Rashika Jaxon, MEA Marketing Partner at Meta.
Today's digital landscape gives marketers access to a seemingly limitless amount of data. But are they making the most of this data? This is not the case in many organizations. Over the past decade, direct marketers have used a variety of data sources, such as cookies and mobile device identifiers, to measure advertising results. Accessing this data is increasingly difficult, making measurement systems such as channel-level reporting and cross-channel attribution less effective.
True digital brands, accustomed to making rapid and frequent changes based on granular data feeds, need to rethink their measurement strategies. This category includes disruptive brand marketers: born-on-the-Internet startups that bring value to customers in innovative new ways. Now is the time for marketers, including digital natives, to build on existing marketing analytics. Looking ahead to 2023, Marketing Mix Modeling (MMM) marketing and business statistical analysis will evolve as a relatively scalable measurement solution for brands seeking actionable insights into the effects of cross-channel marketing.
As privacy laws change the way marketers collect and use data, this is how brands can use marketing mix modeling to uncover insights and measure performance.
Marketing simulation in a new era
Thanks to innovative machine learning algorithms, MMM allows you to publish useful information with the desired granularity and speed. Because it is not based on individual data, it proves to be the only truly holistic and reliable measurement system in the growing data ecosystem. It's great to see marketers moving towards sustainable MMM, developing skills and recruiting talent to address this challenge internally. A good indication of this change is the growth of the Robyn MMM open source community on Github and Facebook . More than 1,000 marketers, analysts and data enthusiasts are already connecting, discussing and learning as they continue their MMM journey.
In a recent study by Accenture, they conducted several experiments with a specially designed MMM to test its suitability for brand marketers who need a reliable and affordable system for detailed multimedia optimization.
Their results show that when used with advanced machine learning techniques and innovations, MMM offers the following benefits:
1) Robust MMM is accessible to marketers of all sizes and categories: The success of MMM is largely measured by the model's ability to predict dependent outcomes. Accenture's experiment, which uses 1,200 characteristic data from 5 different sources commonly collected by traditional brand marketers and disruptors, demonstrates a high level of predictive accuracy and meets the industry standard of 90% R-squared and 5 values. % or less than the mean absolute percentage error (MAPE).
2) MMMs can provide detailed and useful results: Thanks to advances in machine learning, MMMs can now use techniques such as gradient descent algorithms to analyze data and extract useful information based on these variables. In Accenture's research, the model breaks down two years of data down to the day-of-the-week level, a size needed by marketers who change budgets daily (as shown in Figure 1). This is an example of how MMM can provide detailed and actionable information that marketers can use to optimize marketing performance.
3) MMM demonstrates cross-channel synergy without user tracking: Breakthrough brands are often interested in understanding their customers' conversion paths to optimize their cross-channel marketing efforts. Marketing mix models when integrated with advanced machine learning techniques can provide similar insights into cross-channel synergies. Accenture's experience provided clear insight into cross-channel impact. The results are summarized below in the "grid" of enabling factors in Figure 2.
Methods that consistently meet the most important measurement requirements
MMM integrates and evaluates all online and offline marketing activities, creates an overview of their relationships and extends to tracking factors such as promotion, season or competition. Today, MMM requires less resources and budget to implement and uses aggregated data to provide fast and detailed analysis across all channels, making it suitable for brand marketers of all sizes, including those focused on direct mail. Breakthrough brand marketers who often face business changes and tend to use a variety of non-paid marketing tactics also benefit from the comprehensive insight that MMM provides.
Here are four best practices for effective, accurate, and long-lasting results:
1) Agree on the key objectives before the simulation: Given the wide range of questions that MMM can answer, it is important to create an MMM training plan and focus on answering one question at a time. Alignment of key objectives is an important first step in the MMM process. All subsequent stages of MMM construction will benefit from a clear understanding of MMM's core objectives.
2) Make sure the data is current and complete: Create separate variable requirements for each strategy the marketer wants to measure ROI. In addition to preparing for media-related variables, it's also important to include a comprehensive list of non-media variables that can also affect a brand's business results. These variables vary from brand to brand, but some common ones are economic factors, seasons, competition, etc.
3) Choose an MMM option that answers your question - the most common questions are usually already answered in the various MMM options on the market, whether it's an open source solution like Robyn, Solutions Partner, or MMM Self-Service. A SaaS solution. When evaluating MMM options, ensure that their skills can effectively answer the questions identified in Step 1 so that brands can gain useful insights from MMM.
4) Regularly update and calibrate MMM to reflect business changes: Investing in a data infrastructure that automatically feeds new data into the MMM model helps marketers and modeling professionals update their MMM models more effectively in the long term. It is important to build a framework for calibrating and selecting the most powerful MMM model. Further research is the industry's gold standard for measuring ground truth. Marketers should conduct incremental research into their marketing channels along with launching MMM to improve model accuracy and build confidence in MMM adoption.
Reliable measurement solutions require a solid foundation
MMM and its progress will remain. It's time for breakthrough brands to stop waiting and start their MMM measurement journey. The best way for marketers to understand advertising effectiveness is to introduce quick models along with causality tests. Implementing marketing mix modeling now, before the data floodgates close one by one, will help prepare brands for future changes in online privacy. Time and resources are now invested in its construction; will pay for whoever does it.