340
Printing and Publishing 47% 4,230
Utilities 9% 2,880
Medical 22% 2,640
Motoring 19% 1,995
Food and Drink 16% 1,600
Engineering and Construction 10% 425
Total
22.3% 168,380
In the past, there have been a large number of smaller contact centres which performed market
research and outsourced telemarketing activities on behalf of other clients. (Large outsourcing
companies are likely to be heavily involved in inbound client care as well as outbound campaigns).
The past few years have put great pressure on the profit margins of these types of 'commodity'
contact centres, with the result that a significant proportion have ceased trading, or at least,
reduced their headcount considerably. Furthermore, the effects of the Telephone Preference Service
(TPS), which also allows businesses to opt out of receiving telemarketing calls, puts pressure on the
outbound sector. As take-up of the service has grown (around two-thirds of UK households are
registered with the TPS), businesses which use cold-call telemarketing have been left with two
decisions: to continue their existing call-per-prospect activity (thus reducing the overall number of
calls they make, as more people opt-out), or make more calls per prospect, keeping the overall
volume up. The latter option may well alienate more people through pestering them with too many
calls, which may drive up the numbers registering on the TPS, making the pool of available prospects
even smaller. The Privacy and Electronic Communications (EC Directive) Regulations, which has been
in force since the end of 2003, banned all unsolicited email and SMS messages to individuals, cutting
businesses’ direct marketing opportunities at a stroke. In previous years, there had been limited
fines for nuisance or silent calls, but the ICO and Ofcom are actively looking to improve compliance
through greater powers and fines.
16
THE NATURE OF OUTBOUND ACTIVITY
For a number of years, the single most popular outbound activity been proactive customer service -
a strong brand builder as well as an effective inbound call avoidance tactic.
Having said that, the overall proportion of sales calls grew from 31% in 2014 to 38% in 2015,
although making sales calls to potential new customers continues its decline to 12% in 2015 (it was
24% in 2013), demonstrating that cold calling is not the force that it used to be.
Cross-selling/upselling is an important outbound activity (and bear in mind that this figure does not
include those many inbound service calls that are turned into cross-selling opportunities), with 14%
of outbound calls being made for this purpose.
Renewals calls account for 12% of outbound activity, and debt collection makes up 10% of the
industry’s outbound calls. Customer satisfaction surveys remain at levels last seen before the
economic downturn.
Sales to both new and existing customers are obviously still key reasons why companies carry out
outbound calls, and the hybrid method - customer service leading to a cross-sell/up-sell opportunity
- is seen a good way of circumventing the increasing numbers of people joining TPS. However,
businesses must be careful not to pester customers or abuse the relationship they have built up with
frequent calls about products and services that are not tailored to the customer. Increasingly,
turning an inbound service call into a cross-sell or upselling opportunity has become a widely-used
tactic.
Figure 5: Outbound activity (UK)
Proactive customer service (e.g.
notification of delivery, delays, problems, etc)
25%
Cross-selling or up-selling sales calls to existing customers
14%
Sales calls to potential new
customers (cold calls)12%
Renewals (sales calls to existing
customers)12%
Customer satisfaction surveys
10%
Debt collection10%
Other17%
Outbound activity
17
The ‘Other’ segment accounts for a significant part of the outbound mix, and include the following
areas:
Call-backs (from call-back requests and answerphone messages)
Case management and ongoing customer care
Chasing contractors and suppliers
Fraud investigations
Gathering missing order information
Market research and surveys
Responding to complex customer questions sent via other channels
Return calls where the agent was unable to answer the question on first call
Saving customers that are or are at risk of cancelling contracts
Seeking authorisation from clients
Updating complaint status.
18
OUTBOUND: VERTICAL MARKETS
82% of survey respondents carry out some form of outbound calling, with the TMT (technology,
media and telecoms), transport & travel, utilities and outsourcing & telemarketing sectors leading
the way.
As usual, the public sector lags quite some way behind the rest of the contact centre industry in
terms of its outbound activity, although for once, a majority of the 2015 survey respondents carry
out at least some outbound work.
Figure 6: Use of outbound calling, by vertical market
Vertical market
Proportion of respondents using outbound calling
TMT 95%
Transport & Travel 88%
Utilities 88%
Outsourcing 86%
Services 85%
Insurance 85%
Housing 83%
Retail & Distribution 79%
Finance 77%
Manufacturing 75%
Public Sector 61%
Average
82%
Vertical market patterns of outbound activity are very different from each other, and there is not
even a great deal of homogeneity within sectors, so these figures should be treated with some
caution. However, there are some interesting findings to bring out from the results of our 2015
survey of UK contact centres.
The majority of retail & distribution and utilities, and 44% of the manufacturing sectors’ outbound
activity is proactive customer service, advising of status, delays and deliveries, and providing
information up and down the supply chain, with the transport and travel and housing sectors are
also very involved in this.
Outsourcing and transport & travel respondents all report cross-selling and upselling being a
significant part of their outbound activity.
19
The insurance, TMT and outsourcing sectors are the hungriest for new business and most likely to
cold-call (within the law, of course), with the former sector possibly driven by the uptake in web-
based sales lead capture via comparison sites, which provide qualified leads to be acted upon
immediately.
Once again, the utilities and housing respondents report very significant outbound activity
connected to debt collection.
The contract-based business models, often found in insurance (17%) and TMT (21%), are amongst
those most likely to be carrying out renewal sales calls to existing customers.
Figure 7: Outbound activity by vertical market
Outbound activity
FS HS INS MN OS PS RD SVC TMT TT UT Avg
Sales calls to potential new customers (cold calls)
11% 0% 16% 8% 22% 6% 2% 11% 18% 0% 15% 12%
Cross-selling or up-selling sales calls to existing customers
14% 0% 17% 13% 27% 1% 12% 7% 19% 43% 2% 14%
Renewals (sales calls to existing customers)
4% 0% 17% 22% 12% 3% 14% 7% 21% 31% 3% 12%
Proactive customer service (e.g. notification of delivery, delays, problems, etc.)
20% 30% 25% 44% 17% 4% 55% 24% 19% 11% 51% 25%
Debt collection 9% 41% 2% 5% 8% 1% 1% 4% 10% 9% 25% 10%
Customer satisfaction surveys
1% 15% 15% 4% 3% 27% 1% 17% 7% 7% 1% 10%
Other 41% 14% 8% 4% 10% 59% 16% 29% 6% 0% 3% 17%
20
OUTBOUND: CONTACT CENTRE SIZE
In 2015, there is little real pattern between contact centre size and the propensity to make
outbound calls: in the past, large contact centres were more likely to do so, but this has not been the
case for a number of years.
Figure 8: Use of outbound calling, by contact centre size
Contact centre size
Proportion of respondents using outbound calling
Small 81%
Medium 87%
Large 78%
Average
82%
A case could be made that variations in outbound activity are more closely linked to the type of
business (i.e. vertical market) than the contact centre size.
However, it is worth noting that sales calls in general are more likely to be made by larger operations
(52% of activity is sales-related, as opposed to only 32% in small contact centres). Correspondingly,
small operations spend far more time on proactively updating the customer.
Figure 9: Outbound activity by contact centre size
Outbound activity
Small Medium Large Average
Sales calls to potential new customers (cold calls) 9% 12% 17% 12%
Cross-selling or up-selling sales calls to existing customers
11% 10% 25% 14%
Renewals (sales calls to existing customers) 12% 10% 15% 12%
Proactive customer service (e.g. notification of delivery, delays, problems, etc.)
33% 19% 19% 25%
Debt collection 10% 12% 8% 10%
Customer satisfaction surveys 9% 14% 6% 10%
Other 16% 24% 11% 17%
21
DRIVERS OF OUTBOUND AUTOMATION
EFFICIENCY & PRODUCTIVITY
Automated dialling solutions, especially in predictive mode, have traditionally been about reducing
the time between the agent making live contacts, by eliminating manual dialling time, the time spent
waiting for the customer to answer the call and eliminating calls passed through to answer phones.
It is interesting to note that industry statistics on dialling mode efficiency suggest the following to be
reasonable estimates, if the list quality and time of day that calls are made were the same in each
instance:
• manual dialling: 12 minutes of talk time per hour (20%)
• preview dialling: 18 minutes of talk time per hour (30%)
• progressive dialling: 25 to 35 minutes of talk time per hour (42% - 58%)
• predictive dialling: 35 to 45 minutes of talk time per hour (58% - 75%).
In reality, each dialling mode will be used in different circumstances, making direct comparison very
difficult. The following section, on implementation and usage, gives an explanation and greater
detail into the advantages and disadvantages of each dialler mode, and how they are typically used.
Productivity is not simply about maximising live contacts: the effectiveness of the results is of course
of the utmost importance. Automated outbound solutions allow the business to analyse the most
effective times of day to call customer segments, tailor their scripts in order to increase positive
outcomes, and even to ascertain whether voice, text or email is the correct channel for the specific
customer or campaign. This management information also identifies the agents’ effectiveness,
identifying training requirements for underperforming agents, and providing opportunities to share
best practice from the top performers.
22
END-USER QUESTION #1:
APART FROM INCREASING THE NUMBER OF OUTBOUND CALLS BEING MADE, WHAT BENEFITS CAN AUTOMATED OUTBOUND SOLUTIONS PROVIDE?
An outbound solution delivers a lot more than simply increasing the number of
outbound calls being made. It also provides the tools to target activity and
deliver the management information needed to measure outbound business
benefits. rostrvm Outbound enables you to contact your audience:
at the right time – the best time to call;
with the right message and channel – modern campaigns mix telephone, text messages and
email;
and the right people – making sure your calls are handled by appropriately-skilled agents.
Outbound solutions routinely collect data such as the number of call attempts that don’t get
answered, connections to live callers etc. They also capture information that will give you insights
into campaign success and whether your contact is proving to be effective.
These solutions not only improve contact rates - they also support agents with integral agent
scripting systems to help them do and say the right thing, capture the relevant information and
provide links to supporting IT systems.
Today’s outbound solutions are highly flexible and user friendly, whatever you use them for, from
following up on abandoned inbound calls, upselling/cross-selling, customer satisfaction surveys etc.
They can ensure that real progress is made and customers feel appreciated.
23
FLEXIBILITY
Call blending - the process by which agents are switched between inbound and outbound queues
depending upon the operational requirements of the contact centre - offers businesses the
opportunity to deploy their resources flexibly to meet service levels. This capability can be enhanced
by automatically providing the relevant customer data to the agent desktop - whether in inbound or
outbound mode - reducing the need for manual searches of information, reducing handle time and
maximising productivity.
In terms of workforce management, real-time adherence and intraday functionality will mean that
the dynamic nature of much contact centre work will not impact upon the accuracy of management
information systems, being able to track and allocate resource to the necessary activity without
requiring significant amounts of manual reallocation.
Over many years of research, ContactBabel has analysed key performance metrics in operations that
use blended agents or dedicated outbound, and has found that there is often a direct correlation
between superior KPIs and a blended environment. This is investigated in more depth within the Call
Blending section of this report.
CUSTOMER EXPERIENCE
The idea of outbound calling being beneficial for the customer experience goes against the old-
fashioned view of outbound as making cold calls to sell home improvements or encourage PPI or
litigation claims.
In fact, the following section on outbound activity shows that proactive outbound service - keeping
customers informed - has grown very strongly over the past few years. Increasing volumes of
outbound calls are being made at the request of the customer, whether through a website ‘call-me’
button, or as a result of a callback request being made while within the inbound telephony queue. In
all cases, these forms of outbound communication are positive for the customer experience, and
when aligned with a judicious mix of non-voice outbound communication, such as email or SMS,
goes a long way to improving the customer experience while deflecting significant volumes of
inbound calls from otherwise-frustrated customers.
The section on call-me and callback outbound activity later in the report gives more detail on the
importance of this variety of outbound in improving the customer experience.
24
AGENT ENGAGEMENT
As well as improving certain key performance metrics, call blending seems to have a positive impact
on agent attrition rates. More information is given in the dedicated Call Blending section of this
report, and it can be hypothesised that a greater variety of work can improve agent engagement and
thus reduce staff attrition. It should also be noted that the outbound sector has experienced a
significant and ongoing drop in agent attrition rates over the past few years, and this may be linked
to a reduction in repetitive, highly-scripted calls, in favour of more personalised customer service
and sales calls.
By removing the tedious elements of manual dialling, outbound automation can improve the agent
experience by providing more live contacts and thus more opportunities to make sales or otherwise
reach performance targets. Of course, overly aggressive use of predictive dialling for long periods
can exhaust agents, which will quickly have a very negative effect on job satisfaction.
25
HR FACTORS IN OUTBOUND
SALARIES AND BONUSES FOR OUTBOUND AGENTS
As they become more senior, outbound contact centre employees have historically tended to get
paid a lower basic salary than their inbound equivalents, mainly because the former will usually
receive a much higher performance-related bonus.
Agents in inbound environments are reported to receive similar basic salaries as those agents more
involved in outbound work.
