Case Overview

Marketing Mix Modeling (MMM) is the use of advanced analytics to estimate the past impact and predict the future impact of marketing tactics on sales. Any MMM study needs to have defined goals, and businesses should establish the goals and get answers to all of them. Businesses should know if their vendor can answer all the questions they are looking at.

Here, we provide a summary of all the questions that can be addressed through a Market Mix Modeling (MMM) study.

1. Base Sales & Short Term Sales: Base Sales is the revenue a business would have generated without any marketing. And the Total Sales minus the Base Sales gives us the Sales driven by Marketing (i.e. Short Term Sales)

2. Adstock: Adstock or carry over effect measure the prolonged effect of advertisement on consumer’s purchase behavior. Adstocks from previous advertisements reinforce brand awareness, which eventually decay to negligible levels in the absence of further exposure. Brands can make use of Adstock to estimate the impact of their marketing through each channel and plan their channel activities accordingly

3. Channel Sales Contribution: MMM provides us the model co-efficients, which when multiplied (as per the model equation) with the marketing activity provides contribution at weekly or monthly levels. This provides businesses with the sales contribution generated through each marketing channel over the study period (e.g. Sales generated through TV for Q2)

4. Channel Deep Dive: Channels like TV or Print have many attributes associated with the ad spend. Taking TV as an example, the genre of the program/channel – Sports, News, Entertainment etc., Creative type – Creative 1,2 etc. are a couple of such attributes. And not all of these ads will have the same impact on short term sales. A deep dive analysis of channels having significant spend can help businesses with the media buying

5. Campaign Deep dive: Brands have a well defined campaign calendar and the media spend is often focused on specific campaigns. It is important for brands to have a campaign-channel mapping to be able to measure the effectiveness of each campaign and know which campaigns work better

6. Marginal RoI: Average Marketing Return on Investment provide the $ earned for each dollar spent. However, it is more important for businesses to know the impact of the next 1$ spent on marketing. Marginal RoI provides businesses with the capability to perform what-if analyses and plan future media buying

7. Response Curves: Response curves predict the expected incremental sales with ad spend through a given channel. And when multiple channel curves are combined together, it allows businesses to compare the marginal return on investment (MRoI) and make the buying decision. It would be worth noting that the response curves is a predictive element in the MMM study

8. What-if Analyses: Once a business has generated the response curves, it is straightforward to build a what-if analyses tool. What-if analyses tool should not be confused with optimizer, as it allows testing different channel spend scenarios retrospectively e.g. What would my short term sales have been if I increased spend on Direct Mail by 20% and reduced TV spends by 10%

9. Optimization: Or Market Mix Optimization or MMO in short. While knowing effectiveness of past spends and sales generated from them is insightful, a market mix study can only be complete if the outputs are used to come up with:

a. Optimized spend distribution if the Sales is defined

b. Optimized spend mix to reach a KPI (e.g. Ad spend mix to generate $20Mn of Short Term sales

10. Direct & Indirect Impact of Marketing: Gone are the days of disconnected marketing activities. It has been accepted now that consumers are used to having omni-channel interactions with brands. Traditional MMMs (e.g. Regression based) are not equipped to measure the interaction effects (one can introduce interaction terms, which is highly inefficient). It is therefore important for brands to commission an MMM which has the capability to test out all possible interactions and estimate the direct and indirect impact on short term sales

11. Long Term Impact: Marketing activities are not only focused on generating short term sales. The goal is to create brand equity that stays with the consumers. And this brand equity then drives demand for the product. MMMs can be used to estimate the sales being generated through brand attributes like differentiation, innovation, stylishness and from the brand equity as a whole

Building a comprehensive marketing mix model holds the key to an effective marketing and growth for a brand. Brands need to know the complete span of a market mix modeling (MMM) study and plan their MMM study to address all their questions. Only through this will brands be able to measure the effectiveness of numerous marketing variables working together towards the common goal – maximizing revenue

Xtage Labs is an advanced analytics and machine learning based decision insights company. We work with businesses to derive insights from data, and improve the decision making processes.

Get in touch with us to find out what we can do for you.