Topics In Demand
Notification
New

No notification found.

Marketing Mix Modeling: The Cornerstone for CPG Success
Marketing Mix Modeling: The Cornerstone for CPG Success

October 16, 2023

57

0

Marketing Mix Modeling: The Cornerstone for CPG Success

 

As digitalization is picking up pace, companies are emphasizing on trying, validating, and adapting to processes and technologies for data-driven business decisions. Within businesses, be it any department from sales and marketing to IT and HR operations, the focus on data has evolved in recent years. Since COVID-19, the demand for data-driven insights for optimizing the performance of decision-making processes and the utilization of budgets across teams has been a key focus and challenge.

This has also impacted the traditional allocation of budgets for the marketing domain. Strategies for crafting the optimal marketing budget allocation across the mix of marketing activities have pushed marketers to adopt the usage of Marketing Mix Models (which evaluate spending at three pillars: Pricing, Promotion, and Media decisions). These models help identify which marketing channels will be effective and efficient, may drive the highest returns, and optimize spend allocations to maximize business outcomes.

The growing need for digitalization, the way consumers are consuming digital content, and how it is influencing their buying decisions (influencing shopper decisions); have pushed companies to look deeper into the 3rd pillar of their marketing mix – The Media.

The Rise of Media Mix Modeling

A bigger debate has emerged on the usage and consumption of Media Mix Models, which are essentially a subset of traditional marketing mix models. Brand managers and category managers are struggling to optimize their budgets, to have the right allocation for media activities. Allocating the media budget based on data-backed insights can help them both in the short term by driving sales, and in the long term by strengthening brand equity, brand awareness, and brand loyalty.

More often than not, marketers/brand managers/category managers have to deal with scenarios where they have multiple methodologies and multiple vendors running these programs (market mix, media mix, 360 mix models). And they struggle to choose the right model, right methodology, etc. The question that emerges here, is if all of these methods are conducted for the same reason, why are the results different, which one is to be believed, and how do we compile/compare the results of one with the other?

Well, I am going to present my view on how we can answer a few of these concerns with a certain level of confidence, to move towards a convergence of the results irrespective of the models or vendors involved. Especially in the context of running media mix models, and comparing the results to the marketing mix outputs.

Marketing Mix Vs Media Mix

Marketing Mix
We model all the variables which impact sales at the most granular level. Starting with UPC/SKU level promotions and pricing changes which are run weekly at the store level, to the media activities which are done at a category or manufacturer or brand or sub-brand, or SKU level.

Media Mix Models
Usually in media mix models, the emphasis is on modeling the media variables. And more often than not, the impact of the price and promotional variables are neither measured nor considered. Even when they are considered, it is only at a level that may not capture all the attributions as they should have; as these individual promotions on individual SKUs behave very differently when they are on promotion. The behavior is dependent on multiple other factors such as category, brand, region, and the store in which the product is on promotion, etc.

Hence, what you have is an over-estimated media variable in terms of attribution which is not desirable to any marketer, or category/brand manager. Thus, we face the issue of results being different despite using similar models and variables.

Now, the media mix model has multilevel reporting (depending on the business question which has to be answered):

Total TV
  • By timing
  • Spot length
  • Campaigns
  • Tactics
  • Networks
  • Type etc.
Total Digital
  • Online
    • Social/Influencer
    • Audio (Podcasts/Radio)
    • Video
      • Cinema
      • Hulu/YouTube (ad-based OTT) etc.
  • Display
Total Non- Digital (print)
  • Print – Newspaper/Magazine/Flyers
  • Billboards
  • OOH
Example of bringing the consumer lens (by defining and adding variables like below in models)
Customer segment (Using POS estimates OR Analyzing Shopper cart/basket)
  • Heavy chocolate buyers’
  • Heavy Holiday buyers’
  • Regular consumers’
  • Shifters etc.
Result: Media deep-dive to understand attribution, and optimize the budget allocation of media activities based on Category > Brand > Individual media type/vehicle/channel etc.

 

So, what is important to understand?

Even though you are just running a media mix model you will have to understand the impact of the promotion and pricing variables at the store level to ensure that the promotion/pricing impacts are not inflating (overestimating) the media contributions to the media models.

Since you are not running a complete marketing mix model, you will choose not to investigate and report promotion and pricing impacts at an individual SKU or promotion type level; but give a total contribution impact of promotional activities due to price changes over time.

Now, why is it important to do so?

Often you may have struggled to consolidate the findings from two models (run by two different vendors or maybe the same vendor but using a different definition of modeling).  If you do not model the promotions, your media contributions will be overestimated in the weeks when promotions are happening, and will disrupt the total media contribution or ROI, depending on the variability in base promotion will impact the media contributions (i.e., base/dips & falls).

If we take the above approach, we will be able to not only able to integrate or disintegrate results from various models, but still be able to compare the models which use multiple methodologies or are built by multiple vendors and will be able to generate more or less similar results in any independent effort for model/vendor selection.

   

    Author

Imran Saeed

Imran Saeed

That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.


C5i is a pure-play AI & Analytics provider that combines the power of human perspective with AI technology to deliver trustworthy intelligence. The company drives value through a comprehensive solution set, integrating multifunctional teams that have technical and business domain expertise with a robust suite of products, solutions, and accelerators tailored for various horizontal and industry-specific use cases. At the core, C5i’s focus is to deliver business impact at speed and scale by driving adoption of AI-assisted decision-making. C5i caters to some of the world’s largest enterprises, including many Fortune 500 companies. The company’s clients span Technology, Media, and Telecom (TMT), Pharma & Lifesciences, CPG, Retail, Banking, and other sectors. C5i has been recognized by leading industry analysts like Gartner and Forrester for its Analytics and AI capabilities and proprietary AI-based platforms.

© Copyright nasscom. All Rights Reserved.