Here’s an interesting metric - 60% of US advertisers currently use Marketing Mix Modelling (MMM). And 58% of those not using MMM are considering doing so in the near future. And there’s a good reason for this. In today’s competitive market, every marketing dollar counts and maximizing ROI is critical. So what exactly is MMM?

Marketing Mix Modelling (MMM) is a time-tested analytical technique used by companies for measuring the impact of their marketing investments. First adopted by large multinational Consumer Packaged Goods (CPG) companies in the early 1990s, this comprehensive analytical approach takes a close look at the historical relationship between marketing spending and business performance.

It determines the effectiveness of different marketing elements like TV advertising, print advertising, digital marketing, pricing discounts, trade promotions, etc. in terms of their contribution to sales volumes, revenue, profitability or other relevant Key Performance Indicators (KPIs) and has become a crucial tool for guiding strategic and tactical marketing budget decisions.

This article takes an in-depth look at MMM – what exactly it is, why more and more companies are exploring it as a preferred measurement technique, the methodology and how it works, latest advancements and trends in MMM, and importantly how brands can benefit from leveraging this time-tested technique.

Marketing Mix Modelling (MMM) also referred to as Media Mix Modelling, helps brands determine the actual impact of each marketing input or activity on KPIs such as volumes, revenue or profits. With Marketing Mix Modelling, companies get data-driven insights that allow them to adjust their marketing mix to achieve their sales forecast targets. It also provides inputs for optimal budget allocations for each particular marketing campaign or channel, based on how different channels are contributing to sales.

MMM platforms helps companies to streamline all the factors of a marketing mix such as product, pricing and promotions, and also determines the effect of each of the factors on sales. An important feature of Marketing Mix Modelling is that it takes into account not just the impact of marketing and media drivers on sales but also non-media drivers such as competitor activity, seasonality, weather, and holidays as well as macro drivers such as GDP, unemployment, inflation, purchasing power etc. This makes it an extremely comprehensive technique to understand and measure how different business drivers impact sales and other KPIs.

Marketing Mix Modelling can help companies make better decisions by comparing the trade-offs between marketing mix elements. The ultimate goal is to estimate the contribution of each marketing element to the company’s overall performance. This technique is used to help identify which of these elements are most important for a company’s success.

While Marketing Mix Modelling has been used by CPG companies for many years, the technique of Marketing Attribution has gained popularity in recent times with the rise of digital marketing and digital-first businesses where most or all of the customer journey takes place online. MMM and attribution have co-existed, though the choice of which tool to use for marketing measurement was primarily a question of the industry you were in.

However, recent changes in privacy laws, corporate policies, and other factors are forcing companies to look towards adopting Market Mix Modelling, since it is privacy friendly and is not affected by changes such as GDPR and privacy regulations, phasing out of cookies on various browsers and Apple’s restrictions on IDFA on iOS devices. Using automated Marketing Mix Modelling is quite effective because it considers aggregate data rather than user-level data for modelling and analysis. In addition, Marketing Mix Modelling considers the impact of both internal and external factors in modelling, thus providing a more comprehensive look at the drivers that impact sales. This is especially beneficial for CMOs in companies, as Marketing Mix Modelling takes a holistic approach to marketing trends and provides a 360º view into the impact of various marketing and non-marketing drivers on sales.

The MMM SaaS tool by Analytic Edge enables effective ROI measurement without the limitations of traditional MMM such as being completely dependent on external consultants, being time and resource intensive, expensive, not easily scalable, and lack of transparency into the Modelling process (black box). Analytic Edge’s MMM software allows brands to run predictive marketing analytics in-house and on-demand, with the cost, scale, and speed advantages they need.

When we talk about the Marketing Mix Modelling methodology, the principle of multi-linear regression is the most important aspect. Sales or market share is usually considered the dependent variable whereas independent variables could be the factors like distribution, price, TV spends, digital spends, website visitors, outdoor campaign spends, newspaper, and magazine spends, below-the-line promotional spends, consumer promotions information, etc.

Marketing Mix Modelling helps to effectively quantify the impact of each of the marketing inputs from the betas that are generated from the regression analysis performed. The beta shows that one unit increase in the input value would increase the sales by beta units while keeping the other marketing inputs constant. The variables can have either a linear or non-linear relationship with the sales. Deep Dives Analysis and Budget Optimization are the other key things that MMM software assists marketers with.

A typical MMM process involves the following steps and modules:

• Data Input – Manual data loading and alteration via a User Interface. This supplements automated ETL for new and ad-hoc data.
• Data Review – Review and quality check of the data to identify errors and check outliers. This step ensures complete and correct modelling inputs.
• Modelling – Build, test and iterate to arrive at a final model that explains past business performance based on all available data.
• Reporting – This involves standardized reports such as Contributions, Due-Tos, ROIs, Lifts and Response Curves.
• Simulation – This step allows for what-if scenario testing and marketing plan optimization.
• Planning – The final step of MMM revolves around forecasting and variance analysis and tracking progress vis-à-vis the plan.

Here’s a look at the basic Marketing Mix Modelling components:

• Base Sales – The base sales of a product are often driven by economic factors like trends, pricing, seasonality, or qualitative factors such as brand awareness or brand loyalty.
• Incremental Sales – This is the additional sales that come about as a result of promotional activities or marketing campaigns across different mediums.

Implementing Marketing Mix Modelling SaaS in your organization can bring numerous benefits to your brand. Here are the steps to follow to implement an MMM program.

