Estimate the investment in media e briks infotech
- 1. Market Mix Modelling Estimate the effectiveness of investment in media
- 7. Agenda • Business application of Marketing Mix modelling • A case study • Strengths and weaknesses • Brief introduction to more advanced approaches: pooled regressions and structural equations
- 8. Making BP’s media dollars work harder • “Mindshare helped BP to make the most of their media investments across the many states of the USA.” • “BP engaged Mindshare to develop enhanced media investment strategies to maximise sales and boost revenue performance.” • “Drivers of performance were quantified (e.g. media, promotions, distribution, competitor effects) in seven USA states, over three years” • “Return on investment figures were calculated - both short and long term - for 40 campaigns.”
- 9. Marketing Mix modelling • Statistical methods applied to measure the impact of media investments, promotional activities and price tactics on sales or brand awareness • Used to assist and implement a marketing strategy by measuring: – Effectiveness: contribution of marketing activities to sales – Efficiency: short term and long term Return-On- Investment of marketing spend – Price elasticity – Impact of competitors
- 10. MMM How does it work? • A statistical model is estimated on historical data with sales as a dependent variable and list of explanatory variables as marketing activities, price, seasonality and macro factors • The simplest and broadly used model is linear regression: Salest 1 var 1t 2 var 2t ... t • The output of the model is then used to carry out further analysis like media effectiveness, ROI and price elasticity and to simulate what-if scenarios
- 11. Factors that could drive sales Advertising Promotions Competition TV Sponsorships Seasonality Radio Events Weather Print Price Economic Outdoor Adv quality Demographic Internet Distribution Industry data Merchandising Salest 1 var 2 var ... t 1 t t 2 Sales
- 12. MMM project process Set out objectives Data preparation -Define scope •Collect data -Discuss data •Validate, harmonize availability and consolidate data -Design data-warehouse •Present exploratory analysis to client Presentation Model development •Interpretation of •Estimation results •Diagnostics •Learning and •Calculate ROIs, Price recommendations elasticity and response curves
- 13. Case study • An energy company SPetrol wants to evaluate the advertising investments of its retail business in the US from 2001 until 2004. • Client’s questions: • How much have we made through advertising? • What is the return on investments of our media activities? • Which marketing drivers have had the greatest effect? • What’s the influence of price on our sales? • Are we optimally allocating our budget across products ?
- 14. Target variable
- 15. Advertising data • The performance of TV and radio advertising is expressed in terms of Gross Rating Points (GRPs) . A rating point is a percentage of the potential audience and GRPs measure the total of all rating points during and advertising campaign. – GRPs (%) = Reach * Frequency – Example: Let’s assume a commercial is broadcasted two times on TV 1st time on air 2st time on air GRPs 25% of target 32% of target 57% televisions are tuned in televisions are tuned in
- 16. Advertising data • Spetrol has deployed 5 TV campaigns over the sample with a total expenditure of 300 million $ • Each campaign lasted from 4 to 8 weeks • Is there any relationship between sales and TV advertising?
- 17. Carry over effect of TV
- 18. Carry over effect of TV • The exposure to TV advertising builds awareness, resulting in sales. • ADStock allows the inclusion of lagged and non linear effects ADStockt ( ) GRPt ADStockt 1 0 1 • Alpha is estimated iteratively using least squares. The estimate is then validated by media planners
- 19. Advertising data 300 M 164 M 160 M TV Spend Radio Outdoor
- 20. Below the line promotions • It may include – sponsorship – product placement – sales promotion – merchandising – trade shows • Usually represented by dummies (variables equal to 1 when a promotion takes place and 0 otherwise)
- 21. Below the line promotions Sponsorship World Rally Championship Sale promotion Sale promotion 5% Discountt
- 22. Price
- 23. Seasonality August seasonal dummy 5% Discountt Peaks every year Sale promotion in August
- 24. Exploratory analysis Scatter plot Unit root test 32 Histogram and desc stats Correlation matrix Series: SALES 28 Sample 1 209 Observations 209 24 Mean 154403.1 20 Median 153960.2 Maximum 183102.5 16 Minimum 125997.0 Std. Dev. 9476.290 12 Skewness 0.053546 Kurtosis 3.456209 8 Jarque-Bera 1.912312 4 Probability 0.384368 0 130000 140000 150000 160000 170000 180000