Figure 10: Salaries by contact centre activity type (UK)
£16,916
£18,130
£24,897
£39,530
£16,827
£18,618
£26,106
£40,082
£16,340
£18,851
£22,809
£35,864
£16,844
£19,043
£24,923
£39,375
£0 £5,000 £10,000 £15,000 £20,000 £25,000 £30,000 £35,000 £40,000 £45,000
New agent
Experienced agent
Team leader
Contact centre manager
Salaries by contact centre activity type
Average
Outbound
Mixed
Inbound
26
83% of outbound staff are paid a bonus, which is often around 10-15% of annual salary.
Figure 11: Annualised bonuses paid to outbound agents
Figure 12: Annualised basic salaries & bonuses paid to outbound agents (mean/median/1st & 3rd quartiles)
Basic annual salary
Annual bonus
1st quartile £18,000 £3,000
Median £16,000 £1,562
3rd quartile £14,798 £600
Mean £16,394 £2,192
Figure 13: Bonuses paid to outbound agents as a proportion of basic salary (mean/median/1st & 3rd quartiles)
Bonus as a proportion of salary
1st quartile 17%
Median 10%
3rd quartile 2%
Mean 13%
None17%
Less than £5007%
£500 - £1,0009%
£1,001 - £2,00029%
£2,001 - £3,00017%
£3,001 - £4,0005%
£4,001 - £5,0006%
Over £5,00010%
Annualised bonuses paid to outbound FTEs
27
AGENT ATTRITION
It is generally acknowledged that outbound customer contact can be a very difficult, high-pressure
job, which leads to stress and burnout, and thus more likely to see high levels of attrition.
Historically, there has been an acceptance that large volumes of outbound calling will often come at
the cost of high staff attrition, and that this is something which just has to be managed, and
outbound respondents will tend to be more geared-up to cope with high staff turnover.
However, recent year’s outbound attrition figures have been much lower than the historical norm,
which may be to do with the reduced amount of cold calling taking place in UK contact centres, with
growing levels of outbound customer care, and ‘warm calling’ being far more prevalent than 10
years ago.
Figure 14: Mean and median agent attrition by contact centre activity type (UK)
20%
26%
29%
22%
16%15%
20%
16%
0%
5%
10%
15%
20%
25%
30%
35%
Inbound Mixed Outbound Average
Mean and median agent attrition by contact centre activity type
Mean Median
28
Between 2007 and 2011, outbound operations saw higher levels of attrition, but the gap between
inbound and mixed/outbound contact centre attrition rates has been much lower since. 2013 and
2014’s figures – indicating little difference between inbound and outbound operations’ attrition
rates – appears less of a statistical anomaly rather than perhaps being something structural, and
although a gap has re-emerged this year, it is not as notable as in the past.
Figure 15: Historical mean agent attrition by contact centre activity type (with 2018 projection) – (UK)
Contact centre activity 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2018
Inbound (75%+ IB) - - - - 33% 30% 18% 14% 19% 17% 16% 17% 20% 21%
Mixed (25-75% IB) - - - - 28% 23% 23% 18% 23% 26% 18% 26% 26% 24%
Outbound (75%+ OB) - - - - 40% 40% 56% 36% 35% 26% 19% 18% 29% 27%
Average 14.5% 14.9% 22.0% 23.0% 32.0% 28.0% 23.9% 16.0% 21.0% 20.0% 16.9% 18.9% 22.3% 24%
AGENT ABSENCE
Conventional wisdom would expect to find that those in high-pressure outbound jobs would have
higher unauthorised absence rates, but this difference has reduced considerably over the years in
line with attrition rates.
Figure 16: Short-term absence by contact centre activity type (UK)
Contact centre activity type
Mean agent absence rate (2015)
Median agent absence rate (2015)
Agent absence rate (2018 projection) – mean (median)
Inbound 6.4% 5.1% 5.2% (4.7%)
Mixed 6.6% 5.2% 5.4% (4.8%)
Outbound 5.9% 4.8% 6.2% (5.8%)
Average 6.4% 5.1% 5.5% (4.9%)
29
AGENT RECRUITMENT COSTS
As usual, outbound operations will tend to spend less to recruit a new agents compared to inbound
or mixed contact centres, with 59% of outbound respondents reporting that they spent less than
£500 per agent.
Figure 17: Recruitment costs by contact centre activity
19% 21%
59%
23%
30%
18%
12%
26%
16%32%
18%
19%
14% 3%
6%
11%
9%
6%
8%
8%
15%
6%
9%
4% 6% 4%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Inbound Mixed Outbound Average
Recruitment costs by contact centre activity
Don't know
Over £2,500
£2,000 - £2,500
£1,500 - £2,000
£1,000 - £1,500
£500 - £1,000
Less than £500
30
OUTBOUND AUTOMATION: IMPLEMENTATION AND USAGE
OUTBOUND DIALLING PRACTICES
USE OF DIALLERS
At the core of most outbound operations is the automated dialler, which can be deployed in
different modes depending upon the requirement of the business, its agents and the type of calls
that are being placed.
Automated outbound diallers are most often found in large operations which carry out reasonable
amounts of outbound work, as the efficiencies over manual dialling are so considerable that it will
often make commercial sense.
Outbound automation in the cloud is becoming increasingly widely-used, and this means the barriers
to usage are even less, with smaller operations also showing increased interest.
Although dialler usage in respondents from small UK contact centres is currently only 9%, a further
7% are actively planning implementation within the next 12 months. Many suppliers of this
technology are able to offer low-cost, scalable functionality in the cloud, and we would expect this
figure to increase considerably in the next couple of years.
Figure 18: Use of automated outbound diallers, by contact centre size (UK)
Looking at the type of outbound activity carried out, those involved in debt collection are most likely
to use automated diallers, and those involved in customer service the least.
8%
23%
49%
23%1%
10%
10%
6%
7%
8%
10%
8%
3%
4%
2%
3%
73%
46%
27%
53%
8% 10%2%
7%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Small Medium Large Average
Use of automated outbound diallers, by contact centre size
Don't know / NA
No plans to implement
Will implement after 12 months
Will implement within 12 months
Use now, looking to replace/upgrade
Use now, no plans to replace/upgrade
31
DIALLER MODES
The following definitions of automated dialling are provided as guidance, although there is an
acknowledgement that some in the industry may refer to these modes differently.
Preview dialling: once an agent has indicated that they are ready for a call, information about the
call is presented to them. The number is then automatically dialled after a predefined period; the
agent is given time to preview the customer details before the call is launched. Calls where agents
need to be familiar with the context of the call or the customer history, or where the call may be of a
more complex and personalised nature are particularly suitable for the preview dialling mode.
Progressive dialling: once an agent has indicated that they are ready for a call, information about
the call is presented to them and the number is then dialled immediately. Call progress is monitored
by the dialler technology. Calls that do not result in 'ringing' are automatically and immediately
disconnected, whilst 'no answers' are disconnected after a predefined number of seconds. This
dialling mode enables contact centres to avoid abandoned calls, and may be used in conjunction
with predictive dialling in order to keep abandoned call rates within acceptable boundaries.
Predictive dialling: a predictive dialler launches calls at a rate such that the system connects to live
callers as soon as an agent completes the previous transaction, meaning that agents do not need to
listen to engaged tones or answer machines. The dialling rate for each campaign is controlled by a
pacing algorithm, which automatically monitors activity, and calculates when the next contact
should be dialled. The dialling rate is automatically adjusted to maintain a contact rate that is
theoretically synchronised with operator availability. The accuracy of predictive diallers improves as
more agents are using the system, as it means there are more data from which to estimate average
call length. As such, the best results are often seen in larger contact centre operations where
efficiency and cost per live call are of key importance to profitability.
32
The chart below shows that in surveys carried out with UK-based outbound contact centre
operations1, it was found that larger operations (200+ seats) are more likely to be using full
predictive mode (73% of their outbound activity), as they will tend to have the agent and data pools
deep enough to make better use of this, although the gap between small and large operations’ use
of predictive is not as wide as might be thought, and is perhaps more dependent on the type of
calling being made. (It should be noted that these figures apply to the outbound sector of the
industry, not the entire contact centre industry).
Figure 19: Use of dialling modes, by contact centre size (outbound operations only)
1 ContactBabel research on behalf of Ofcom, 2015
42% 43%
73%
55%
22%28%
16%
21%7%
15%
6%
8%16%
10%
4%
10%13%
5% 1%6%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Small Medium Large Average
Use of dialling modes, by contact centre size
Manual
Other
Preview
Progressive
Predictive
33
When considering dialling mode by activity type, warm sales and debt collection are somewhat more
likely to have been calls made predictively, with customer surveys and service being less so.
Figure 20: Use of dialling modes, by outbound activity (outbound operations only)
In interviews with outbound contact centres, respondents were asked why they used particular
modes of outbound dialling.
Preview dialling was stated to be chosen in cases where:
there was a danger of breaching Ofcom regulations through imperfect use of predictive
dialling
campaigns were small and/or where data was of high worth
activity involved handling callback requests, where the customer and their requirements
could be checked before the call was placed.
Progressive dialling was more often used where:
there is a blended environment which may need rapid changes between inbound and
outbound work, and which thus risk high spikes of abandonment as agents move between
inbound and outbound more quickly than a predictive dialler can handle
campaigns are smaller than those which typically use predictive dialling, but management
still wish to gain efficiencies.
14%4% 8% 4%
50%
11%13%
6%
8%
21%17% 35%
28%
64% 62%55%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Service Sales Debt collection Customer surveys
Use of dialling modes, by outbound activity
Predictive
Progressive
Preview
Manual
34
Predictive dialling is said to be chosen where:
there are very high volumes of data in the campaign
the campaign or data is in a late stage and is the most efficient way to contact hard-to-reach
customers
where data quality is low, so it screens out unusable numbers
where there is a pressing need to speak with customers quickly (for example, to stop them
falling down the debt chain), where non-predictive calling would not support this.
Other automated dialling:
‘ratio’ dialling2 is used by a small number of respondents, when calls are very short and
predictive dialling will slow things up especially at the start of the day when contact rates are
the best. This method is said to allow a quicker response to increases in call drop rate.
While the productivity benefits possible through predictive dialling are considerable, businesses
need to be aware that unless managed correctly, predictive diallers can quickly lead to high levels of
call abandonment and silent calls as there are not enough agents available to handle the live
contacts. Ofcom regulations currently state that no more than 3% of callers can be abandoned in this
way, and there is currently at the time of writing (March 2016) an extensive ongoing review into how
silent and abandoned calls can be minimised further.
Additionally, overuse of aggressive predictive dialling algorithms can put extreme pressure on
agents, leading to burnout and high levels of agent attrition and consequent recruitment costs, as
well as generally lower levels of service quality.
2 Ratio dialling: a ratio of lines-to-agent is set, i.e. for each agent, the dialler will have been told to place a number of calls, so a 3:2 ratio will have 3 calls placed for 2 agents. This is a simple method of automated dialling as it does not need a pacing algorithm, but needs close monitoring by the dialler manager as the ratio will not alter unless it is manually changed and risks making abandoned calls if it is not regularly updated.
35
OUTBOUND OUTCOMES
A series of questions was asked to respondents from outbound contact centres about the typical
results that they had from their ongoing outbound activity and/or campaign. Slightly over half of
calls made were answered - either by an answer phone or by a person - with those carrying out debt
collection activities having the lowest answer rate and those carrying out customer service activities
the highest.
Figure 21: Proportion of calls answered, by outbound activity type
Outbound activity type
% of calls answered
Customer service 68%
Customer surveys 65%
Warm sales 60%
Cold sales 59%
Debt collection 44%
Other 62%
Average
53%
Large operations were less likely to have their calls answered – they are more likely to call for longer
hours rather than just the peak hours.
Figure 22: Proportion of calls answered, by size
Contact centre size
% of calls answered
Small 60%
Medium 51%
Large 47%
Undisclosed 51%
Average
53%
36
Of calls answered, responses showed that 49% were answered by a person and 51% by an answer
machine / voicemail.
Figure 23: Of the calls that were answered, how many were by a consumer? (by outbound activity type)
Outbound activity type
% of calls answered, that were answered by a consumer
Customer surveys 85%
Customer service 64%
Cold sales 57%
Debt collection 41%
Warm sales 38%
Other 65%
Average
49%
Figure 24: Of the calls that were answered, how many were by a consumer? (by contact centre size)
Contact centre size
% of calls answered, that were answered by a consumer
Small 51%
Medium 53%
Large 45%
Average
49%
37
1 in 8 of the calls that were answered by a consumer were then hung up by that consumer, rejecting
the call. This is far higher in those carrying out debt collection and cold sales activities.
Figure 25: Of the calls that were answered by a consumer, how many were hung up by the consumer when an agent was available? (by
outbound activity type)
Outbound activity type
% of calls answered by a consumer, that were hung-up by a consumer when
an agent was available
Debt collection 22%
Cold sales 10%
Warm sales 2%
Customer surveys 1%
Customer service 0%
Other 3%
Average
12%
In 1.1% of cases where the call was answered by the consumer, a recorded message was played as
no agent was available, with this figure being the highest in the heavily-predictive worlds of debt
collection and cold sales.