• Defining business priorities and scoping the study - The first and most important step to getting MMM right is defining your business priorities. Identify top business challenges and encapsulate them under two broad categories of profitability and growth. Translate challenges into specific questions that define the scope. Identify how to answer these questions through MMM.
• Assessing data readiness - Data readiness is about assessing how data ready you are as an organization to embark on the MMM journey. Do you have a data infrastructure in your organisation? If yes, is it efficiently managed? Are data feeds automated? What is the frequency of data collection for sales, app downloads, media spend and other important variables? Typically for MMM you need 3 broad categories of data – performance data, media data, and other drivers.
• Securing alignment - Align with marketing, commercial and data science teams to get buy-in and allocate responsibilities. Any successful program will typically have a senior sponsor who will engage with and commission the work.
• Using MMM recommendations for maximum impact - MMM not only provides media mix recommendations but also non-media e.g., how promotions and pricing impact sales, the role of affiliates, how much in-store distribution contributed to sales, the true impact of seasonality etc. Therefore, the findings of the MMM can be used to inform the business on both marketing and non- marketing decisions in order to take relevant action.

With the right DIY Marketing Mix Modelling , businesses can maximize their advertising mix and promotional tactics to generate higher sales or revenues. MMM findings and insights allow marketers to determine the optimal mix of product, distribution, pricing, and promotion to create the ideal marketing mix to boost sales.

Factors such as traditional channels, promotions, seasonality, or other variables are also factored into the MMM by marketers. Once the data is collected from various sources, advanced statistical analysis and Artificial Intelligence (AI) are applied to the data. The insights can help to determine the effectiveness of the current marketing campaigns.

While MMM offer numerous benefits to brands, traditional MMM solutions do have a few typical challenges and limitations.

• Time and resource intensive – MMM solutions have traditionally been time and resource intensive, making them prohibitively expensive to scale across the organization or update regularly.
• Reliance on external specialists – Companies are reliant on external vendors every time they need to run analytics and do not have option to do this in-house.
• Black box approach - Thirdly, MMM solutions are like ‘black boxes’ and third party vendors do not provide visibility into the models used and basis for insights. This results in low confidence and trust In MMM insights and recommendations.

Here are some Marketing Mix Modelling examples from the business world:

• A leading bank wanted to move away from print and increase digital spending. Before doing that, they used a customized Marketing Mix Modelling SaaS to understand the effectiveness of different campaigns in specific regions, across different service channels and age cohorts, and optimize budget allocations. They were also able to identify the best channels to target certain unique offerings as well.
• A large supermarket chain wanted to devise an integrated marketing strategy based on customer purchase behavior in stores. They wanted to change the merchandising based on the items that were purchased by the customers. They used automated Marketing Mix Modelling for better customer segmentation analysis.
• A multinational CPG company’s automated Marketing Mix Modelling offered them a continuous and ‘always-on’ MMM framework to aid frequent business planning decisions.
• A Spanish hotel chain with a global presence observed a significant 36% increase in Return on Advertising Spend (ROAS) by measuring the effectiveness of its different advertising channels and implementing changes based on the Marketing Mix Modelling insights.
• A global foods and nutrition company used Marketing Mix Modelling to determine the optimal allocation of their marketing budget across several brands in order to improve marketing Return on Investment (ROI). They also wanted to understand how marketing programs of one brand impacted the other brands

With Marketing Mix Modelling businesses can better understand how changes in their marketing efforts affect their sales and profits. Establishing business objectives is the first step to getting started with Marketing Mix Modelling optimization. Following this, the required data is gathered and analyzed to determine the optimal combination of marketing efforts to maximize the company’s profits.

Marketing Mix Modelling was traditionally seen as a technique for more traditional brands, with a higher proportion of offline spend. But even digital advertisers can significantly benefit from the wider view that MMM delivers. Modern MMM software offers marketers a holistic, telescope view, allowing them to optimize marketing allocations across various digital and offline channels and maximize marketing ROI.

Marketing Mix Modelling uses predictive Modelling to quantify the impact of various marketing activities on any Key Performance Indicator (KPI) such as sales, revenues, number of customers, number of installs, etc. With the help of such insights, your company can allocate marketing budgets optimally for maximum ROI.

Until recently, the choice of which solution to use for measurement, whether Marketing Mix Modelling or attribution was largely driven by the industry companies belong to. Digital-native advertisers typically opted for the speed and granularity of attribution, while more traditional or omnichannel advertisers opted for the accuracy and wider view that Marketing Mix Modelling platforms delivered.

With the improvements that next-gen Marketing Mix Modelling platforms now offer, digital advertisers can have the best of both worlds. They can combine and leverage the benefits of both attribution and Marketing Mix Modelling for a much more comprehensive measurement and insights through a future-proof measurement stack.

About Analytic Edge

Analytic Edge, a part of C5i, is a global analytics company that leverages technology and advanced analytics to help companies make data-based marketing decisions. The company’s flagship platform Analytic Edge Qube offers a suite of marketing analytics solutions with a Software as a Service (SaaS) model. The solutions include DemandDriversTM for always-on Marketing Mix Modelling (MMM), SynTestTM for AI powered Test and Learn, PriceSenseTM for pricing and promotion analytics, and PowerViewTM for analytics visualization. Analytic Edge works with clients across industry verticals such as e-commerce, mobile apps, gaming, consumer packaged goods, retail, automotive and many others.