- 26. Estimated equation Salest = 167412 + 168* AdStock(GRPsTVt,0.75) + 161* AdStock(GRPsRadiot,0.35) + 166* AdStock(Outdoort,0.15) + 580* PromotionDummyt + 6507* Seasonalityt + -12631* Pricet + Errort
- 27. Model diagnostics • Model: – Significant F-stat and high R-squared • Variables: – Significant T-stats – Coefficients must make sense – Variance inflation factor low • Residuals: – Normality (Jarque-Bera) – Absence of serial correlation ( Durbin Watson, correlogram)
- 28. Residuals diagnostics 16 Series: RESID 14 Sample 1 209 Observations 209 Durbin Watson = 1.69 12 Mean -2.31e-11 DW>2 positive autocorrelation 10 Median Maximum -66.11295 8049.987 DW<2 negative autocorrelation 8 Minimum -11378.69 Std. Dev. 3612.711 6 Skewness -0.158326 Kurtosis 2.624286 4 Jarque-Bera 2.102443 2 Probability 0.349511 0 -10000 -5000 0 5000 y y ˆ ˆ
- 29. Estimated factors contribution to sales Fitted Salest = estimated Intercept = 167,412 Can be interpreted as Brand Equity: •Volume generated in absence of any marketing activity •Indicator of the strength of the brand and users’ loyalty
- 30. Estimated factors contribution to sales TV Contributiont(000’ Gallons) = coefficient *Adstock(TV)t Fitted Salest = 167,412 + 168* TVt + 161*Radiot + 166* OOHt + 580* Promotiont
- 31. Estimated factors contribution to sales in August Peaks every year Peacks every year in August Fitted Salest = 167,412 + 168* TVt + 161*Radiot + 166* OOHt Equity = Promotiont + 6507* Seasonailityt + 580* estimated Intercept = 167,412 Can be interpreted as Brand Equity
- 32. Estimated factors contribution to sales Fitted Salest = 167,412 + 168* TVt + 161*Radiot + 166* OOHt + 580* Promotiont + 6507* Seasonailityt - 12631* Pricet Negative price effect
- 33. Marketing mix (sample output)
- 34. Estimated factors contribution to sales
- 35. Estimated factors contribution to sales N TotSalesContribution coeff Factori i 1
- 36. Estimated factors contribution to revenue N ibution coeff Factori Pr icei Tot Re venueContr i 1
- 37. ROI TOT Re venueContr ibution ROI TOTCost
- 38. Does it really make sense? TheDiminishing in more I invest media, returns I sell the more
- 39. Response curves NegExp a (1 exp(b GRPs )) S a (1/(1 exp(b (GRPs mean(GRPs )))) Taking into account diminishing returns
- 40. Price elasticity • Assumption: constant elasticity across the sample which implies a linear relation between volume and price • By using the coefficient of the regression, it is possible to derive an estimate for price elasticity: – Price coefficient = -12631 – Average price = 1.51 $ – Average volume sales = 154,000 Gallons Avg Pr ice A 10% drop in price Elasticity * coeff 0.12 increases sales by 1.2% AvgSales
- 41. Dynamic price elasticity Elasticity changes with price 200,000 Weekly Volume and $ Sales vis-à-vis price of 1.75L 180,000 Volume (9L Cases) 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 Price (750 ml) 9 20.0 11 13 16 18 25 27 29 10 12 14 15 17 19 21 22 23 24 26 28 30 Volume Elastic (>1): Demand is sensitive to price changes. Estimated through non Inelastic (<1): Demand is not sensitive to price changes linear regressions
- 42. Client’s questions How much have we made through advertising? • 1 billion $ driven by TV • 500 million $ due to radio • 200 million $ generated by Outdoor and promotional activities Investments in media generated 1.7 billion $ in revenue
- 43. Client’s questions What is the return on investments of our media activities? For each dollar invested in TV you get 3.5 dollars back
- 44. Client’s questions What’s the influence of price on our sales? A 10% drop in price increases sales by 1.2%
- 45. Are we optimally allocating our budget across products ? Maximum Optimal GRPs Marginal Return Over Optimal GRPs Point of Saturation Sub –Optimal GRPs Maximum Average Return Invest more in Radio and less in OOH
- 46. Marketing Mix – Sample Output Marketing mix (sample output) 45 Carry Over Effect 5000 Diminishing Returns 40 4500 35 4000 30 Promo TV Saturation 3500 Weekly GRPs Weekly Sales 25 3000 20 2500 Current Optimal 15 2000 1500 10 1000 5 500 0 0 0 20 40 60 80 100 120 140 160 180 Week1 Week2 Week3 Week4 Week5 Avg. Weekly GRPs Diminishing Returns is the point were spending additional GRPs does not results in additional Simultaneous Effect sales. Carry Over Effect (Ad Stock) relates to the Volume residual effect of an ad. When all the components are layered on Base Base/Seasonal TV/Radio/Print Direct Marketing Rates/Promotions sales, it is clear what drivers contribute to sales Time and when and their Simultaneous Effect.