Figure 26: Of the calls that were answered by a consumer, how many had a recorded message is played as no agent was available (by
outbound activity type)
Outbound activity type
% of calls that were answered by a consumer, and an
agent is not available, where recorded message is played
Debt collection 1.7%
Cold sales 0.9%
Warm sales 0.8%
Customer surveys 0.3%
Customer service 0.0%
Other 0.3%
Average
1.1%
38
15% of calls answered by an answer machine ended in a recorded message being left on the answer
machine. Those carrying out debt collection and warm calling activities appear to be the only activity
types doing this of the survey respondents.
Figure 27: Of the calls that were answered by an answerphone, where a recorded message is played, by outbound activity type
Outbound activity type
% of calls that were answered by an answerphone, where
a recorded message is played
Debt collection 25%
Warm sales 17%
Cold sales 0%
Customer service 0%
Customer surveys 0%
Other 0%
Average
15%
It is interesting to note that 85% of respondents answering this question did not play a recorded
message to an answerphone. The majority stated that they did not use AMD and therefore the call
will be passed through to an agent who would either leave a message or hang-up depending on the
business requirements.
39
AGENT UTILISATION RATES
The majority of respondents calculate agent utilisation rate as a proportion of the time spent talking
to end-users, divided by the amount of time available to talk. For example, if an agent spends 30
minutes in the hour talking, and they are logged in to the system for 50 minutes (with a 10 minute
break), then the majority of contact centres calculate this to be a 60% utilisation rate (30/50).
Most respondents do not include spending time before the call looking at contact data, or adding
wrap-up notes after the call to be a part of agent utilisation, stating that utilisation rates are only
calculated based on the % of time spent in a 'talk' state on the dialler as a % of the total time the
agent was working, and does not include any time updating customer records or waiting for calls.
However, most of the higher (70%+) results in the chart below also include wrap-up time (e.g. Talk +
Wrap, as a % of Talk + Wrap + Idle). For most respondents that split out wrap-up time from call time,
wrap is usually between 15-20% of the call time.
Figure 28: Average outbound agent utilisation rates
Agent utilisation rates
1st quartile 70%
Median 56%
3rd quartile 42%
Mean 56%
30% or lower5%
31-40%14%
41-50%25%
51-60%20%
61-70%13%
71-80%16%
Over 80%7%
Average agent utilisation rate
40
AGENT UTILISATION RATES AND DIALLING MODES
Analysis of agent utilisation rates were carried out against the main dialler mode employed by
respondents, in order to investigate whether any dialling mode has a correlation with agent
utilisation. For example, as predictive dialling is a highly automated process, it might be expected
that this mode would clearly yield the highest utilisation rates.
However, there are more variables than just dialler mode that will affect utilisation rates, and which
themselves may encourage an outbound operation to use one mode over another. For example,
respondents state that predictive diallers are more likely to be used than other modes in cases
where data quality is low, and there have already been multiple attempts made to contact a
customer. Different businesses will have their own views on the balance between efficiency and
quality, and the amount of wrap-up time that may be required by various business activities will be
different, and will affect utilisation as well (if included in the utilisation calculation).
The overall agent utilisation rate (i.e. all responses, regardless of dialling mode and method of
calculation) is 57%, or 34.2 minutes per hour.
When considering only those respondents who consider talk-time, but not wrap-up in their
calculations, this figure is 51% (30.6 minutes per hour)
For respondents who include both talk time and wrap-up in their calculations, the utilisation
rate is 68% (40.8 minutes per hour).
In order to see whether and by how much dialling mode can be said to affect agent utilisation, the
utilisation rates were calculated for respondents using particular dialling modes. It was found that
amongst contact centres which calculated utilisation through only talk-time and which used mainly
predictive dialling had an average utilisation rate of 62% (37.2 minutes per hour), compared to the
respondent base as a whole, who had a 51% utilisation rate. This implies a clear increase in agent
utilisation time through predictive dialling.
It is interesting to note that industry statistics on dialling mode efficiency suggest the following to be
reasonable estimates3, ceteris paribus:
manual dialling: 12 minutes of talk time per hour (20%)
preview dialling: 18 minutes of talk time per hour (30%)
progressive dialling: 25 to 35 minutes of talk time per hour (42% - 58%)
predictive dialling: 35 to 45 minutes of talk time per hour (58% - 75%)
It should also be noted that many respondents stated that predictive dialling is very likely to be used
for more difficult ‘end of project’ calls where customers have been called many times already, with
the initial attempts made using IVM, and high quality data being originally called in preview or
progressive modes. As the quality and age of the data is different depending on the mode used, the
fact that there are numerous variables changing should strike an even greater note of caution.
3 ContactBabel estimates, based on a mix of secondary research with dialler manufacturers and primary ContactBabel research
41
DIALLING ACTIVITY: TIME OF DAY
Contact centre respondents provided information about the hours in which they carry out outbound
calling, and indicated the peak times where they were making most calls.
In the working week, calling tends to start at 0800 or 0900 and carry on for 12 hours until 2000 or
2100. Calling before 0930 was almost always carried out for debt collection.
By 10am, 85% of respondents are making outbound calls, with 72% finishing by 8pm. By 9pm, all
respondents have finished calling.
Figure 29: Cumulative opening and closing hours, Mon-Fri
29%31%
71%74%
85%88%
91% 92%
99% 99% 100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 1% 1% 1% 1% 1%4%
7%10%
12%
25%
29%
72%
76%
100%100%100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0800 0830 0900 0930 1000 1030 1100 1130 1200 1230 1300 1330 1400 1430 1500 1530 1600 1630 1700 1730 1800 1830 1900 1930 2000 2030 2100 2130 2200
Cumulative opening and closing hours Mon-Fri
M-F start M-F end
42
Figure 30: Cumulative opening and closing hours, Sat
Only around 30-40% of outbound respondents stated that they called on Saturdays.
Saturday calling was usually carried out in the morning, tailing off after 1pm. By 9am, most
respondents were calling, with around one-third stopping by 1pm, with a gradual tailing-off of
activity throughout the afternoon, with debt collectors the most active. By 6pm, almost all outbound
calling had stopped.
20%
27%
80% 80%
100%100%100%100%100%100% 100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%
0% 0% 0% 0% 0% 0% 0% 0%
7%10%
37% 37%
47% 47% 47% 47%
60% 60%
67%
73%
93% 93%97% 97% 97% 97%
100%100%100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0800 0830 0900 0930 1000 1030 1100 1130 1200 1230 1300 1330 1400 1430 1500 1530 1600 1630 1700 1730 1800 1830 1900 1930 2000 2030 2100 2130 2200
Cumulative opening and closing hours, Sat
Sat - start Sat - end
43
The great majority of respondents do not call on Sundays. Of those that do, calling begins at 10am
and tends to finish between 2pm and 6pm.
Figure 31: Cumulative opening and closing hours, Sun
When looking at peak calling hours, this differs somewhat depending on the type of calling being
carried out:
Debt collection respondents state that they have an outbound calling peak in the early evening,
between 5.30pm and 8pm, with many operations also stating that there are peaks in the morning as
well, particularly between 10am and 12 noon.
Amongst respondents that carry out sales to existing customers (for example contract renewals,
cross selling and upselling), an evening peak is even more noticeable. Many respondents report the
period between 4.30pm and 7.30pm has more volume, with very few reporting significant sales call
volumes before 12 noon.
For respondents carrying out cold calls, there were less obvious peaks, with calls spread out across
the day, perhaps as for these types of sales campaign / outbound activity, there are likely to be more
targets to aim for.
0% 0% 0% 0%
100%100%100%100%100%100% 100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%100%
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
33% 33%
50% 50%
67% 67% 67% 67%
83% 83% 83% 83% 83% 83%
100%100%100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0800 0830 0900 0930 1000 1030 1100 1130 1200 1230 1300 1330 1400 1430 1500 1530 1600 1630 1700 1730 1800 1830 1900 1930 2000 2030 2100 2130 2200
Cumulative opening and closing hours, Sun
Sun - start Sun - end
44
Different segments of the contact list are likely to be available at different times: at the simplest
level, retired people are likely to available all day, and people of working age during lunchtime and
the evening. Of course, for anything remotely associated with B2B, evenings and weekends should
not be considered. Published research4 suggests that the best time for B2B cold calls is the early
morning (before 9 am) and towards the end of the working day (4-5pm), outside of core meeting
hours.
It is important to take into account external events, such as sporting events within a city (for
example, calling specific demographics during an important football match is unlikely to yield
anything positive), and if at all possible, getting the customer to state their preference early in the
business relationship in terms of availability and preferred channel will be positive for all concerned.
‘Warming them up’ through an earlier SMS message is also a tactic to be considered.
To make the customer segmentation strategy really work requires considerable amounts of data on
customer demographics, purchasing history and preferred channels and contact times, and
businesses should look to extract such information from their sales intelligence application and
match it with outbound calling outcomes and if possible, the elements of dynamic scripting that tend
to work best with such customers.
4 Sung Hyun Kwan Business School in association with Kellogg School of Management
45
DIALLING ACTIVITY: MAXIMUM NUMBER OF ATTEMPTS
As might be expected, the maximum number of attempts made to contact a customer or prospect
differs significantly depending upon the activity being carried out. Sufficient data on three outbound
activities - debt collection, warm calling (sales to existing customers) and cold calling (sales to new
prospects) were available to allow detailed analysis of call attempts and the time left between calls.
Those involved in debt collection, as we might expect, are determined to speak to specific
individuals, and almost half of them make multiple calls on the same day (with a few making as
many as six calls per day to the same individual). In the main, they will wait a minimum of four hours
between calls, although several mention that if an engaged signal is received, they will ring back
within 15-30 minutes as the chances of successful connection is higher. There are instances of
respondents making up to 100 calls per month to a specific number, although on average, making
two or three calls per week is much more common. Several respondents pointed out that if a
message is left or a call dropped, they would not call back within 72 hours.
Figure 32: Call attempts by activity type (debt collection)
Maximum calls per
consumer per day
Length of
campaign / live data
(days)
Maximum calls
in total campaign time
Time left
between calls (if engaged tone) -
hours
Time left
between calls (if no engaged tone)
- hours
1st quartile 3 28 20 2.25 72
Median 2 21 10 0.5 4
3rd quartile 1 7 6 0.38 4
Mean 2.2 21 18 1.3 24
High 6 90 90 4 72
Low 1 2 3 0.25 2
46
Those respondents involved in calling existing customers, for example in order to cross-sell, upsell or
renew contracts, are less likely than debt collectors to call as many times, although they have a
higher average attempt per customer per day figure (although the research base for this was only 6
respondents, so should be treated with caution).
Typically, calls will be placed every two or three days for a three- or four-week period, with around 8
calls over this period being average. Having said that, some respondents do indicate that they call
more frequently, although none make more than 12 calls in total.
Figure 33: Call attempts by activity type (warm calling – sales calls to existing customers)
Maximum calls per
consumer per day
Length of
campaign / live data
(days)
Maximum calls
in total campaign time
Time left
between calls (if engaged tone) -
hours
Time left
between calls (if no engaged tone)
- hours
1st quartile 3 28 10 n/a 14
Median 3 28 8 0.5 4
3rd quartile 3 18 8 n/a 4
Mean 3.5 22 8 0.5 9
High 8 35 12 0.5 24
Low 1 3 3 0.5 3
Respondents carrying out sales calls to new prospects will call multiple times in a week - around
three or four being average - with a couple of respondents carrying out very significant focused
calling, with around 10-15 calls being placed to the same potential customer.
Figure 34: Call attempts by activity type (cold calling – sales calls to new prospects)
Maximum calls per
consumer per day
Length of
campaign / live data
(days)
Maximum calls
in total campaign time
Time left
between calls (if engaged tone-
hours)
Time left
between calls (if no engaged tone
- hours)
1st quartile 3 32 18 n/a n/a
Median 3 28 10 0.38 4
3rd quartile 2 11 3 n/a n/a
Mean 3 25 18 0.38 3.4
High 3 56 100 0.5 4
Low 2 5 3 0.25 2
47
RING TIME AND CONNECTION TIME
MINIMUM RING TIME
The survey showed that the mean average ring time industry-wide is 22 seconds, with cold selling
tending to ring the longest and ‘other’ the shortest. It can be seen that the small sample size of some
outbound activity types of skewing the mean results somewhat (particularly in the case of customer
surveys), so median averages have been included as well for greater clarity. It seems fair to note that
those involved in sales to new customers (i.e. cold calling) are more likely than other sectors to let
calls ring longer.
Figure 35: Minimum mean and median ring time before call termination, by outbound activity type
Outbound activity type
Mean minimum ring time before
call termination (seconds)
Median minimum ring time
before call termination (seconds)
Cold sales 26 25
Customer surveys 23 15
Debt collection 22 24
Customer service 21 20
Warm sales 20 21
Other 17 18
Average
21.9 20.0
The majority of respondents who stated why they set a particular minimum amount of time before
call termination explained that there was a sweet spot around 18 - 20 seconds: they referred to
Ofcom’s policy on calls ringing for a minimum of 15 seconds, and many answer machines kick in
after 21 seconds. It was also noted that studying historical maximum wait times indicated that
ringing for longer than 20 seconds rarely results in much improvement, and increasing answer
machine rates would also ‘lock out’ the record with regard to redial rules.
