Econometric modeling
3 likes1,298 views
Econometrics models sales and other key metrics like market share and brand awareness based on factors like price, distribution, advertising spend and more. The document discusses building an econometrics model including the needed data, factors for analysis like media activations and promotions, and how the model works. It can then be used to evaluate relationships between variables, forecast sales based on plans, compare ROI of different media, and analyze competitors. Building the initial model takes 1-2 months and then it requires quarterly updates and debugging over a year for ongoing use.
Econometric modeling
- 2. What is econometrics • How it works • Data that is needed for analysis SOM Model • Factor Analysis • ROI Evaluation Comparison with main competitors • Factor & ROI Forecast possibility 2 Agenda
- 4. Around a summer campaign sales are 20% up YOY. Why? Large discounts available? Competitors spending less? We moved TV spend to Door Drops? The weather was good? 4 How would you answer to this problem? Econometrics
- 5. Is an attempt to understand how our dependent variable: • Sales value • Market Share • Shops traffic • Brand awareness is changed due to variation of other parameters: • Distribution • Price • Adv spends etc. 5 Econometric model
- 6. Jan-13 Feb-13 Mar-13 Apr-13 May-13 Jun-13 Jul-13 Aug-13 Sep-13 Oct-13 Nov-13 Dec-13 6 How it works… Input Communication – TV – Print – OOH – Radio – Internet Competitors – Advertising – Price – Distribution Influence on result (sales/awareness) Input Market – Seasonality – Weather – Macro economics Consumer – Promo – Price – Distribution
- 7. 7 Econometrics in media helps to answer a lot of questions What the relationship is between sales, seasonality, price, distribution and Advertising How much Advertising influences sales value How long advertising effect lasts What results should we expect from increase/decrease of Advertising budgets What periods will give the highest Advertising ROI: •How to distribute Adv activity in time •What media mix to choose
- 8. Sales and Market by SKU for the whole category Sales data – Absolutes or SOM Price – Price index as deviation from average Category price Distribution *Data should by at least on monthly basis for 3-4 years. Geographical split is appreciated 8 Data that is needed for analysis Market Activations Sales Promo National Consumer Promo Launches of Competitor’s SKU Media Activations TV by type (image/promo) and by copy TV Sponsorship by channel Internet by type OOH by type (traditional, metro, interactive)
- 11. 11 SOM change with time
- 13. 13 Sales and National Consumers Promo
- 15. 0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0% Jan-11 Feb-11 Mar-11 Apr-11 May-11 Jun-11 Jul-11 Aug-11 Sep-11 Oct-11 Nov-11 Dec-11 Jan-12 Feb-12 Mar-12 Apr-12 May-12 Jun-12 Jul-12 Aug-12 Sep-12 Oct-12 Nov-12 Dec-12 Jan-13 Feb-13 Mar-13 Apr-13 May-13 Jun-13 Jul-13 Aug-13 Sep-13 Oct-13 Nov-13 Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14 Oct-14 Nov-14 Dec-14 SOM,% Non-promo factors TV copy 1 TV copy 2 TV copy 3 TV copy 4 TV copy 5 TV copy 6 TV copy 7 TV copy 8 ООН Non-Standart OOH/Metro Interactive OOH Sponsorship 1 Sponsorship 2 Pre-rolls Banners Context SMM Promo (Bonus+Discount) Promo (Sale out) Promo (DDP Drive) Promo (Buy&Get) NCP Competitor1 activations Competitor2 activations SOM 15 Factors contribution Negative influence of competitor 1 Negative influence of competitor 2
- 16. 16 Media and Promo factors output change over time Change of average Media & Promo factors output from year to year Detailed Media & Promo factors output with analysis of reasons Comparison of Media budget split with Media output split Average ROI evaluation
- 18. -5% 0% 5% 10% 15% 20% 25% 30% SOM,% Base Price Distribution 21 PositiveNegative Influence SKU launch Non Promo factors showing Price & Distribution influence SKU launch SKU launch
