Analytics-Based Marketing Mix Model for Improved ROI
The availability of multiple platforms for marketing means that businesses have a plethora of choices when it comes to marketing. But how do they determine which of the marketing techniques are effective? For instance, if a business can invest $100 in marketing, how does it apportion towards advertisements and consumer promotions? This, ideally, should be determined by the returns obtained from each of these areas.
Analysing marketing solutions, simulating the returns on advertising, and optimizing these techniques is referred to as marketing mix modelling (MMM). Marketers can make informed decisions in terms of the spending on marketing inputs based on how much they contribute to actual sales. MMM analyses the past impact of marketing techniques on sales, and predicts the future impact on sales, thus influencing marketing decisions, and leading to an improved ROI.
Why use marketing mix modelling (MMM)
Today, with so much media available for advertising and promotions, it becomes vital that companies optimise their marketing spending to get maximum ROI to stay competitive. Marketing mix modelling has been around for a long time and was traditionally applied to the consumer-packaged-goods (CPG) or Fast-Moving Consumer Goods (FMCG) industry. With the explosion of data and tech infrastructure, it has come a long way from the initial 4 Ps of marketing - Price, Promotion, Product and Place, and now employs statistical and analytical aggregate-level modelling, deriving data from various sources including digital platforms, point of sale and company internal data.
With Google phasing out third-party cookies by the end of 2023, marketers are staring at a cookie-less world. Apple introduced App Tracking Transparency (ATT) from iOS 14.5 onwards, a consent protocol, limiting how advertisers use individual data. Consequently, a sustained disruption is expected in digital advertising, and marketers need to re-strategize ad measurement techniques to determine the best platforms for promotions and advertising.
Also, the New Normal has caused a shift in digital behaviour, and some proactive action is needed, with the ability to quickly realign based on changing circumstances. MMM helps with planning and taking pre-emptive actions. With the vastness of marketing, businesses need proven techniques to help with determining the best possible avenues for marketing, and MMM is important for brand success. Digital technologies in marketing will also make organizations more sentient – capable of sensing and feeling what the next customer is looking for.*
This particularly applies to the FMCG industry, one of the largest industries. Products are characterized by short-shelf life mostly meant for personal use. There is intense competition amongst retailers, and profit margins are low, which are offset by selling large volumes of products. In such a scenario, Marketing mix modelling can help establish product loyalty amongst consumers, and possibly improve profitability by increasing product prices.
Marketing mix models are based on statistical analysis
MMM uses multi-linear regression principles. The equation is formed by using dependent variables and predictors. Regression is based on how the independent variables impact the dependent variable. For instance, price, distribution, outdoor campaign spending, print media spending, etc., can be independent variables, and sales or market share can be the dependent variable. Also, these variables can have a linear or non-linear relationship with sales. Digital marketing inputs such as website visitors, social media spends etc. also become inputs to the marketing model.
Businesses can use MMM to evaluate the marketing channels used and determine which of them are the most effective and can provide maximum ROI.
Building the model
The marketing mix model helps analyse past performance and also determine the optimal mix of elements. The model also should support organisational and marketing objectives, performance measures and metrics.
While the marketing mix model is very useful, building the model requires some groundwork. Once it is built, marketers can interpret the model. Follow these steps to build an optimal model:
- Define goals and objectives by organisational and marketing objectives. Decide on key performance indicators (KPIs) and measurement metrics
- Organize resources - marketing mix models require experts and trained analysts/data scientists
- Decide on the data that needs to be collected for the model to build the prototype of the model - customer demographics, industry data, competitor data, marketing spends, sales data, distribution data etc.
- Identify data sources and merge them to prep for analysis
- Build the prototype, test predictions on a sample test, and rework the model
- Prep and collect the data
- Model the influence of individual factors and simulate the impact of different marketing activities
- Develop and deploy the optimised model
- Use the results by arriving at recommendations and fine-tune the model
- Keep revising the model every quarter and rebuild the model annually.
Choosing the right platform allows brands to run their own in-house MMM, where the models have integrated automatically with systems. Developing an optimised marketing mix model requires advanced analytical and modelling skills and can be done by trained data scientists and analysts. Using the right platform and expertise to build the model, collecting and aggregating the data is important for an effective marketing mix.
*For organisations on the digital transformation journey, agility is key in responding to a rapidly changing technology and business landscape. Now more than ever, it is crucial to deliver and exceed organisational expectations with a robust digital mindset backed by innovation. Enabling businesses to sense, learn, respond, and evolve like a living organism, will be imperative for business excellence going forward. A comprehensive, yet modular suite of services is doing exactly that. Equipping organisations with intuitive decision-making automatically at scale, actionable insights based on real-time solutions, anytime/anywhere experience, and in-depth data visibility across functions leading to hyper-productivity, Live Enterprise is building connected organisations that are innovating collaboratively for the future.