48
DIALLING ACTIVITY: THE ROLE OF MOBILE
The majority of mobile users have access to a smartphone, meaning a large proportion of customers
will want to contact businesses through these devices, whether via the telephony element of the
device, or via the company’s website or mobile app. Taking into account the use of tablet computers
to access the Internet, the ‘mobile channel’ may actually be the first port-of-call for many customers,
especially those in the younger demographics.
The dual, mutually-supporting drivers of high-speed mobile networks and the proliferation of
smartphones means that provision of services via a mobile channel offers businesses and consumers
the opportunity to make a step-change in the way that they communicate with each other. This new
world of communication allows businesses to consider whether functionality such as multimedia
streaming and videoconferencing could give them a competitive advantage in the customer service
world.
Gathering, understanding and using the contextual data that can surround the mobile consumer will
be key to pushing the uptake and functionality of this channel forward. The plethora of channels
immediately available to the mobile consumer - including voice, web browsing, SMS, social media,
and web chat - encourages the customer to act immediately for all their service or information
requirements, rather than waiting until they are in front of a desktop computer.
This year’s UK survey respondents report that it costs almost twice as much to call a mobile phone
than a landline, a gap that is shrinking each year:
Mean cost of outbound call to landline (ppm): 2.2p
Median cost of outbound call to landline (ppm): 1.0p (1st quartile 0.2p, 3rd quartile 2.7p)
Mean cost of outbound call to mobile (ppm): 4.1p
Median cost of outbound call to mobile (ppm): 3.0p (1st quartile 1.0p, 3rd quartile 6.5p)
Although it is more expensive to call a mobile than a landline, the former’s ubiquity and
responsiveness has a great advantage over landline phones and even emails, as the business is far
more likely to be able to reach the customer by calling their mobile number, although there is no
guarantee of an answer, which is why SMS is often used as part of the outbound mix.
The ubiquity and massive increase in the use of smartphones also means that the mobile channel
has become a crucial part of the customer contact mix, and businesses should look to engage with
mobile customers, not worrying too much about cost differentials. Besides, increasing the use of
SMS as a channel to the customer would reduce many of the costs associated with mobile
communication (including agent time, as well as transmission costs), while keeping the advantages
of contacting a customer’s mobile number.
49
Just over half of calls made by survey respondents are placed to mobile phones, which while they
are more expensive, are more likely to be answered by the correct person regardless of time of day.
Figure 36: % of outbound dialling carried out to mobile phones
% of calls made to mobile phones
1st quartile 73%
Median 60%
3rd quartile 42%
Mean 58%
There is little difference between activity types, with 55% of debt collection calls being made to
mobiles, and 61% of sales calls.
Further information about servicing the Mobile Customer can be found in “The Inner Circle Guide to
Multichannel Customer Contact”, and “The Inner Circle Guide to Self-Service”.
Both reports are available free of charge from www.contactbabel.com.
END-USER QUESTION #2:
WE’RE INTERESTED IN INCREASING PRODUCTIVITY, BUT CONCERNED ABOUT AGENT BURNOUT AND INCREASED ATTRITION. DO YOU HAVE ANY POINTERS ON BEST PRACTICE?
If you just turn up the productivity dial agent burnout can be a problem. To
avoid it, make sure that you keep a close eye on effectiveness as well as
productivity – for example, if you are making lots of calls but not reaching your
target audience then your agents will be unhappy.
Use the data management tools incorporated in modern outbound and blending systems: these help
you address the right audience at the right time and be able to mix your communication channels to
give you agent productivity, the right business outcomes and a work blend that avoids agent
burnout.
50
ANSWER MACHINE DETECTION
Answer Machine Detection (AMD) is a type of technology that allows diallers in predictive mode to
recognise calls that are picked up by an answer machine, and stop the outbound agent taking that
call, keeping them free to speak with actual customers. The purpose of this was simply to present
agents with the greatest number of live contacts. The negative experience of the customers
answering the phone that were incorrectly identified as being answer machines and then hung up on
(‘false positives’) was rarely considered.
Most AMD solutions usually work by:
Detecting a long string of words in one burst such as “Hello, we’re not available, please leave
a message after the tone …”, deducing from the patterns of noise and silence on the line
whether an automated message is being played when a call is connected. This is known as
the ‘cadence’ method, but there are other ways in which AMD can work
Identifying a live caller who may answer by saying “Hello?” and then waiting for a reply.
However, AMD may cause problems and is not 100% accurate:
There will be a pause or delay while the system tries to connect an agent, which may lead to
the called party hanging up
the AMD solution may incorrectly flag a live person as being an answering machine, hanging
up the call. The effect of this ‘false positive’ is that the customer will effectively get a silent
call which can be frustrating and worrying for the majority of the public who are unaware of
how outbound technology works.
In recent years, Ofcom has had to respond to increased numbers of complaints about silent and
abandoned nuisance calls made as a result of AMD: as a result, a set of rules was designed in order
to reduce the number of abandoned or silent calls.5 However, at the time of writing (March 2016),
Ofcom is coming towards the end of a new regulatory review which is expected to be considerably
tighter and potentially more punitive than the existing regulations, and which may also insist that all
outbound calls include a CLI to identify the calling organisation.
Cadence AMD – analysing from the patterns of noise and silence on the call whether an answer
phone message is being played – relies upon an accurate assessment of whether a call has been
answered by a human or answer machine within the two-second window currently allowed by
Ofcom, which does not give enough time to both detect 100% of answer machines and exclude all
live contacts, leading to silent and abandoned calls. In order to remain compliant with Ofcom (and
prove it), there is also a great deal of management and reporting to be carried out, reducing the
productivity benefits.
5 http://stakeholders.ofcom.org.uk/consultations/silent-calls/statement/
51
In terms of estimating the accuracy of the AMD, the majority of UK outbound survey respondents
tested the accuracy by comparing agent or supervisor experience with what the dialler is telling
them. For example, a dialler manager may listen to a substantial number of calls each month or
quarter, and analyse how many are AMD false positives compared to what the system is stating. In
this way, a more accurate view of the accuracy answer machine detection is given. Other
respondents switch AMD on, but still pass all calls through to agents for a specific period. The agent
marks each call as answer machine or not, and this is then compared to the AMD statistics to see if
they match up.
When respondents were asked to estimate AMD accuracy, results given were:
Figure 37: Estimation of AMD accuracy
The majority of respondents include AMD false positives in their overall abandoned call rate, with
these estimated AMD false positives being factored into the dialler settings to reduce the maximum
percentage of abandoned calls before they occur. The resulting number of abandoned calls can be
added to the result set to achieve a true abandonment figure.
Some contact centres do not include false positives in their abandoned call rate, as playing an IVM
message triggered by AMD is the same as choosing to play an IVM to the customer (e.g. through a
broadcast message), which a debt collection agency considers they are allowed to do. If the
customer chooses not to press 1, then that is their decision.
The rate of silent calls - where AMD falsely detects that an answer machine is present - can by their
nature only be estimated, otherwise they would not have been abandoned. This requires companies
to test their systems and analyse recordings to estimate the rate of silent calls, which gets taken
<50%4%
50-75%4%
76-90%12%
91-95%8%
Over 95%8%
Did not answer64%
AMD estimated accuracy
52
from their overall abandonment rate, which is a maximum of 3% of calls, meaning the dialler can
only be run at a maximum rate of 3% minus the estimated false positive rate, further reducing
productivity gains. An alternative solution to trying to detect every answering machine is only to
remove calls if the systems is 100% certain of that an answer machine is present.
Respondents to a ContactBabel survey of UK outbound contact centre operations estimated that
0.4% of calls answered by consumers were believed by outbound operations to be AMD false
positives (where the dialler incorrectly guesses a real person is an answer machine and hangs up the
call, producing a ‘silent call’), although this was from a relatively small sample size and should be
treated with caution.
Of those outbound respondents that use AMD, 44% employ it all of the time, particular the case in
medium and large operations. A minority pick and choose the times they use it (some contact
centres only employ it in peak calling times, where customers or prospects are most likely to be
available, so as to maximise live contacts). Only 15% of outbound users state that they do not have
AMD functionality in their solution, with a higher figure from those in small operations.
Figure 38: Use of answer machine detection (AMD), by contact centre size (UK) – only those using automated outbound solutions
57%
47%
44%
33%
7%
11%
13%
14%
5%
8%
33%
21%
15%
33%
14%
11%
15%
7%
5%
5%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Small
Medium
Large
Average
Use of answer machine detection (AMD), by contact centre size
Yes, and we always use it Yes, and we sometimes use it Yes, but we rarely use it
Yes, but we never use it No AMD functionality Don't know
53
The use of AMD may be written into the client’s request in the case of an outsourcer, or it may be an
internal decision driven by dialler managers and compliance directors. AMD and IVM (interactive
voice messaging) can be used in tandem to minimise false AMD abandonment. A respondent to
ContactBabel’s survey of outbound contact centre explains:
“If we place an outbound call and within two seconds of line activation we believe it to be a
machine we will play an IVM message to ask the recipient to confirm they are the person we
are required to speak to, before giving the option to press one and enter the call centre. If
the recipient does not press 1, then they are played the same information as an abandoned
call so that the company name and identity has been provided.”
In this way, there is a two-step process to determining whether there is a consumer or answer
machine at the end of the line, and by giving the consumer the option to speak to an agent, AMD
false positives should be reduced.
As part of this survey, respondents were asked what their experience had been when they had
restricted or switched AMD off and on to see what the impact upon productivity and compliance had
been. It was generally felt that there was a significant improvement in live contact rates when AMD
was in place, and that agent morale and performance dropped considerably when AMD had been
turned off, due to a large increase in the instance of answer phones presented to agents. Some
respondents noted that they used AMD only at specific times during the day, and when the data that
they were calling on was of questionable quality. However, there was a general acknowledgement
that the nature of most AMD solutions being used today will produce some false positives, and that
legislation requiring recorded messages and/or an IVR option to speak to an agent would assist the
customer experience without damaging productivity.
USE OF CLI / CLID
The presentation of CLI / CLID (calling line identification) is required for US telemarketers, and it is
illegal to ‘spoof’ CLID (make it appear that the call is originating from a different number). In the UK,
DCMS (the Department of Culture, Media and Sport) want to make outbound callers provide an
accurate CLI and the ICO (information Commissioner’s Office) supports this, so it is likely to become
law at some point in 2016.
CLI is in any case presented by 93% of UK respondents to ContactBabel’s outbound survey, so any
change to regulations is likely to affect rogue calling organisations. Some respondents, often from
the debt collection sector, mentioned that localised CLI (i.e. those that appear to have originated
from the same town or city as the called party) tend to have improved answer rates, and that calls
clearly originating from UK operations are also likely to have improved results.
Around a third of UK outbound contact centres use multiple CLIs, which allow them to route a
customer calling back to a specific department, client, product, brand or campaign, and also to
provide local support. Some of the debt collecting respondents state that debtors learn to recognise
specific CLIs, and that alternatives are used to improve answer rates.
54
THE USE OF IVM (INTERACTIVE VOICE MESSAGING)
Interactive Voice Messaging (IVM) is often used alongside live outbound calling, with an automated
message inviting the recipient to enter into conversation, either now or at a later time.
Around a third of outbound contact centres in the UK use IVM, with those in the debt collection
industry most likely to do so. Respondents within the industry state that they use this in order to
give as many people as possible the chance to be connected to an agent before the downstream
collections process takes over: unlike calls to new customers (cold sales) for example, there is a
limited time period for a debt to be settled at a specific stage before it is escalated. It is also seen as
a way to clean lists and collect the easiest debts before the respondent switches dialling mode to
predictive dialling.
By far the most prevalent use of IVM is where a recorded message is played that asks specifically by
name for a particular customer, asking them to press 1 if it is them in order to be transferred to an
agent. If no key presses are detected, a message will be left with the caller’s name and telephone
number, asking them to call back. This may be looped to avoid partial recordings being made.
Businesses will ask the customer to identify themselves through their date of birth or account
number before any information is shared with them.
Some contact centres state that they will make sure an agent is available to take the call if 1 is
pressed, others that any IVM returns that are abandoned will not be called for another 72 hours at
which point a live agent will be made available.
IVM is seen as less intrusive than live outbound dialling. It can be personalised, and the customer
can choose to interact with the company at that time, advise a better time to contact or simply hang
up. For the debt collection industry, a significant proportion of customers that enter collections are
able to make a payment to bring their account up to date, exiting the collections process within the
same month. IVM enables companies to automate outbound interactions with this segment of
customers and thus enable live call centre resources to focus on speaking to customers with more
serious financial difficulties or more complex arrears queries, which is beneficial to both customer
and company.
IVM is less expensive than employing agents and can be particularly useful in small teams where
predictive dialling is not possible due to the risk of high call abandonment rates. IVM is also effective
at cleaning data, for example, when attempting a phone number for the first time, IVM helps cleanse
the database of bad or dead numbers and provides an instant ‘opt-out’ process for removal of Do
Not Contact numbers.