- 47. Pros and cons • Simple and intuitive • Correlation doesn’t imply • The outcome is backed by causality qualitative expertise and in • Risk of spurious regressions field research especially when modelling • Constructive way of running in levels different scenarios and • Model highly depends on evaluating past variables chosen performance • Poor in forecasting • Better with granular data • Very successful method – high turnover
- 48. Spurious statistics • A high correlation between sales and TV could mean: Sales Media – Either media causes sales – or sales causes media – or a third variable causes Income both sales and TV What is the truth?
- 49. Non sense correlations • Some spurious • On the other hand, a correlations: low correlation doesn’t – death rate and rule out the possibility proportion of marriages of a strong relation: Corr = 0.95 Corr = 0.0 – National income and sunspots Corr = 0.91 – Inflation rate and accumulation of annual rainfall •Correlations must support a theory •Calculate correlations both in levels and differences •Always look at scatter plots
- 50. What variables should have been included?
- 51. New media • Digital Marketing – Display Marketing – Search Engine Marketing (SEO & PPC) – Affiliate Marketing – Mobile Marketing – Social Media
- 52. New media • Data availability – Impressions – Clicks – Post event activity – Bespoke engagement metrics • Example of a tracking centre: – Double-click
- 53. Alternative methods • Linear regression • Logistic regression • Discriminant analysis • Factor analysis • Cluster analysis • Structural equations modelling
- 54. Pooled regressions Sales Local media Nat media Local Price California California USA California + ... + error sa Nevada Nevada USA Nevada + ... + error Oregon Oregon USA Oregon + ... + error
- 55. Pooled regressions example 1. SalesCalifornia = c11*TVCalifornia + c12*TVOregon+c13*RadioCalifornia +c14*RadioOregon + ErrorColifornia 2. SalesOregon = c21*TVCalifornia + c22*TVOregon+c23*RadioCalifornia +c24*RadioOregon + ErrorOregon TVC SalesC c11 c12 c13 c14 TVO C Sales c Radio O 21 c22 c23 c24 C O RadioO Media effect is also tested across regions
- 56. How advertising effects consumers? Understanding: – the process by which advertising affects consumers – How the effects of advertising are spread over time – The role of different media – The role of competitors
- 57. The purchase funnel • A basic process that leads to the purchase of Awareness a product consists in: – Awareness – costumer is aware of the existence of a product Consideration – Consideration – actively expressing an interest in the company – Purchase Purchase
- 58. Working on survey data • A sample of the target audience is interviewed about brand awareness, consideration and choice • Research agencies provide awareness, consideration and purchase time series in % terms – i.e. A purchase of 10% means that 10 out of 100 interviewed people purchased the product
- 59. Testing the purchase funnel Awareness Consideration Purchase Advertising first exercise its influence on awareness. Via awareness there is an effect on Media consideration which drives the consumer to purchase
- 60. Testing the purchase funnel • Awarenesst=c11+c12*TVt+c13*radiot+c14*OOHt+error1t • Considerationt = b1*awarenesst + c21 + error2t • Purchaset = b3*Considerationt + b2*Awareness +c31 + error3t a1,a2,a3 must be insignificant to confirm theory Const 1 a1 a2 Awart c11 c12 c13 c14 1t b 1 a3 Const c21 0 0 0 TVt 1 Radiot 2 t b2 b3 1 Purcht c31 0 0 OOH 3t 0 t
- 61. Agenda • Business application of Marketing Mix modelling • A case study • Strengths and weaknesses • Brief introduction to more advanced approaches: pooled regressions and structural equations
- 62. References