MAXIMIZE ADAPTABILITYTHROUGH DATAMANAGEMENT
DOCUMENTATION
By Dan [email protected]
@kiwidankun@dqmatters
SESSION OBJECTIVES
1. Clarify the similarity and distinction between Business Analysis and Data Management domains
2. Identify which Data Management tools enable the Business Analyst to be efficient and highly adaptable
3. Identify who has the information you need to get the job done
This session will use the data management lens to identify key datamanagement resources that business analysts can use to ensure theyare agile and nimble.
(c) Dan Myers, DQMatters.com 2016 2
A FEW ASSUMPTIONS BEFORE WE GET STARTED
• This presentation is geared to provide a high-level overview of the data/informationmanagement domain and synergies, similarities, and differences with the business analysisdomain.
• I assume that you know business analysis and its variants (e.g. Strategic Planning, BusinessArchitecture/Model Analysis, Process Design, Systems Analysis)
• Different domains may claim ownership of some activity which is an area of overlap(similarity) between other domains, but I believe we can still learn a lot from from each other
(c) Dan Myers, DQMatters.com 2016 3
DEFINITION OF BUSINESS ANALYSIS
What is Business Analysis?
Definition: “Business Analysis is the practice of enabling change in an organizational context, bydefining needs and recommending solutions that deliver value to stakeholders. The set of tasksand techniques that are used to perform business analysis are defined in A Guide to theBusiness Analysis Body of Knowledge® (BABOK®Guide).”
(c) Dan Myers, DQMatters.com 2016 4
Definition: “A business analyst is someone who analyzes an organization or business domain (realor hypothetical) and documents its business or processes or systems, assessing the business modelor its integration with technology.” (Wikipedia, 10/2016)
DEFINITION OF DATA MANAGEMENT
From Data Administration and ManagementAssociation (DAMA): What is DataManagement?
“Data management (DM) is the businessfunction of planning for, controlling anddelivering data and information assets. Thisfunction includes: The disciplines ofdevelopment, execution, and supervision ofplans, policies, programs, projects, processes,practices and procedures that control, protect,deliver, and enhance the value of data andinformation assets.” DAMA DMBOK, p.4.DMBOK 2 download URL
(c) Dan Myers, DQMatters.com 2016 5
SIMILARITY AND OVERLAP BETWEEN BUSINESSANALYSIS AND DATA MANAGEMENT?
(c) Dan Myers, DQMatters.com 2016 6
Areas of Overlap: Both define needs and recommend solutions, document business processes, query andanalyze data, and data anomalies and manage projects…etc.)
Business Analyst Data Management RolesA Collect/document systems requirements Focus on data requirements, not functionalB Develop Business Models Data Modeler: Develop Logical and then Physical Data ModelsC Document processes, systems Architect: Document, Design and Govern Systems and DataD Query, analyze and create reports Data Tester: Test and ensure data at rest; reports are correctE Analyze data anomalies, document data
quality issues, report data qualitydefects
DBA: Administer data and access securityDQ Analyst: Assess and report DQ levels, conduct root causeanalysis, propose resolution
F Document DM requirements DM Analyst: Implement DQ rules, controls, define datagovernance policies and procedures
G Project Management Manage DQ remediation projects
OpenDiscussion
DM RESOURCES AVAILABLE TO YOU DURING THE SDLCWaterfall SDLC
Initiation
Requirements
Design
Build
Test
Train
Deploy
Maintain
MetadataManagement
Business Glossary
Glossary, Search,Lineage, Impact Analysis
Reuse of Models/Data/Definitions,Clarity afforded Developers
Technical metadata developed andstored in enterprise repository
Pseudo code for SystemsIntegration Testing, Clarity of Test
Cases
Definitions and lineage documentssupport training
Comprehensive communicationabout what and where new data
exists
Glossary and lineage facilitaterelease processes
Data Governance
Executive support throughgovernance committee
Stakeholder groups known androles and responsibilities
documented
Data Architecture standards exist &rules followed
Technical naming conventions &use of non-Production data to unit
test
Stakeholder groups defines UserAcceptance Testing andcommunication targets
Governance team explainsescalation processes for data
quality issue escalation
Stewardship; cross dept/silocommunication ensures successful
rollout
Ensures currency of documentationthrough governance audits and
committee involvement
Information/Data Quality
Correct Data=> CorrectQuestion=>Correct Project
Reusable DQ Rules for FormValidation, IT completeness
controls
Design for reuse, consistency andflexibility
Use of real-life test data ensuresrealistic unit testing
System Integration Testing includesuse of Dimensions of Data Quality
Training includes data qualityawareness components and fitness
for use discussion
Communication about appropriateuse of data & context provided
with release & DQ control reports
Data Quality Scorecards for keymeasures and dimensions over
time
The focus of this presentationis to explain what DataManagement resources areavailable to you as a BusinessAnalyst.