It also allows customers to know who the organisation is that are trying to contact them, and to
press 1 to speak to them which is especially useful in the case of false AMD detection. IVM can also
be used to assist with fraud prevention and credit card checking, as it is a timely and rapid method of
informing the customer that there is a potential problem.
55
If a customer receiving an IVM call indicates they wish to speak to an agent, survey responses show
that target connection times are substantially shorter than the usual 20-30 second targets employed
in most ‘standard’ inbound environments, with a mean of 15 seconds and a median of only 8
seconds. The actual average time per call that inbound IVM-prompted customers are on hold is a
mean of 9 seconds, and a median of 8 seconds, so companies tend to exceed their target.
It is reported that 27% of IVM calls were answered by a customer, and 73% by an answerphone.
Conjecturally, this may be because IVM is more likely to be used in situations where an answering
machine is more likely to pick up, or the consumer is less likely to answer the call (e.g. debt
collection).
USE OF BROADCAST MESSAGING
Automated messaging or ‘broadcast’ messages are calls made with the sole purpose of making a
recorded announcement, rather than to connect a consumer to a live agent immediately. Examples
include a recorded sales or marketing message, or a recorded information message which is not an
‘Agent Unavailable’ message.”
Broadcast messaging is carried out by far more of the large operations than small, with one third of
large UK outbound operations using broadcast, compared to only 3% of the sub-50 seat category.
Debt collection operations respondents in particular stated that they use broadcast messaging to
drive inbound call volumes, as it is seen as another approach to take with customers who are
difficult to contact directly.
56
COMPLIANCE WITH REGULATIONS
UK regulations
The Telephone Preference Service (and the EC Regulations on Electronic Communication which deals
with email and SMS) are part of the general social and political drift towards allowing consumers and
businesses the right not to be contacted by companies.
In the UK, Ofcom is getting progressively stricter in its outbound regulations, a summary of which
appears here. Please note that regulations are subject to change and readers should refer to the
Ofcom website for the most up-to-date information.
Summary of Ofcom's Dialling Regulations
Call abandonment rate shall be no more than 3% of ‘live calls’ on each individual campaign over
any 24 hour period;
In the event of an ‘abandoned call’, a very brief recorded information message is played within
one second of the call being answered, which;
1. identifies the company on whose behalf the call was made;
2. identifies the intended purpose of the call (i.e. “unsolicited sales call”, “call as part of debt
recovery”, etc.)
3. offers the called person the possibility of declining to receive further calls from that company by
contacting a no charge or basic rate number;
4. includes no marketing content and is not used as an opportunity to market to the called person;
calls which are not answered should ring for a minimum of 15 seconds before being terminated;
when an ‘abandoned call’ is made to a particular number, that number is not called again in the
following 72 hours, unless a dedicated operator is available;
for each outbound call a CLI number is presented, to which a return call may be made which is
not charged at a higher rate than the national call rate;
either a recorded message or a live operator is available at the CLI number presented to inform
called persons of the identity of the organisation that called them, the intended purpose of the
‘abandoned call’ and that the called person’s number will be deleted from the organisation’s
database and added to its in-house suppression list at the called person’s request if they leave
their name and telephone number;
any call made by the called person to the contact number provided shall not be used as an
opportunity to market to that person;
records are kept that demonstrate compliance with the above procedures.
57
Ofcom has prioritised the reduction or elimination of silent and abandoned calls, and the most
recent (December 2015) document is downloadable from:
http://stakeholders.ofcom.org.uk/consultations/review-of-how-we-use-persistent-misuse-powers/
Proposed regulations are available here:
http://stakeholders.ofcom.org.uk/binaries/consultations/review-of-how-we-use-persistent-misuse-
powers/annexes/Annex_5.pdf
As part of the overall review into outbound calling, the Department of Media, Culture and Sport
(DCMS) wants to make outbound callers provide an accurate and returnable CLI, and the ICO and
Ofcom support this.
58
DIALLING ACTIVITY: TPS
The Telephone Preference Service (TPS) is a free service and is the UK opt-out register on individuals
and those working in organisations (under the CTPS) can record their preference not to receive
unsolicited sales or marketing calls. It is a legal requirement that all organisations (including
charities, voluntary organisations and political parties) do not make such calls to numbers registered
on the TPS unless they have consent to do so.
In a survey of outbound UK contact centres, around half of calls were stated not to require TPS
checking (as they were to existing customers or others who had given permission, or who were
allowed to be called through other legislation).
For those who were conducting TPS tests, checking numbers before they were loaded into the
dialler was the most commonly used method by far. (NB – ‘other’ outbound activity, not shown
specifically, was included in the ‘average’ column).
Figure 39: Methods of checking TPS (n=91)
6%11% 11% 8% 10%
12%
44% 47%
78%
29%
32%
11%
3%
3%
4%
79%
44%
32%
14%
71%
51%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Debt collection Customerservice
Warm sales Cold sales Surveys Average
Methods of checking TPS
TPS check notrequired
TPS check not carriedout
Manually checkedbefore dialling
Checked beforenumbers were loadedinto the dialler
Automaticallychecked as dialled
59
END-USER QUESTION #3:
WHAT IMPACT DOES CURRENT AND FUTURE LEGISLATION HAVE ON OUTBOUND, AND HOW DO YOUR SOLUTIONS ADDRESS THIS?
At the time of writing the consultation phase of Ofcom’s “Review of how we use
our persistent misuse powers: Focus on silent and abandoned calls” has just
ended. Assuming that the final guidance will reflect information contained in the
consultation, Rostrvm Solutions supports the review and the clarification that it will provide.
We particularly welcome the added guidance on Interactive Voice Messaging (IVM) which, if used
sensibly, extends managed-contact opportunities with customers.
The detail in the consultation document suggests several changes to be taken into account. In
reality, any contact centre that is following the current (2010) guidelines – and the spirit of them too
- shouldn’t be overly concerned. That said, contact centres operating outside, or even on the outer
limits of, the current guidelines need to take a long, hard look at their operational practices. The
rostrvm dialler has all the controls you need to achieve Ofcom-compliant dialling.
There are other regulators to consider such as the Information Commissioner’s Office (ICO). Any
outbound system needs to manage data-related issues. Again rostrvm includes the facilities you
need to comply – checking your lists against Telephone Preference Service, de-duplicating data and
managing Do Not Call lists.
Also, dialler managers need to consider carefully their campaign strategies and, if they haven’t done
it already, move from the old 1990s-style ‘load-up-the-dialler and step-on-the-accelerator’
techniques and adopt a more precise and targeted approach – we call it Precision Contact.
Precision Contact uses the available tools to target your audience
at the right time – the best time to call;
with the right message and channel – modern campaigns mix telephone, text messages and
email;
and the right people – making sure your calls are handled by appropriately-skilled agents.
60
COMPLYING WITH REGULATIONS
Outbound contact centres were asked to give their opinions of the reasons for non-compliance with
regulations, and what they considered to be the main drivers of silent and abandoned calls.
A top-level categorisation of drivers is shown below (placed in order of frequency):
poor dialler management
rogue operators
agent behaviour
targets and KPIs
poor data.
Poor dialler management
Outbound contact centres note that abandoning calls can happen as a result of dialler pacing being
too aggressive, or agents spending too long in unavailable time, in conjunction with too little
monitoring and slow dialler readjustment. It is recommended that the dialler is warmed up in
preview mode before moving to full predictive mode, to avoid causing dropped calls at the start of
the shift due to the mass logging-in of the resource pool. This should ensure that dropped calls are
low from the outset, and abandonment rate will be under the threshold throughout the shift rather
than having to try to recover this in a suboptimal way towards the end of a shift.
Overdial settings should only be used where dialler managers are confident that the creation of high
connect rates can be sustained by agents. Where an outbound-only campaign is taking place, agent
levels can be fairly well predicted, but in a blended inbound/outbound environment, the necessity to
balance the outbound abandonment rate while reacting to inbound call spikes requires very careful
management, and the introduction of minimum agent volumes for the outbound campaigns can
assist with this process.
It is also important to have a sound understanding of the data that is being called, and wherever
possible analyse the correct calling periods for this. Understanding the customer prospect list to
optimise the contact preferences and availability times if possible can greatly assist with contact
rates. Moving between pools of variable data quality can very quickly increase call abandonment
rates, so extra management time at these threshold points is crucial.
Rogue operators
It is an unfortunate fact that some companies blatantly ignore regulations, withholding CLI in order
to obviate complaints from consumers. Proposed regulations around the accuracy and ubiquity of
CLI aim to reduce these issues, although rogue operators calling from offshore are unlikely to be
deterred by this. The planting of monitored /seeded telephone numbers and the aggressive tracing
and investigation of any contact centre calls that have failed to present a valid CLI will assist in
identifying the rogue operators who are responsible for considerable proportion of nuisance calls.
61
Agent behaviour
An increase in abandoned calls can be in part attributable to agent activity, for example not adhering
to correct procedures when logging out for breaks (closing the application rather than requesting
the dialler to log the agent off means that the dialler algorithm is working off incorrect assumptions).
This can be exacerbated by the lack of constant monitoring, and unfocused dialler and line
management.
Targets and KPIs
Judging the success or otherwise of cold calling or marketing campaigns through assessing volumes
or call attempts per day can lead to overly-aggressive dialler management, which is likely to push up
the abandonment rate considerably. There can be pressure put on operational and dialler
management if an outsourcer is rewarded solely through results rather than agent hours billed, for
example: the risk is that the focus changes to call throughput rather than call outcome. This is a
particular risk in organisations which are calling large volumes of untargeted data, for example
generalised PPI or personal injury campaigns. Low contact rates can also occur if there has been a
long period of time since customer consent was given, or if consent was assumed rather than stated
explicitly.
Many outbound solution providers offer their customers a set of tools to offset risk and ensure
compliancy. Detection tools can apply different outbound strategies automatically depending on
whether a landline or mobile number is being called, the time of day, day of week and any public
holidays.
Some countries specify the maximum number of attempts that can be made to reach a customer in
a specific timescale, and any dialler solution should have this functionality. A multichannel solution
can identify and deliver campaigns based on the customer’s original opt-in information and
observing behaviours such as a consumer replying to SMS messages but not to mobile calls.
It is also necessary to be able to screen against national Do Not Call databases, data expiration dates
and to comply with state and regional dialling rules. The increased use of mobile and smart phones
means that it is possible to reduce or eliminate the number of calls made to a device that is any
different time zone, for example when the customer is on holiday.
62
CALL BLENDING
Call blending is an element of outbound calling which has had to fight against the conventional
wisdom of the traditional contact centre industry, which implies that the more one can segregate
the contact centre into a series of production lines, the better-run the operation will be.
A contact centre handling different processes involving customer service, sales orders, and
outbound telemarketing will have different groups of agents with specific skills for these areas.
Some agents are more capable and adaptable than others, and can be used as blended agents. For
example, these agents may have a primary responsibility to handle inbound calls, but when the
inbound call volume drops, the dialler will send a message to these agents indicating that they have
been switched to outbound mode and start offering outbound calls to them. Where relevant,
systems will prompt a script for the outbound calls to run on the agent desktop and depending on
the call volume in the inbound queue, the agents will be switched automatically, improving
productivity. Results can include increased agent productivity, streamlined staffing, and improved
customer service. However, this process needs to be understood and managed carefully, as not all
agents are adept at dealing with both inbound and outbound calls, and if there is a constant
switching from inbound to outbound and back again, the agent may lose concentration and the
productivity may go down.
A structured blended environment, where agents are moved seamlessly and dynamically between
inbound and outbound, is used in only 12% of the UK contact centres which carry out both inbound
and outbound work. As usual, medium and large contact centres are more likely than smaller
operations to use this type of approach. Small operations are more likely to operate blending on a
manual, ad-hoc basis. A substantial proportion of respondents from medium operations use
dedicated teams to handle only either outbound or inbound.
Figure 40: Use of call blending by contact centre size
7%
18%11% 12%
62% 38%
67%
55%
31%
44%
22%
33%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Small Medium Large Average
Use of call blending by contact centre size
No, we have dedicated and separateinbound and outbound agents
Yes, but in an informal, manually-assigned way
Yes, in a formal and structured way(automatically switching based oncall volumes)
63
Looking at the effect on productivity, it is interesting to put the use and type of call blending against
key contact centre performance and operational metrics, such as average speed to answer, which is
- as usual - somewhat lower in formal, blended environments than in dedicated and more ad-hoc
environments, as agents are moved between tasks quickly and seamlessly.
Figure 41: Average speed to answer, by call blending environment type
Call blending environment
Average speed to answer (seconds)
Blending used in a formal and structured way 24
Blending used in an informal, ad-hoc way 38
Dedicated and separate inbound and outbound agents 33
64
MULTIMEDIA BLENDING
When considering call blending, it is also worth looking at how non-voice interactions can fit into a
blended environment too.
