2. Why should marketers use MMM?

Marketers need to adopt marketing mix modeling to gain a holistic view of their diverse advertising stack and understand the effectiveness of each channel. MMM provides a comprehensive analysis of the entire marketing funnel, from top-of-funnel brand awareness campaigns to bottom-of-funnel conversion tactics. Gaming companies—subject to frequent shifts in consumer preferences, technology, and competitive actions—can leverage MMM across the entire marketing mix to quantify the impact of digital and traditional channels, optimize marketing spend, and make data-driven decisions.

Next-generation MMM solutions have significantly reduced the technical complexity and resource requirements for implementation. Gaming advertisers can now easily integrate MMM into their workflows, with daily or weekly model updates providing highly actionable insights. MMM offers several advantages over traditional last-touch attribution, including the ability to reveal channel saturation and make more informed budget allocation decisions. Gaming advertisers often run campaigns across multiple channels simultaneously, and MMM can reveal synergies among these channels, showing how they work together to boost overall performance.

As the gaming industry faces increasing privacy restrictions and data availability challenges, MMM has become a future-proof measurement strategy bridging the gaps left because of evolving privacy policies. By adopting a data-driven, experimental approach to MMM, gaming companies can unlock powerful benefits, such as understanding the true drivers of their success and maintaining a competitive edge in such a rapidly changing market while adapting to a privacy-forward world.

5. What performance improvements can one expect when implementing MMM?

Implementing marketing mix modeling can lead to significant performance efficiency boosts in marketing strategy and execution. A primary benefit is enhanced return on ad spend (ROAS), as MMM helps pinpoint the marketing channels yielding the highest returns, allowing for strategic reallocation of budget based on analysis of cost curves. MMM can drive down cost per acquisition (CPA) by identifying and focusing on cost-efficient channels that engage the most users for the least spend.

MMM optimizes budget allocation by taking into account the interplay among channels and recognizing saturation points to prevent diminishing returns. This helps identify inefficiencies and opportunities within the marketing strategy, such as areas of overspending or potential channels that could drive higher returns with additional investment.

MMM provides a clearer understanding of the impact of organic efforts versus paid marketing, essential for accurately attributing growth and understanding the true value of paid campaigns. The tool’s forecasting and scenario planning capabilities allow marketers to anticipate future outcomes and adapt strategies accordingly. As the gaming market and consumer behaviors evolve, MMM’s data-driven adaptability ensures that insights remain current, guiding marketers through market shifts. Finally, MMM fosters alignment across different marketing teams within an organization, ensuring that all efforts are coordinated toward common goals and that each team understands the impact of their work on overall marketing success.

6. How hard is it to set up MMM?

While setting up marketing mix modeling requires a thoughtful approach and access to quality data, it is no longer the complex, resource-heavy endeavor it once was. Setup is more streamlined and user-friendly than ever, thanks to advancements in technology and the emergence of next-generation MMM solutions such as AIM by Kochava.

An MMM onboarding process begins by feeding the model comprehensive historical data, such as daily aggregated marketing spend, sales or conversion data, and other relevant metrics. This process has become more straightforward and doesn’t require the technical integrations of the past, such as SDKs or tags. Beyond collecting and ensuring the quality of historical data, marketers consult with an MMM provider to understand business nuances, develop a tailored statistical model, and integrate various data sources through user-friendly APIs.

Once the model is developed, it is calibrated and validated to ensure accurate market reflection and reliable predictive capabilities. The MMM partner and advertiser work together to interpret outputs and apply insights to marketing strategies. The model benefits from ongoing support and iterative improvements as new data comes in and market conditions evolve, ensuring that takeaways remain relevant and actionable.

7. What are the inputs needed to run MMM?

Webinar panelists emphasized the importance of data quality and comprehensiveness, which fortunately is a hallmark of the gaming industry. Historical data, ideally spanning one to three years or more, is necessary to capture long-term trends, seasonality, and business cycles, all of which impact the model’s accuracy.

An accurate and actionable model starts with aggregated market-level data encompassing sales and/or conversion figures compiled daily or weekly and provided to the model on an ongoing basis. This data serves as the dependent variable that enables the model to assess the influence of various factors on these outcomes.

Next, ad spend data comprises expenditures broken down by channel and aggregated to the same frequency as sales data. External factors affecting sales but beyond the scope of marketing efforts, including economic indicators, weather conditions, holidays, and significant events, should also be incorporated. For gaming companies, conversion funnel data is key, as it maps out the customer journey from app installs to in-app purchases, helping to attribute value to different marketing activities.

The panelists noted that modern MMM solutions such as AIM by Kochava have made the process of gathering and integrating these inputs more streamlined. The use of APIs and automated data ingestion processes allows for efficient, painless data collection.