Resources can be indexed bythe phases of the SDLC andsubdivided into the domainsof Data Management, such asMetadata Management, DataGovernance and InformationQuality.
(c) Dan Myers, DQMatters.com 2016 7
DEEP DIVE OF A FEW DATAMANAGEMENT TOOLSHere is a list of some of the tools used in DataManagement. Note that this is by no meanscomplete, but rather illustrative.
1. Metadata Managemento Metadata Repository
2. Data Qualityo Data Profiler
3. Data Modeling & Designo Data Modeling Tool
(c) Dan Myers, DQMatters.com 2016 8
DataManagement
DataModeling& Design
DataArchitecture
Data Storage &Operations
Data Security
Data Integration&
InteroperabilityDocuments &
Content
Reference &Master Data
DataWarehousing &
BI
Metadata
Data Quality
DAMA, DMBOK2 Knowledge Area Wheel
IDENTIFYING THE PEOPLE YOU NEED
(c) Dan Myers, DQMatters.com 2016 9
DataManagement
DataModeling& Design
DataArchitecture
Data Storage &Operations
Data Security
Data Integration&
InteroperabilityDocuments &
Content
Reference &Master Data
DataWarehousing &
BI
Metadata
Data Quality
Database Administrationteam, Data modelers
Enterprise, System, and DataArchitects
• DBAs• Off-site document storage team (e.g. Iron
Mountain)• Records Management dept• Legal dept
• Information Security team• Records Management dept• Legal dept
• DW, BI team• Data as a Service team (e.g. FTP, API,
web-services
• Sharepoint, Wiki, Document storageteam
• Metadata Repository, Metadatamanagement team
• Customer or Product domainmanagement team
• Sometimes in Marketing dept
• Data Warehousing;Business intelligencedepartment
• Data consumers, knowledgeworkers; data scientists
• Data Governance team• Metadata tool
administrator• Data stewards
• Data Quality score cards• DQ Center of Excellence• DM Council
METADATA REPOSITORY
(c) Dan Myers, DQMatters.com 2016 10
PrivacyLabels (PII)
TechnicalMetadata
Tables &Columns
Data Marts,Warehouses
Data Lineage &Impact Analysis
BusinessMetadata
BusinessJargon
Acronyms
ReferenceCodes
DataGovernance Stewardship,
Ownership
Data QualityLevels
Original Purpose;Fitness for reuse
LearningTools
BusinessGlossary
Metadata Repository isbasically a digital card catalogwhere documentation is storedabout an organization’s data.
DATA LINEAGE EXAMPLES
(c) Dan Myers, DQMatters.com 2016 11
DATA MODELING TOOLS
12
How can you easily communicate a concept? Usean illustration or model. Speed that up with a toolthat outputs your model in a familiar notation.• Data Modeling Styles
o Information Engineering (IE), IDEF1X,Object Role Modeling (ORM), UnifiedModeling Language (UML)
• Levels of Modeling (from DAMA DMBOK, Chapter 5)
o Conceptual Model- A conceptual data model is a visual, high-level perspective on a subject area ofimportance to the business. It contains only the basic and critical business entities within a given realm andfunction, with a description of each entity and the relationships between entities.
o Logical Model- A logical data model is a detailed representation of data requirements and the businessrules that govern data quality, usually in support of a specific usage context (application requirements). Alogical data model often begins as an extension of a conceptual data model, adding data attributes to eachentity.
o Physical Model- A physical data model optimizes the implementation of detailed data requirements andbusiness rules in light of technology constraints, application usage, performance requirements, andmodeling standards.
DATA QUALITY-
(c) Dan Myers, DQMatters.com 2016 13
Data Profiling process of using software tools tocollect qualitative and quantitative informationabout the characteristics of a dataset (such as:count of cells null for a column,average/min/max/mode for numeric data,distinct list of categorical data).
Dimensions Of Data Quality
http://dimensionsofdataquality.com
Completeness Accuracy
Lineage
Timeliness Representation
Integrity
Currency
Consistency
PrecisionPrecision
Data Quality “The quality of data is defined bytwo related factors: how well it meets theexpectations of data consumers (how well it isable to serve the purposes of its intended use oruses) and how well it represents the objects,events, and concepts it is created to represent.” -
--Laura Sebastian-Coleman*
*Measuring data quality for ongoing improvement : a data qualityassessment framework, Sebastian-Coleman (2013), p. xxx (28)