There is no general agreement within the industry on how best to deal with email, although there
are genuine reasons to encourage email/voice blending. On one side, there is a case made that
letting agents answer email makes the job more interesting for them, lowering attrition and
improving skills. The other side to this says that the skills required by email agents are different from
voice agents, and that it is difficult to find the agents to do both jobs. Both sides make sense
logically, and historically, of those contact centres which use voice/email blending, only around 1 in
5 have experienced problems finding the right staff for these types of role, a figure that decreased
each year that it was surveyed.
The great majority of respondents in most sectors allow at least some of their agents to carry out
both email and telephony. However, email requires certain skills, including grammar and
punctuation, which not every agent has, even with assistance from an email management system's
response template.
On average, 66% of agents in a blended multimedia environment are allowed to do both email and
voice work, a figure which had been growing year-on-year, but which has steadied recently.
Those in small and medium operations are much more likely to use the same agents to handle email
and telephony, probably because there is not the option to have the specialised teams found in large
contact centres, which are much more likely to have a dedicated group handling email.
Simply because a contact centre uses the same agents for email and voice does not mean that all
operations use the same level of multimedia blending. For some operations, multimedia blending is
a strategic decision which has been invested in with the right levels of technology and training being
provided. For others, it is a necessity, with agents encouraged to answer emails in slack call times.
Small and medium operations - which in the past may not have had sufficient email volumes or the
investment available to formalise the blending by forming a universal queue to deal with all types of
interaction - are now as likely to use a universal queue as the ad hoc method. Many larger contact
centres prefer to use dedicated email groups.
However, this preference of many larger contact centres to form specialised multimedia groups may
not provide the same levels of service. 2014 data indicated a formalised blending environment, such
as a universal queue, has a beneficial effect on email response times. Respondents using a formal
blended environment reported that 32% of emails are handled within 1 hour, with a further 41%
being dealt with inside a day.
The ad-hoc approach is less successful at very rapid response, with only 16% of emails having had an
average handle time within 1 hour, although a further 58% were handled in a day.
Dedicated email-only agents of the kind used in many larger contact centres answered an average of
17% of emails within the hour, with a further 48% handled within a day.
65
END-USER QUESTION #4:
WHAT BENEFITS ARE THERE FROM CALL BLENDING? WHAT EFFECTS CAN WE EXPECT TO SEE ON PRODUCTIVITY AND AGENT ENGAGEMENT?
Blending helps to ensure that every opportunity to please customers, boost
productivity and optimize resources is taken by prioritizing contact to suit
demand and resources. Being able to choose what you blend enables greater
output to be gained from the same number of staff, as they can be switched to different tasks.
Call blending generally takes a couple of forms:
Blending customer contact by direction - most typically increasing outbound calling when
inbound traffic is low, to ensure that agents are always active.
Blending different communication channels – for example combining telephone activity with
handling emails, text messages and social media.
Both forms of blending lead to increased agent activity and ‘raw’ productivity gains. But experience
shows that translating those gains into clear business benefits requires consideration of your
audience.
Simply increasing outbound calling when inbound traffic is low will enhance agent productivity but
won’t result in successful contact if you’re calling at the wrong time. Usually, the best time to make
outbound contact with your target is the same time as your inbound call peaks.
The most effective blending is achieved with a mix of contact channels. For example, during an
inbound call volume trough, send text messages or emails to attract more inbound calls. At the same
time ensure that your dialler campaigns are targeting the right audience, such as segmenting your
audience towards retired people during the day.
Blending ensures real progress is made, customers feel appreciated and ‘idle time’ is a thing of the
past.
66
MULTICHANNEL OUTBOUND STRATEGIES AND TECHNIQUES
While the vast majority of targeted outbound contact is carried out by agents, the opportunity exists
for automated outbound service to expand - such as sending reminders and notifications to
customers through an automated process - thus significantly reducing the cost to the business while
improving the overall customer experience. Many customers will choose to seek clarification or a
status update at some point in the buying process through making an inbound interaction. By
sending a pre-emptive outbound message, the business is proactively assisting the customer to
manage their interaction.
63% of UK contact centres do not use automated recorded voice messages for any purpose, whereas
SMS messages are used by over half of respondents, mainly for notifications and reminders. Only
31% of respondents do not send automated emails at all.
Email is used most for outbound customer satisfaction surveys, and 15% of respondents use
recorded messages (which will usually include an IVR session to capture customer input) for this
purpose as well.
Figure 42: Use of automated outbound communication
Businesses should consider how customers prefer to receive information: for example, SMS
messages for delivery notifications or appointment reminders may be preferred, whereas email may
be more suitable for order updates. Live outbound calls might be reserved for circumstances
requiring immediate action or for more complex situations.
10% 11%
15%
63%
26%
16%13%
45%
23%21%
26%
31%
0%
10%
20%
30%
40%
50%
60%
70%
Reminders (e.g. appointments,contract renewals)
Notifications (e.g. deliveries) Customer satisfaction surveys No automated interactionsthrough this channel
Use of automated outbound communication
Recorded messages SMS Email
--- Thought Leadership ---
The contact centre’s purpose has changed from transactional activity to one where your customers
can contact you for a wide range of reasons and through a variety of communication channels – you
can talk to your clients on the phone, by email, text message, chat, social media … the list goes on.
Your customers are demanding choice but your contact centre needs to balance meeting their
wishes, satisfying service needs and optimising costs.
Let’s take an example: your customer sends you an enquiry by email; how do you respond? If you
focus on the channel then it would seem obvious to reply by email. Whilst this may be an effective
response channel for a simple enquiry, when you are dealing with customer- and case-specific
enquiries an email response may open the door to a time-expensive series of emails. Text based
‘conversations’ with your customers pose other problems: If your client tweets about your business
how do you respond professionally and effectively in 140 characters in a public environment? How
do you verify the identity of a web chat caller? How do you link text messages to clients in your
Customer Relationship Management System?
The answer is often to shift the channel back to the telephone. If your client writes to you with a
complex question then the most effective customer service response is to pick up the phone and
talk.
All multichannel contact centres need a dialler and blending
Diallers in contact centres used to be about mass contact – for example using a predictive dialler to
support debt collection.
Today’s precision dialling technology means that requests for contact such as emails, tweets and
chats can be fed into the dialler to deliver a fast, personal response to customers whilst managing
and monitoring your most important resource – your agents.
rostrvm OutBound lets you blend SMS (text message, web, social, chat and email contact with your
OutBound dialling campaigns.
68
END-USER QUESTION #5:
ARE OUTBOUND AND CALL BLENDING SOLUTIONS SEPARATE FROM OTHER MULTICHANNEL INTERACTION, OR HAS THERE BEEN ANY MOVE TO INTEGRATE THESE CHANNELS TOGETHER?
Modern outbound and call blending solutions incorporate multichannel
interactions. Increasingly dialler campaigns combine telephone calls with text
messages and emails to cover the complete customer journey and experience.
For example, if your customer calls after receiving your email your contact centre agent needs to be
able to see the email concerned at the time of the call. Breaking down channel silos is a key
component of customer satisfaction.
69
THE ROLE OF SCRIPTING
The variable capability of agents is a contributory inhibitor to quality improvements, and the
increasingly complex and variable nature of agents’ work means that rigid, linear scripting is no
longer appropriate for many contact centres. One possible solution is to look at dynamic scripting
applications within an overall unified desktop environment, as well as understanding agent training
needs through call recording and analysis, and delivering the right training and in-call information.
DYNAMIC SCRIPTING
Scripting has a mixed reputation, rightly or wrongly, with inflexible outbound sales scripts being
particularly disliked. However, a dynamic scripting environment, sometimes referred to as a
workflow - where the help provided on the agent's screen changes depending on who the customer
is and what they actually want, applicable to both inbound and outbound work - can be beneficial to
agent and customer alike, supporting complex interactions where multiple systems and media are
required. Applications that define each step of the call interaction flow and present the relevant
data to the agent allows agents to take advantage of pull-down menus, on-screen buttons, pop-up
windows, inheritance mechanisms, and fill-in-the-blank parameters. Workflows guide agents
dynamically through dialogues with customers which change as required, while managing contacts,
controlling interactions, and updating databases. In a large operation, there may be distinct groups
of agents (e.g. a large dedicated sales outbound team) that use scripting whereas inbound agents
may not. Large contact centres, by their nature and also because of generally higher attrition rates,
may have more requirements to help new agents learn the ropes.
Insurance, transport and travel, public sector and outsourcing respondents are amongst the greatest
users of scripting, with around well over half of respondents from each of these sectors stating that
agents used the application (although not necessarily for 100% of calls, or agents - scripting may be
used widely in financial industries for legal disclaimers, terms and conditions, etc.).
Those in TMT, retail and outsourcing are most likely to use scripting to assist with outbound sales
campaigns, and those in transport and travel and outsourcing are amongst those most likely to use it
dependent on the call type or specific campaign.
None of the manufacturers responding to our survey use scripting in any way. While this appearance
of the absolute rejection of scripting in this sector should be placed in statistical context, and
manufacturing vertical market consistent large part of B2B sales and consumer helplines (which may
have a great variety of queries to answer), neither of which are typically suited to a heavily-scripted
environment.
70
Figure 43: Use and type of scripting, by vertical market
45%
42%
38%
33%
30%
29%
17%
14%
13%
13%
27%
19%
25%
17%
10%
14%
17%
14%
25%
14%
9%
13%
10%
21%
17%
10%
13%
8%
9%
15%
19%
33%
7%
17%
5%
13%
38%
12%
36%
35%
31%
33%
60%
43%
67%
67%
63%
50%
100%
51%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Insurance
Public Sector
Outsourcing
Transport & Travel
Housing
TMT
Utilities
Services
Retail & Distribution
Finance
Manufacturing
Average
Use and type of scripting, by vertical market
For all agents For new agents
For outbound agents Sometimes, dependent on call or campaign
No scripting
71
Small contact centres are less likely to implement scripting for all agents, but are almost twice as
likely to use it as assistance for their new agent intake. Larger contact centres are far more likely to
use scripting for outbound and also depending upon campaign or call type. As agent in smaller
operations will tend to have to handle a wider range of queries - as agent groups tend to be less
specialised than in larger contact centres - it could be viewed that the relatively low use of scripting
is disadvantaging them.
Figure 44: Use and type of scripting, by contact centre size
21%
30%
31%
27%
19%
10%
11%
14%
4%
8%
14%
8%
6%
14%
20%
12%
56%
50%
46%
51%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Small
Medium
Large
Average
Use and type of scripting, by contact centre size
For all agents For new agents
For outbound agents Sometimes, dependent on call or campaign
No scripting
72
Those respondents who used scripting were generally enthusiastic about the benefits they had
received. In particular, the assistance with compliance was particularly well-thought-of, as was the
consistency of data capture (as the same information is collected each time, with no short-cuts or
missed questions). 75% of respondents found that scripting was effective or very effective at
standardising interactions with customers - of course, this only helps quality of the 'average'
interaction is high to begin with.
Figure 45: The effectiveness of scripting
Secondary benefits around reducing call handling times (through a structured call flow) and
particularly as an agent training tool were also reported, with 58% of respondents also saying that
scripting was an effective tool for assisting with cross-selling and up-selling. Using scripting to
facilitate call wrap-up was still a net positive, but is seen as being less useful than other factors.
It seems fair to say from looking at these results that, while overly-scripted conversations are
frustrating for both caller and agent, a dynamic script or prompt which guides an agent through a
conversation and provides relevant information and suggestions is certainly worth considering and
tends to yield positive results, particularly if the level of ‘hand-holding’ that the script provides can
be tailored to the experience of the agent.
12%
15%
18%
20%
25%
33%
40%
45%
33%
40%
35%
50%
50%
45%
30%
25%
30%
30%
20%
8%
5%
8%
23%
7%
10%
5%
5%
10%
5%
4%
5%
5%
4%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Reducing agent training needs
Reducing call wrap-up time
Supporting cross selling and upselling
Reducing call handling times
Standardising interactions with customers
Capturing data consistently
Assisting with compliance and auditability
The effectiveness of scripting
Very effective Effective Neutral Ineffective Very ineffective
73
Despite these positive responses from users, scripting has had a bad name, engendered by the
uncomfortable-sounding scripts often used by outbound sales agents, particularly those from
offshore. Part of this problem is that the agent can sound robotic after delivering the same spiel
dozens of times, but perhaps as important, there is also the fact that outbound sales calls are
generally not well-received by customers, so anything that is associated with them (be it scripting,
offshore agents, silent calls, etc.) will have a negative connotation to overcome in any case.
Figure 46: Inhibitors to scripting
One of the biggest objections to scripting is that every conversation is different. While this is
certainly true, readers should be aware that contemporary solutions offer a whole spectrum of
scripting, from tight to loose, with the latter simply offering checkpoints within the conversation:
much of the negativity around scripting comes from familiarity only with the tight, constrained
version, and as this chart shows, contact centre management strongly believe that neither
customers nor agents like scripting (and seemingly, neither do the people at the top of the
organisation, who appear to hand down non-scripting edicts).