8. What outputs does one get from MMM?

The multifaceted outputs from marketing mix modeling are designed to provide a full understanding of marketing performance across various channels. When you engage with an MMM solution, you receive actionable insights that are comprehensive, actionable, and effective for optimizing your marketing mix in a privacy-first, data-driven environment:

• MMM offers detailed sales attribution, revealing the contribution of each marketing channel, both online and offline. This granular understanding of the marketing mix enables gaming companies to make informed decisions about where to allocate budgets for maximum impact.
• MMM generates cost curves, identifying the point at which additional spending in a particular channel yields diminishing returns. This insight is crucial for optimizing marketing spend and avoiding overinvestment in channels that have reached their peak effectiveness.
• MMM analyzes channel effectiveness and saturation to provide data-driven budget recommendations for desired marketing goals—maximizing overall sales, improving cost-per-acquisition, or optimizing for other key performance indicators (KPIs).
• MMM separates the incremental impact of marketing efforts from baseline sales, providing invaluable insights into the true value added by your marketing investments. It reveals synergies among various marketing channels, helping businesses understand how various initiatives work together to drive sales.
• MMM offers powerful forecasting and scenario planning capabilities, allowing gaming companies to predict future marketing performance and simulate the potential impact of various budget allocation strategies.

Ideally, the model’s output should be tailored to align with existing workflows and decision-making processes within an organization. This enables marketing teams to integrate MMM insights seamlessly into their strategies and make informed, data-driven decisions.

9. Who are the clients of an MMM model?

Clients of MMM models are typically marketers who need to understand the effectiveness of their marketing spend and seek insights into the performance of their marketing investments. These clients can include:

• Performance marketers: Marketers focused on optimizing ad spend to achieve the best possible return on investment (ROI), needing to understand the impact of different marketing channels on conversions and sales.
• Brand marketers: Marketers concerned with building brand equity and awareness over time, needing to measure long-term effects of branding campaigns alongside short-term performance metrics.
• Media planners and buyers: Professionals responsible for allocating budgets across various media channels, needing to justify media mix decisions with data-driven insights.
• Marketing executives: CMOs and other marketing leaders who require a high-level view of marketing effectiveness to inform strategy and report to other company stakeholders.
• Data analysts and scientists: Individuals who delve into the data to extract meaningful insights and trends, requiring a robust analytical tool to aid in their analyses.

10. What does an MMM UI offer?

Modern MMM tools present data in a format that is easy to interpret and act upon. This means that the insights are not only interesting but also actionable, allowing marketing teams to implement recommendations with confidence. The user interface (UI) offered by contemporary MMM solutions is designed to be intuitive and user-friendly, enabling users to navigate and interpret the data easily. The UI may include:

• Dashboard view: A centralized dashboard that provides a snapshot of key metrics and performance indicators, allowing clients to quickly assess the overall effectiveness of their marketing efforts.
• Data visualization: Graphs, charts, and other visual tools that make complex data more digestible and clearly depict important trends and patterns.
• Scenario planning tools: Features that enable clients to simulate different marketing strategies and budget allocations to predict potential outcomes and optimize spend.
• Customizable reports: Readily generated reports that focus on specific KPIs, channels, or time periods, tailored to the needs of the user.
• Actionable insights: Data presented in a way that advises on steps for optimization, such as recommendations for budget reallocation.
• Integration with other sources: The ability to integrate MMM insights with other data sources, such as attribution models, to provide a more complete view of marketing performance.

The UI is crafted to empower marketers with the ability to make informed, strategic decisions without requiring deep technical expertise in data science or statistical modeling. It bridges the gap between complex modeling techniques and practical marketing applications, ensuring that insights generated by the model are accessible and actionable for clients.

11. How would you suggest advertisers integrate MMM results together with attribution results to make strategic decisions?

Advertisers can effectively leverage the complementary strengths of MMM and attribution data to make informed, data-driven budget decisions aligned with both short-term and long-term marketing objectives. This integrated approach—inherent to a large degree in next-generation MMM tools—enables advertisers to navigate the complexities of today’s marketing landscape and effectively allocate their budgets for maximum impact.

By integrating the granular, real-time insights into campaign impact from attribution data with the strategic, long-term insights about marketing effectiveness over time from MMM, advertisers can understand a complete picture of marketing performance. Advertisers can use MMM results to identify saturation points, measure incrementality, and understand channel synergies. They can complement these insights with attribution data on short-term conversions and immediate ROI. This integrated approach helps ensure that budget allocations across their marketing mix are aligned with both immediate and long-term business goals.

Tools that integrate these results assist advertisers in creating a comprehensive, data-driven strategy that accounts for the complexities of the modern marketing landscape. This holistic approach empowers them to make better-informed budget decisions and drive improved return on their marketing investments.

12. Is AIM a dedicated measurement platform partner for game UA?

Yes, AIM (Always-On Incremental Measurement) by Kochava is a platform designed specifically to address the kinds of unique challenges faced by user acquisition (UA) marketers in the gaming industry.

AIM leverages advanced machine learning to provide real-time insights, crucial in a dynamic, fast-paced gaming environment where trends and user behaviors shift rapidly. By continuously learning from daily UA results, AIM adapts to new market information, ensuring that insights are always current and accurate. This empowers gaming marketers with immediate, powerful data for next-level growth strategies.

