Marketing mix modeling (MMM) is an analytic approach, throughhistorical information, that quantifies the impact of marketingelements on sales or market share. This practice has been inexistence since 1949 (Source:Wikipedia) Thanks to thedevelopment of technology and the availability of in depth data,MMM has been witnessing a quantum progression in terms of theapproach that's been adopted. MMM as a concept was used byconsumer packaged goods (CPC) industries only, a few years ago,but in recent times, its popularity is spread to other industries aswell. As the penetration of MMM in various categories and marketsexpanded, the discipline also witnessed advancements.Increasingly MMM is considered a combination of art and science.Domain expertise in marketing and media is one of the necessaryfactors in successful implementation and new methodologies haveexploited this connection
MMM, helps us to quantify the effect of marketing inputs onbusiness metrics- sales/market share. This is usually done througha statistical technique called "regression". Till recently, regressionwas used by treating most of the marketing variables as linear andmedia, especially TV as non linear. By treating the influence of TVas non linear, it embraced the theory of diminishing returns on TVdeployment- as the input on TV increases the sales/market shareincrease at an increasing rate in the beginning and then increase ata diminishing rate. This approach helped the user to set an optimalTV weight for a desired sales/market share. Increasingly thisapproach was followed when TV was considered as a lead mediumand also this technique was used for setting TV weights.
As the usage of MMM expanded, the utility of the same also expandedbeyond just TV weight setting. The approach found increasing usage indetermining the type of distribution outlet that influence sales/marketshare, determination of optimal price for the brand and the interrelationship of media especially between the paid, owned and earnedmedia on sales [market share movement. Increasingly MMM is being usednot only for deriving the insights but as a "live" scenario planner answering"what if" questions to the marketing practitioner.
Auto regression was used as a technique in implementing MMM bymany practioners in the past and many continue to do so. Thisrelated the relationship of set of independent variables, themarketing inputs, on sales. But the underlying assumption Of thistechnique was the variables taken in to consideration should be"independent". But in reality, especially in marketing, one does notcome across "independent" variables. All variables are interlinked-TV not only could influence sales but also could influencedistribution. Similarly online media and TV could be related in aclose manner as TV exposure could lead to an online visit and thiscould then result in brand purchase. So one of the keydevelopments in this discipline in the recent past is treatment ofthese variables not as "independent" but more realistically asinterdependent variables. Techniques like Bayesian networkanalysis, Bayesian shrinkage, causal model using Bayesian, Vectorauto regression (VAR) are some of the advances from autoregression. These techniques provide us insights on the extent ofrelationship between the so called "independent" variables andthus by embracing the inter relationship, provide us a more preciseestimation of the influence of marketing variables on sales. It notonly provides us just the estimation, but also throws light on thepattern of direct and indirect influence of certain variables onsales/ market share.
Another major advancement that one witnesses in this discipline isadoption of non linear response functions. Response function isthe pattern of relationship of any input variable on sales/ marketshare. This throws light on the pattern of sales movement asone increase the input on the marketing variable that influencessales. Earlier given the complexity all the other variables apart fromTV were considered linear. Now non linearity can be applied to anyvariable and the complexity has not become manageable,therefore leading to better estimation.
But in reality most of the variables including the digital media, neednot display a linear relationship. The relationship could be nonlinear as shown in exhibit two and three
But in reality most of the variables including the digital media, neednot display a linear relationship. The relationship could be nonlinear as shown in exhibit two and three
Thus understanding the pattern of relationship and then mirror thispattern in modeling is very critical for an accurate MMM. Newtechniques like Bayesian network, causality models using Bayesianand Bayesian shrinkage are increasingly used to take care of thispattern.
MMM has evolved from being used as to derive media insight to arobust aid in creating a complete marketing strategy. The outputfrom MMM is increasingly being used to create various scenarios indeploying marketing inputs and monitor the business outcome.Since the capture of data has being automated and is notaccessible without a lag, the scenario planner from MMM is usedmore like "live" scenario! In addition, given the advancements inoptimization algorithms, MMM is also used increasingly to optimizethe marketing and media inputs to deliver sales growth andmarketing effectiveness.
More and more industries are embracing this practice in recenttimes, as marketing accountability is one of the coreresponsibilities of a marketing practitioner. MMM has moved a longway from being used by CPG category to even to most of servicecategories. Irrespective of categories, getting a causal insight,through quantitative techniques from the available data isbecoming a basic norm for marketers. As this discipline penetratesto many categories, more ideas will get injected in to MMMpractice, thanks to category and market dynamics. Thus one couldsee further advancement in this practice in future!