5%
5%
30%
32%
40%
45%
5%
23%
35%
48%
45%
35%
30%
40%
20%
18%
10%
7%
35%
27%
15%
2%
5%
3%
25%
5%
10%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Too expensive
Takes too long to produce the script
Strategic decision by senior management
Agents don't like it
Customers don't like it
Every interaction is different
Inhibitors to scripting
Strongly agree Agree Neutral Disagree Strongly disagree
74
Scripting is much more than a series of dialogue prompts, as it is now more about designing the
customer experience and journey for a particular type of interaction - for example, making an
insurance claim - using design sessions fitting together workflow, data sources and dialogue. People
with less experience of modern-day scripting often think the robotic ‘scripted conversation’
experience is what scripting creates, and indeed, many of the respondents to this questionnaire
believe that there are too many variables to anticipate within a conversation and that letting agents
make the best decision about the conversation is really the only way to handle things. However,
real-time decision engines are an example where online resources can be leveraged to work within
the contact centre as well: real time responsiveness in an online environment is achieved through
automated decision-making built on a set of business rules which identify pre-defined customer
profiles and the solutions, products and data that are suitable for presentation to the customer.
The design environment that new scripting solutions provides can use existing back end resources
without further development, so scripts no longer have to anticipate every possible alternative. Data
gathered during conversations, combined with customer profiling, trigger appropriate responses
which can be immediately presented to the customer, which is beneficial for customer and agent
alike. As such, dynamic scripting can be considered as one part of a group of solutions aimed at
reducing the complexity of the agent's desktop while improving the quality of the interaction. The
next section looks at optimising the quality of customer interactions through a unified agent
desktop, a key supporting feature of which is the ability to define call scripts, either as a flow of data
capture screens, or as a set of actual words or prompts. Simple, non-technical tools enable
supervisors and managers to define these call flows, the data required and the scripts/prompts to be
used.
END-USER QUESTION #6:
HAVE YOU NOTICED ANY CHANGES IN HOW SCRIPTING IS BEING USED? HOW DO BUSINESSES TEND TO USE THIS?
Thankfully scripting has moved on a long way from stilted word-for-word
speeches. Today’s ‘scripting’ systems support contact centre agents by providing
guidance on what they are expected to achieve and presenting relevant data at
the right time. Of course there are times in the process when exact scripts need to be used – such as
compliance statements in a financial transaction; graphical script presentation allows mandatory
statements to be highlighted.
Modern scripts also capture extended Management Information without adding overhead to the
agent’s core task. This means you can use champion-challenger techniques to identify the best
routes through a call to achieve your business aims.
75
CALL-BACK & CALL-ME REQUESTS
ContactBabel carried out a large-scale survey of the public that explored why customers notoriously
hate queuing to speak to a contact centre agent, yet seemed far more acceptant to wait in an actual
physical queue, often for a longer time.
Figure 47: Reasons given for dislike of contact centre queuing
Reason for disliking queue
Average score from 10 where 10 is “extremely frustrating”
% of public scoring this at a maximum 10
Not knowing how much longer you'll have to wait
8.7 61%
Repetitive announcements 8.0 45%
Having to restate account information already given earlier in the call
8.0 45%
Can't do anything else in the meantime
7.9 46%
The music you have to listen to 7.3 39%
Apart from the fact that customers have a lot of strongly felt reasons for disliking phone queues, the
key finding from this table is that 61% of the public hate not knowing how much longer they will be
waiting. This is less of a problem when waiting in a shop to speak to an assistant, as although they
cannot give you an exact statement of when someone can help, the queuing system allows a
customer to see how many people are ahead of them, to estimate their own wait time, and exercise
some level of control over the situation. This makes queuing psychologically easier for the customer,
even if the actual waiting time is significantly longer than it would be in a contact centre queue.
The phenomenon of 'Dentist-Chair Time' - time which seems to stretch out to infinity - is very much
active in the contact centre world. ACD statistics from thousands of contact centres over many years
indicate that an average wait time is around 20-30 seconds. However, when the public was asked to
estimate the time they usually (not exceptionally) spent waiting to speak to a contact centre, the
average answer was 11½ minutes - 27 times longer than the reality.
76
Clearly, trimming 10% off a queue time isn't going to make a lot of difference to the perception of
the caller, even though it may be a very difficult task for the contact centre to carry out. If customers
aren't informed of wait time, they may become discouraged and frustrated as hold time drags on.
This can lead to increased abandonment and even if the caller does decide to hold on, this
experience starts the call off badly leaving the agent with a lot to make up. Customers waste time
complaining about their experiences and may even ask additional questions on the call so that they
‘get their money's worth’.
If customers are given the estimated wait time, they may decide to abandon immediately or may
judge that the wait is acceptable and remain on the line to speak with an agent. This alleviates some
customer frustration but means that some of the callers which abandon may not call back - ever -
and it doesn't solve the fact that customers are still having to wait. One solution is to implement a
virtual queuing system, which not only provides customers with information about current queue
conditions but also presents them with various active options, such as remaining on hold or choosing
to be called back when it is their turn.
There are several different varieties of virtual queuing systems: the "First-In, First-Out" (FIFO) system
keeps the customer's place in line by monitoring queue conditions until the estimated wait time hits
a set target, at which point it intercepts incoming calls before they enter the queue, informing
customers of the likely wait time and offering the option of receiving an outbound call in the same
amount of time as if they had personally waited on hold.
At this point, customers choosing to remain on hold go directly into a queue. Customers who opt for
a call-back are prompted to enter their telephone number and possibly some extra details that can
be used for agent selection and skills-based routing, and are then asked to hang up. Virtual
placeholders keep the customers' places in line and the virtual queuing system launches an
outbound call to the customer at the agreed time. When the call-back is answered by the customer,
the system checks the right person is on the line and ready to talk. If this is the case, the call is
routed to the next available suitable agent, who handles it as a normal inbound call.
By replacing real hold time with this virtual version, customers are free to do other things, thus
removing four of the five problems that they have with queues - unknown queue times, hold music,
the inability to do anything else and repetitive announcements.
Scheduled call-back options differ from a FIFO experience, in that customers do not keep their place
in queue, but are called back at some time in the future that is more convenient for them (for
example, when they know they will be back at their desk and available to take a call).
77
There are several types of scheduled virtual queuing:
Datebook-type scheduling systems allow customers to schedule appointments for days in
the future, with times blocked-out that are unavailable for scheduling, and limiting the
number of call-backs available. This system also allows customers that reach a contact
centre out-of-hours to schedule a call-back during normal working hours
Timer scheduling systems promise a call-back after a specific amount of time, regardless of
queue conditions. While this ensures an on-time call-back for the customer, a surge in call
volume or staff reduction due to a shift change can create problems for the contact centre's
queue, lengthening wait times for other callers
Forecast-based scheduling systems offer appointments during times that are expected to
have low call volumes. These times may not be convenient for the customer, and the contact
centre runs the risk that their scheduling may be inaccurate.
Respondents offering a telephony call-back option were asked to state which types of call-back were
presented to callers. The majority of respondents that offered call-back functionality allowed callers
to request a FIFO call-back (i.e. acting as a placeholder in the queue), with a minority allowing
customers to specify a scheduled time.
Figure 48: Types of telephony call-back offered to customers (only from respondents offering telephony call-back)
Type of call-back
Proportion of respondents offering call-back that use this
FIFO (first-in, first-out) - holds the caller's place in the queue, then calls once they are at the front
75%
Forecast-based (called back at a time to suit the contact centre) 30%
Timed (called back at or before a specific time, regardless of queue conditions)
30%
Datebook (caller can specify a day to be called back on) 18%
More than half of respondents who offered call-back reported that FIFO placeholder call-backs were
far more requested than one of the delayed call-back types. On analysing the contact centre activity
type (i.e. sales or service), those callers making sales enquiries were more likely to want a
placeholder-type of call-back. This could possibly be explained by the differing states of mind of
customers calling to purchase something, or to make a query or payment. The former are more
likely to have chosen to call the contact centre to make a purchase that they are enthusiastic about,
and/or which is time-sensitive, and as such, want to speak to the business as soon as possible.
78
The use of a website 'call-me' button (which initiates an outbound call at a time specified by the
recipient) is weighted towards operations that carry out significant sales, but is present only in the
case of 13% of this year’s respondents. We would expect to see this continue to increase
significantly, and play an important part in providing customer support especially via mobile /
smartphone channels, especially if the call request is delivered to the agent with context for example
the details of the online interaction that resulted in a callback request and/or the customer’s history
and profile.
19% of respondents offer a telephony queue call-back option, and only around 10% of respondents
announce the position of the call in the queue. 23% of respondents use screen-popping
functionality, putting information about the caller and possibly their requirements on the agent
desktop as the call is delivered, which is more likely to be used in larger operations.
Figure 49: Use of website and queue call-back options and queue position announcements, by contact centre size
20%
22%
12%
10%
21%
9% 9%
12%
32%
25%
18%
7%
23%
19%
13%
10%
0%
5%
10%
15%
20%
25%
30%
35%
Screen-pop of customer detailsbased on CLI/IVR information
Callback option in the telephonyqueue
Callback button on the website Announcement of caller position intelephony queue
Use of website and queue call-back options and queue position announcements, by contact centre size
Small Medium Large Average
79
Respondents indicate that telephony call-back tends not to be universally available to callers, with
businesses only offering it after a certain period of wait time or once the queue becomes so long
that it triggers the functionality to be offered. 48% of respondents trigger call-back functionality
based on the actual time that the customer has spent waiting, with almost a quarter looking at the
estimated wait time based on ACD statistics. 28% of respondents use a mixture of actual and
expected queue time.
Figure 50: Triggers for offering call-back functionality
Actual time spent waiting in the queue
48%
Estimated wait time based on ACD statistics
24%
Both of these28%
Triggers for offering call-back functionality
80
42% of respondents using call-back state that it is offered after the caller has spent between one and
two minutes in the queue, although 21% say that it takes longer than five minutes before call-back is
offered, and which point many customers have already given up.
Figure 51: Typical queue time before call-back is offered
Less than 1 minute
12%
1-2 minutes42%
2-3 minutes17%
3-5 minutes4%
5-10 minutes13%
Over 10 minutes
8%
Don't know / NA4%
Typical (actual or estimated) queue time before a call-back is offered to a customer
81
Of those who are offered a call-back, most respondents report that fewer than a quarter of callers
chose this option. This may be because customers lack confidence that the business will call back
when they say they will, are relatively unfamiliar with the technology and/or do not have the call-
back option offered to them early enough and so have already abandoned the call.
Figure 52: Proportion of customers in telephony queue accepting offer of call-back
The previous finding is concerning, as call-back has great potential for both customers and
businesses: virtual queuing and call-back, when implemented - and explained properly to customers
- can be a win-win for both business and customer by:
Increasing customer satisfaction
Reducing average speed to answer
Reducing call abandonment rates
Reducing call lengths as customers should spend less time complaining and adding-on
unnecessary queries "while they're on..."
Reducing toll-free costs, as virtual queuing time does not incur telephone charges.
Less than 25%56%
26-50%26%
51-75%9%
Don't know9%
Proportion of customers in telephony queue accepting offer of call-back
82
Respondents offering telephony call-back functionality stated clearly that is was most useful for
managing call volumes and spikes in busy periods, and thus improving customer satisfaction and
experience. Being able to spread calls out over the day and allow callers to keep their place in the
queue - without actually having to queue – is seen by users as being of great use to both company
and customer.
Telephony call-back is not seen by businesses as having much of a positive effect upon reducing
agent stress and pressure to achieve key metrics, nor is it viewed in the context of maximising sales
opportunities from customers who would otherwise go elsewhere. No respondents considered it
particularly useful in reducing their freephone costs from customers who were queueing at the
businesses’ expense from considerable amount of time. It is not to say that telephony call-back does
not provide these benefits, just that respondents do not use call-back with these in mind.
Figure 53: Effectiveness of telephony call-back functionality
43%38%
9%5%
43%50%
48%
5% 14%
13% 13%
26%
27%27%
17%
64%59%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Managing callvolumes (e.g.
managing & avoidingspikes)
Improving customersatisfaction &
experience
Reducing agent stressand pressure to
achieve KPIs
Maximising salesopportunities
Reduce freephonecosts
Effectiveness of telephony call-back functionality
Don't know
Not useful
Somewhat useful
Very useful
83
Looking at the perceived effectiveness of website call-back functionality, it is worth noting that
respondents believed web call-back was particularly beneficial for improving the customer
experience, and thus satisfaction.
A significant proportion of respondents did not believe that it particularly affected call volumes, and
although a greater proportion noted its use in maximising sales opportunities, more respondents
surprisingly did not appreciate this benefit, which is arguably one of the main features of website
call-back.
Figure 54: Effectiveness of website call-back functionality
62%
31%23%
18%9%
31%
31%
23% 36%
18%
8%
31%
31%
27%
36%
8%
23%18%
36%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Improving customersatisfaction &
experience
Managing callvolumes (e.g.
managing & avoidingspikes)
Maximising salesopportunities
Reducing agent stressand pressure to
achieve KPIs
Reduce freephonecosts
Effectiveness of website call-back functionality
Don't know
Not useful
Somewhat useful
Very useful
84
Queue management and call abandonment rates
Offering an in-queue call-back option seems to have a positive effect on call abandonment rates,
with those doing so having an abandoned call rate around 20% lower than those who do not offer
this option, similar findings to previous year’s figures.