- 19. 22 Evaluation of Price change influence on SOM Aug’14 Price Index=0,96 We can evaluate what Price should be to achieve goal SOM under certain Distribution value
- 20. 23 Evaluation of Distribution change influence on SOM Aug’14 Distribution=76,8 We can evaluate what Distribution should be to achieve goal SOM under certain Price index
- 21. The critical Distribution level after which our Media activations start to work may be evaluated 24
- 23. 26 Output in terms of SOM for media Investments BybudgetorTRP’s ROI curves for each media channel by type based on modeling Detailed analysis of ROI including saturation level for each channel Regional analysis for local media channels (e.g. OOH) TV analysis by creatives
- 24. 0% 1% 2% 3% 4% 5% 6% 0,0% 1,0% 2,0% 3,0% 4,0% SOMChange,% Copy Awareness, % 31 IncasethereisBrandHealthTrackingforbrand–detailed analysisofcorrelationbetweensalesandawareness outputsmaybedone WithoutPriceandDistributionInfluence Correlation = 90% 0% 1% 2% 3% 4% 5% 6% 7% 8% SOM,% TRP’s TVC 1 TVC 2 TVC 3 TVC 4 TVC 6 TVC 7 TVC 8
- 25. Higher ROI Tactic Higher ROI Strategic Lower ROI Strategic Lower ROI Tactic 32 ROI Map for comparison of media channels WithoutPriceandDistributionInfluence Sponsorship M1 TV Promo SMM TV Sponsorship MTV Pre-rolls Interactive OOH OOH Context Banners IdeaX TV Metro OOH Promo (Bonus&Discount) Promo (Sale out) Promo (Buy&Get) ROI Changeofefficiency withtime ROI and duration of influence
- 26. ROI and duration of influence Higher ROI Tactic Higher ROI Strategic Lower ROI Strategic Lower ROI Tactic 33 ROI Map for comparison of media channels WithoutPriceandDistributionInfluence ROI Changeofefficiency withtime
- 27. ROI and duration of influence Higher ROI Tactic Higher ROI Strategic Lower ROI Strategic Lower ROI Tactic 34 ROI Map for comparison of media channels WithoutPriceandDistributionInfluence Sponsorship 2 TV Product TV Image Sponsorship 1 Pre-rolls Interactive OOH OOH Context Banners Metro OOH Promo (Bonus&Discount) Promo (Sale out) Promo (Buy&Get) ROI Changeofefficiency withtime
- 29. 36 For the comparison models of competitors’ sales are created ROI curves comparison for each media channel by type Local results analysis where needed Detailed Media & Promo factors output comparison with analysis of reasons
- 31. Brand plans Price policy for the next period – Your assumptions of category price policy forecast is appreciated Distribution changes plan *Data should by at least on monthly basis 41 Data that is needed for forecasting is your plans Market Activations plan Sales Promo National Consumer Promo Launches Media Activations plan With detailed split by media channels with weekly media pressure plan TV weekly plan by creative with specification
- 32. 42 Forecast should be updated each quarter Sales forecast based on market and media activation plans Should be updated according to actual price and distribution changes Detailed analysis of factors output and media recommendations Forecast of competitive environment
- 33. Media activations (including price seasonality factor) Most effective media mix based on • Price factor • Efficiency of channel • Saturation point of channel • Communication task (strategic or tactic) Appropriate level of distribution for • Media launch of new SKU • Rich of specific SOM (under other stable conditions) Appropriate price policy for Rich of specific SOM (under other stable conditions) 45 Econometricmodelmayhelpinplanning
- 34. 46 Timing & Cost
- 35. 2..3weeks •Brief from client •Data collection 1..2months •Data processing •Model creation •Preliminary results presentation 2..3weeks •Work with comments correction of model & presentation •Actualization of Model with new data Finalpresentation •Presentation of results •Recommend ations •Forecast for the future period 1year •Debugging of the Model •Quarterly reviews 47 Timing From2monthsforfinalModelcreation&Presentation 1yearfordebugging