Home >Marketing >Measuring the “carryover” effects of pricing

Measuring the “carryover” effects of pricing

Date post:01-Nov-2014
Category:
View:107 times
Download:2 times
Share this document with a friend
Description:
Standard pricing models evaluate point in time consumer price sensitivity, looking at the relationship between consumption changes vis-à-vis pricing or promotional changes one week at a time. Consumer price sensitivity is a more gradual phenomenon that builds over time, with shocks on consumption reverberating several weeks following a price change. This is true for own as well as competitive pricing effects- it is easy to underestimate how much impact a competitor’s price has on a brand if one is just looking at one week at a time- the shock carries over or “persists” in later purchase cycles, regardless of price stabilizing to a new level or reverting back. We use a Dynamic Time Series model (Vector Autoregression) to capture the contemporaneous as well as lagged effect of pricing and promotions (own as well as competitive) to capture this “carryover” effect. This can help prevent marketers from underestimating the extent and duration of own as well as competitive pricing action.
Transcript:
  • 1. Pricing Using Dynamic Demand Modeling Measuring the Carryover Effects of Joe Sakach Director - Consumer & Customer Insights Joy Joseph Vice President, Analytics
  • 2. 2 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 2 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Overview Standard pricing models evaluate point in time consumer price sensitivity, looking at the relationship between consumption changes vis--vis pricing or promotional changes one week at a time Consumer price sensitivity is a more dynamic phenomenon that extends over time, with shocks on consumption reverberating several weeks following a price change We use a Dynamic Time Series model to capture the contemporaneous as well as lagged effect of pricing and promotions to capture this carryover effect This can help marketers develop more effective promotion plans that maximized effectiveness of own promotions and minimize the impact of competitive promotions
  • 3. 3 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 3 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Marketers Perceptions on Pricing (Based on Survey Responses from SymphonyIRI Clients) 83% of responders used some level of econometric analysis including elasticity models to drive pricing decisions 50% managed price (promotional and everyday) gaps to key competitors but only 30% accounted for post-event factors influencing pricing effects and lifts 71.4% 28.6% Evaluate Pricing & Promotional Lifts within week of change Adjust lifts for Forward Buying & Repeat Purchases
  • 4. 4 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 4 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. A Primer on Contemporary Pricing Models Current pricing models capture all the major drivers that impact sales, but they correlate the impact of these drivers on volume sales within each week in the analysis time period separately Volume Sales in Week Base & Promoted Price For Week Quality Trade For Week Special Trade Programs (Multiples, Bonus Packs, Retailer Specific events) in Week Competitive Price & Trade For Week Category Trend, New Product Launches and Seasonality For Week
  • 5. 5 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 5 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Why is this a problem? because promotional and pricing effects carry over well-beyond the week of the event itself and looking at just the event week will: Understate or overstate impact of the event Not reveal any insights on optimal time gaps (hiatus) between events, leading to inefficiency Initital Promotional Lift generated by TPR (Includes core buyers and incremental triers) Promotional Events are followed by a dip resulting from forward purchase acceleration effects (a.k.a pantry-stocking) Short-term Post-Promotional Dip could be followed by lifts from incremental repeat buyers inlater purchase cycles
  • 6. 6 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 6 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. So What About Competitive Cross Pricing Effects? Competitive pricing effects also carry over beyond the week of the competitive event driven by Consumers taken out of normal purchase cycles due to competitive pantry- stocking Consumers retained by competing brands through repeat purchases Carryover negative effects from proportion of switchers retained by competitor in later purchase cycles Initial negative impact from competitive promotion Hiatus period before repeat losses set in
  • 7. 7 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 7 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Solution: Dynamic Time-Series Modeling Dynamic Time Series modeling, especially Panel Vector- Autoregressions look at the effects of drivers over multiple time-periods Output from these models measure the effect of consumption shocks due to drivers over consecutive time periods Lag in Weeks StandardizedVolumeImpact
  • 8. 8 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 8 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. So whats a Panel VAR? Ability to measure continuous demand over a period of time VAR models can handle only time-series data, Panel VAR can leverage time periods across multiple geographies Week 1 Week 2 Week n Geo 1 Week 1 Week 2 Week n Geo 2 Week 1 Week 2 Week n Geo 3
  • 9. 9 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 9 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Application to Trade Promotion Planning While just looking at the promotion week, Feature & Display could be most profitable but other tactics may yield overall better cumulative incrementality The post-promotion dip driven by forward-buying is not an ideal period to be promoting again as consumers are still working on the pantry they stocked up in the previous promotions Ideal point to promote again is when the repeat buying and/or forward- buying starts to taper off- in below instance it would be around week 6
  • 10. 10 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 10 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Application to Competitive Response Strategy Successful competitive promotions should not be immediately responded to, but rather leverage modeled insights to determine the optimal hiatus during which counter-promotions will not obtain the intended benefit as consumers already have a stocked up pantry and your category is not on their shopping list
  • 11. 11 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 11 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Case Study
  • 12. 12 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 12 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Background Study objectives Understand promotional impacts beyond the week of execution Use results to inform a more effective trade promotion strategy Analysis framework Market-level data 4 years of weekly data Competitive set of both intra- and inter- category products Control for other major drivers (e.g. Advertising and Economic factors) that are not reported here
  • 13. 13 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 13 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Demand Drivers Summary Managing competitive threat is key for this segment -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 Category Comp X-Category Comp Non-Promoted Price Distribution Own Trade
  • 14. 14 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 14 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Impulse Response Curves - Price Promoted discounts show strong forward-buying behavior with minimal repeat purchases. -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0 2 4 6 8 10 12 14 16 Weeks Discount Depth Forward-Buying Dip Minimal Repeat Purchases Initial Promo Lift
  • 15. 15 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 15 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Cumulative Trade Impact Cumulative impact of promotional price changes are significantly lower than week zero impact would indicate 0 0.05 0.1 0.15 0.2 0.25 Cumulative Impact Week 0 Impact
  • 16. 16 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 16 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Impulse Response Curves - Trade Lagged effects for F&D and Disp offset initial promotional lifts Feature generates significant repeat purchasing Optimal hiatus of 8-9 weeks between promotions -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0 2 4 6 8 10 12 14 16 Weeks Feat & Disp Feat Only Disp Only Forward-Buying Dip Repeat Purchases Optimal Hiatus ~ 8-9 wks
  • 17. 17 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 17 Copyright SymphonyIRI Group, 2011. Confidential and Proprietary. Cumulative Trade Impact Cumulative impact of Features exceeds that of Feat & Disp Initial promotion lift of Displays is almost offset by forward buying impacts 0 0.05 0.1 0.15 0.2 0.25 Disp Only Feat & Disp Feat Only Cumulative Impact Week 0 Impact
  • 18. 18 Copyright SymphonyIRI Group, 2010. Confidential and Proprietary. 18 Copyright SymphonyIRI Grou
Popular Tags:

Click here to load reader

Embed Size (px)
Recommended