However, announcing the position of the caller in the queue seems to have much less of a positive
outcome, with very similar call abandonment rates.
A tentative conclusion can be drawn that call abandonment rates can be reduced best through
actual actions that benefit customers, such as calling then back. Simply providing them with more
information about the queue situation (which may be negative) could encourage customers to
abandon the call and ring back later. However, having this knowledge and actively being able to
choose what to do may well make the customer think more positively about the company (in that
they haven’t had to waste any great amount of time) but without a call-back option the customer
will not have managed to do what they set out to.
85
OUTBOUND IN THE CLOUD
While the pay-as-you-go, or monthly rental model which was one of the main attractions of cloud /
managed services is still an attractive proposition, it is far less important today than it was three or
four years earlier. Solution providers point out that CPE deployments can now often be paid for in an
Opex model, and that the movement to cloud is more about the functionality, agility and scalability
that this model offers.
It is also the case that operations have generally become more complex, certainly in the number of
channels offered and the proportion of interactions going through them. It is no longer possible for
many companies simply to focus upon voice, and the attendant need to provide a single view of
customer regardless of channel acts as an inhibitor to the siloed approach.
Looking at inhibitors to cloud-based solutions, concern over data security is still one of the most
important factors holding back the movement to cloud, although solution providers point out now
that the concerns that potential customers hold seem to be much more specific to their business,
rather than being a wide-ranging, generic fear of losing control.
One of the main differences to have occurred in the past three or four years is that cloud is now
seen as a genuine alternative to CPE for even the largest of enterprises, not just smaller operations.
Some solution providers have stated that they originally focused their product and marketing efforts
on the small and medium contact centre sectors, but have been surprised by the interest shown by
those with many hundreds (if not thousands) of agents, and have changed their strategy and
marketing approaches appropriately. There is an increased familiarity and trust in cloud-based
solutions, both in decision-makers’ personal (i.e. non-business) lives, and in the prevalence of cloud-
based enterprise applications, which is meant that cloud is no longer a particularly tough sell.
Solution providers note that there has been great growth in cloud solutions that support outbound
sales activity, where the pay-as-you-go model allows outsourcers to have some control over their
profit margins depending on the amount of work that they have, as well as benefiting from the
immediate use of added functionality that cloud-based solutions provide. It is common for cloud
providers to tightly integrate with CRM and contact management packages (e.g. Salesforce.com),
and adding dialling functionality supports the consultative style of selling, giving agents the
opportunity to read up about customers before they dial. This is not just the case for traditional
contact centres, but also for organisations which may not see themselves as having a contact centre,
but which carry out large amounts of sales communication with customers and prospects, albeit in
perhaps a more informal environment. Cloud-based solutions offer such enterprises the benefit and
efficiency of the contact centre world, without the high levels of sunk cost and operational and
technical expertise that a CPE solution would demand.
86
Of UK contact centres that use dialling technology, separate ContactBabel surveys indicate that
around 15-20% of dialler functionality is currently deployed through a cloud-based solution, with the
majority being on-premise solutions that are owned and managed by the users. Having said that,
58% of current on-premise dialler users expect to migrate to a cloud-based solution within two
years: while this is a rather dramatic increase which may not come to pass in reality, it certainly
shows that cloud is being taken seriously by users of outbound automation.
Figure 55: Dialler deployment model, by contact centre size
15% 13%7%
16%
15%
13%
13%
14%
84%
70%
88%
73%79%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Small Medium Large Unspecified Average
Dialler deployment model, by contact centre size
On-premisesolution - ownedand run by us
Managed servicesolution/privatecloud - ownedand run by athird-party for ourexclusive useDelivered viapublic cloud i.e.multi-tenancy
87
END-USER QUESTION #7:
HAS THERE BEEN MUCH UPTAKE OF CLOUD-BASED SOLUTIONS? IS THIS MAINLY DOWN TO COST, OR ARE THERE OTHER BENEFITS AS WELL?
Cloud-based solutions are proving to be very popular. In Rostrvm Solutions’ case
the majority of new customers have employed cloud-based systems over the
past year. However, the reasons for selecting cloud aren’t usually cost – indeed
cloud is often more expensive over the long term.
Key benefits of cloud are:
It’s quick and easy to turn on – so users can prove the business benefit of innovative contact
strategies without expending too much time or money.
It’s flexible – for example location-independent, work-from-home strategies are easy to set
up.
It’s up to date – cloud systems always make the latest software and services available to
users, giving a competitive advantage.
But cloud isn’t appropriate for everyone; for example, contact strategies that demand real-time
integration with on-site databases may not be suitable.
88
THE FUTURE OF OUTBOUND
While predicting the future with any sort of accuracy is by its nature impossible, there are various
ongoing themes and drivers that seem to point the outbound industry in a specific direction:
Driven by consumer complaints, there is likely to be a greater focus and regulation on
reducing or eliminating abandoned calls and enforcing accurate CLI/CLID. The forthcoming
EU GDPR (General Data Protection Regulation) is likely to make outbound more difficult
through restricting access to personal information
The boundary between inbound and outbound is likely to merge even further, with
outbound (including email, SMS and smartphone notifications, as well as telephony)
becoming an integral part of customer care
Outbound has already moved a long way away from the traditional cold calling model, and
increased knowledge of customer preferences and personalisation will be vital to
outbound’s rehabilitation with customers. While the majority of businesses currently use
some form of contact list segmentation (for example, looking at gender, age and customer
status), increasing amounts of customer data and the analysis of call outcomes means that
customer segmentation techniques will become more sophisticated and effective
There is likely to be a greater emphasis placed upon the automation of outbound contact:
understanding and analysing the perfect time to communicate with customers and their
preferences for channel may mean that the right agents are either unavailable or that this is
an unsuitable way to communicate with these customers
Although for many businesses it can be enough of a challenge simply to pair a live agent with
a live contact, we would expect to see leading-edge adopters consider personality matching
techniques in the future. Solutions exist where the personality profiles of customers and
agents can be identified, and the resulting customer and agent segmentation can allow the
business to have certain types of agent calling customers with the same personality type,
increasing empathy, customer satisfaction and revenue. We would also expect to see a
greater increase in personalised service by pairing agents and customers who have spoken
together before
As more people use a smartphone as their primary means of communication, businesses
must continue to re-evaluate the best means of communicating with customers based upon
the individual requirements and activities. An omnichannel approach, which can detect the
time and method by which the customer has contacted the business previously, can analyse
and predict the most successful ways of doing so in the future. It is likely that a multiple
channel approach will become popular, such as sending an email or SMS first, seeing
whether the customer interacts with it, and then placing an outbound call at an appropriate
time
89
The analysis of big data and the attendant timely recognition of potential opportunities and
triggers means that outbound communications can become welcomed as adding to the
customer experience: in effect, the business provides the customer with information or an
opportunity to communicate before they realise they even need it. This proactive approach
will be driven by a more sophisticated understanding of the customer’s thought processes
and closer analysis of customer preferences and the results that similar groups of customers
have experienced through this method before. Analysts talk about “the moment of
opportunity”, and although identifying and acting upon this with the right resource will be a
difficult journey for organisations to fulfil, it leads to potentially great benefits for both
customer and business
The inbound contact centre world has changed greatly over the past decade, with the
former focus on call duration having been moved very firmly into the background, with a
focus now being on customer satisfaction and first contact resolution. This change has been
as a result of many of the low value calls having been shifted to self-service (whether via
phone or web), leaving the more complex, high-value, high opportunity calls. We would
expect to see a similar pattern develop within the outbound industry: not only are there far
more personalised customer service calls being made, but also the opportunity now exists to
make personalised outbound sales calls as well. The wealth of information potentially
available about the customer, their likely intentions and even their type of personality gives
salespeople greater opportunity to engage successfully and make a sale. In tandem with
sophisticated customer profiling and solutions that screen out unanswered calls or
voicemails, the opportunity exists for outbound calling to be less of a scattergun, high-
intensity approach, and more about delivering relevant services and offers to the customer
at the correct time. Conjecturally, this may already be happening, as it is noticeable that the
agent attrition rate in outbound contact centres (historically extremely high), has seen a
considerable decrease in both the UK and US over recent years, suggesting that the nature
of the work has changed
There are great opportunities to improve the dynamism of scripting depending upon the
direction of the conversation and the customer profile. Static scripts should perhaps now be
used only for the most inexperienced of agents, and the success or otherwise of specific
dynamic scripts should be assessed and fed back into the system, allowing a greater
understanding of what works, to whom, and when
There will be greater sophistication in the monetisation and quantification of proactive
customer contact. Not only does the proactive engagement of the customer or prospect
reduce inbound call volumes and costs, but it also has a positive effect on revenue and
customer loyalty. At the moment, this can be difficult to quantify but we expect large
organisations to be able to analyse and justify in terms of profitability what can be a
considerable expenditure.
90
END-USER QUESTION #8:
ARE YOU SEEING A CHANGE IN THE NATURE OF OUTBOUND CALLING? HOW ARE YOUR SOLUTIONS DEVELOPING TO ADDRESS THIS AND WHAT CAN WE EXPECT TO SEE IN TERMS OF FUTURE FUNCTIONALITY FROM OUTBOUND SOLUTIONS?
Outbound calling used to be a ‘numbers game’ - make more calls and you’ll
make more sales, collect more debt etc. Outbound calling is changing to improve
customer contact and is incorporating Precision Dialling, which uses intelligence
gained about contacts to optimize dialling activities. Precision Dialling gives you the optimum chance
of making contact with your customers because you dial:
at the right time – the best time to call;
with the right message and channel – modern campaigns mix telephone, text messages and
email;
and the right people – making sure your calls are handled by appropriately skilled agents.
To use intelligence you’ve gained about each customer to programme your dialler you need good
management reports and records. Look at individual customer records and use your information
about when inbound callers called you to decide when to call them. Think about it – if a person calls
you at, say, 4pm it’s a good bet that they will be receptive to you calling them around that time.
One thing about outbound that’s unlikely to change any time soon is that, no matter how a person
contacted you – Twitter, email etc. – if it’s a complaint it’s ALWAYS better to respond using the
phone – social media is too public and a call resolves the issue faster than email.
91
ABOUT ROSTRVM SOLUTIONS
Rostrvm Solutions is a British software company. We design, develop and support rostrvm, a
modular suite of applications for managing and blending all of your inbound and outbound
communications, with clear and accurate reporting, available on any device.
For multichannel, blended contact
rostrvm provides intelligent, data driven, skills-based routing for your contact centre. Whether it’s
precision dialling, effective call queuing or multi-site, multichannel contact handling, everyone is
looked after.
To track your progress, rostrvm’s performance management tools deliver contact centre and back
office performance metrics. You can see business outcomes and real-time and historical
management information across all contact media and supporting business processes. Added to that
is call recording for quality assessment, training and compliance.
Another key rostrvm feature is Desktop optimisation. Streamlined desktops with integrated
functions save time and money. Our applications for the contact centre and back office will ensure
your operation is totally efficient.
The rostrvm suite is a truly integrated platform that supports increased functionality. It will overlay
your existing systems both in the Cloud and on your site so you don’t have to throw away your
investment but can enhance it cost effectively and with minimal risk. We provide support and
improve any IT environment and telephone system.
Our commercial frameworks will really benefit your business; rostrvm works on site or hosted in the
cloud and can be acquired on a capex, opex, subscription or pay-per-use basis. It’s completely
scalable and facilitates requirements like remote working.
As a privately-held company Rostrvm Solutions maintains a strong culture of independence, which is
increasingly rare in our market sector. We see our independence as a major benefit to our
customers and partners - it guarantees the openness of our technology and the objectiveness of our
approach and advice
Our dedicated team of experts, based in Woking, Surrey, have over 30 years’ experience in the field
and are skilled in delivering tailored solutions and a unique combination of benefits that leave our
competition standing.
As a result, we have very loyal and long-standing customers, who rely on us for the provision and
support of their core contact handling and process management functions.
Find out more
Visit www.rostrvm.com to read our case studies and learn more about our products, services and
customers.
Or why don’t you drop us a line or give us a call to arrange a meeting? We’ll make it worth your
while!
Call us on 01483 494 690. Email: [email protected]
Rostrvm Solutions Limited, Dukes Court, Dukes Street, Woking, Surrey GU21 5RT United Kingdom
ABOUT CONTACTBABEL
ContactBabel is the contact centre industry expert. If you have a question about how the industry
works, or where it’s heading, the chances are we have the answer.
The coverage provided by our massive and ongoing primary research projects is matched by our
experience analysing the contact centre industry. We understand how technology, people and
process best fit together, and how they will work collectively in the future.
We help the biggest and most successful vendors develop their contact centre strategies and talk to
the right prospects. We have shown the UK government how the global contact centre industry will
develop and change. We help contact centres compare themselves to their closest competitors so
they can understand what they are doing well and what needs to improve.
If you have a question about your company’s place in the contact centre industry, perhaps we can
help you.
Email: [email protected]
Website: www.contactbabel.com
Telephone: +44 (0)191 271 5269