A key feature of AIM is its seamless integration without disruption into existing UA workflows. This is particularly important for gaming companies with complex systems and processes already in place. The platform also respects data privacy by operating without the need for SDKs, tags, or sensitive data, making it a future-proof solution in the evolving privacy landscape.

AIM addresses the challenges posed by increasing limitations on touch attribution and the struggle of traditional MMM to keep pace with the dynamic UA landscape. It offers features such as optimized budgeting, accurate forecasting, and powerful scenario planning to bolster buying decisions with data-driven insights.

AIM by Kochava is a dedicated measurement platform that brings a modern, privacy-first, and real-time approach to the challenges of today’s gaming market, empowering UA marketers with the tools needed to optimize their strategies and drive growth.

13. How should a gaming company test an MMM model?

As the expert speakers at the webinar emphasized, the true test of a MMM model comes from its application in the real world—that is where the rubber meets the road. Testing an MMM model involves a multifaceted approach to validate its accuracy and actionability. The testing process can begin with applying the model’s initial incremental recommendations to real-world marketing strategy, then observing the impact on key performance indicators (KPIs).

Marketers can implement the changes suggested by the MMM model, such as adjusting marketing spend across various channels, monitoring the resulting changes in metrics like sales, conversions, return on ad spend (ROAS), and cost per acquisition (CPA). By directly observing whether these recommendations lead to expected improvements, marketers can confidently assess the model’s ability to provide accurate guidance that improves results.

In addition to real-scenario testing, conducting backtests—seeing how a model would have performed after the fact—is key validation. By applying the MMM model to historical data, marketers can compare its forecasts to the actual results for that period. This retrospective analysis helps identify discrepancies and pinpoint areas where further refinements might improve accuracy.

A testing process is iterative. Based on the outcomes of implementing recommendations and backtesting, marketers can make adjustments to the model, such as refining the input data, tweaking the parameters, or incorporating additional variables that influence marketing effectiveness. Continual improvement ensures that the model remains accurate and responsive to changing market conditions. Next-generation MMM platforms, such as AIM by Kochava, are designed to update in real-time or near-real-time, allowing for continuous learning from new market data. This ensures that the insights remain current and actionable, reflecting the latest campaign performances and market trends.

Note that understanding the model’s inherent error rate and confidence intervals is crucial for setting realistic expectations in interpreting the results. A robust MMM tool provides this information, allowing marketers to make decisions with a clear understanding of any associated uncertainty.

14. Any suggestions for gaming advertisers who are just starting to explore MMM?

First and most important: Embrace a learning mindset. Approach MMM with curiosity and a willingness to challenge existing assumptions. MMM provides valuable insights that may differ from your current understanding of marketing effectiveness, so be prepared to reassess your premises based on your model’s data-driven findings.

Rather than overhauling your entire measurement strategy at once, start with a pilot program. Test MMM on a specific campaign or subset of your marketing channels to learn how to interpret the model’s outputs and gain confidence in its predictive capabilities. This will enable you to integrate MMM painlessly with your existing attribution and analytics tools and assess a more comprehensive view of your marketing performance.

Ensure that you have processes in place to collect and aggregate data accurately, as MMM relies on accurate, granular data from your media spend and performance as well as external factors that could influence marketing results. If you lack in-house expertise, consider partnering with an experienced MMM provider or consultant who will guide you through the complexities of setting up and interpreting the model. Keep in mind that next-generation MMM tools can be up and running with relative ease.

Check out this post on why many companies are choosing SaaS MMM tools over in-house solutions.

MMM is an iterative process. Experiment and continuously refine and optimize the model, making incremental adjustments to your marketing strategy and testing the impact of these changes. Over time, you can fine-tune your marketing mix to achieve better efficiency and ROI.

Finally, lean into MMM for strategic planning beyond tactical campaign optimizations. The model can help you understand the long-term effects of your marketing investments and inform budget allocation decisions for future periods, empowering you to make more informed, data-driven choices for your gaming business.

By following these suggestions, gaming advertisers new to MMM can set themselves up for success, leveraging the model to gain deeper insights, optimize their marketing spend, and ultimately drive better business outcomes. Remember that the MMM journey is one of continuous learning and improvement, and the insights it provides can become a competitive advantage in the evolving gaming market.

15. How often should gaming advertisers refresh their MMM model? Daily, weekly, monthly?

The frequency with which gaming advertisers should refresh their MMM model is contingent on various factors, including the model’s design and sophistication, the volatility of the market, and the company’s specific objectives and capabilities. Overall, the refresh rate of an MMM model for gaming advertisers should be tailored to the unique needs of the business, the agility of the marketing team, and the dynamics of the gaming market. While some advertisers may benefit from daily updates, others may find weekly or monthly recommendations more practical. The key is to ensure that the output is actionable, timely, and aligned with the company’s strategic goals and operational capabilities.

The technology behind modern MMM solutions has advanced significantly, allowing for more frequent updates, with some tools capable of providing daily or near real-time recommendations. This is particularly beneficial for dynamic industries like gaming, where market conditions and user behaviors can change rapidly. However, the optimal cadence for model refreshes should align with the gaming company’s decision-making cycles. If the advertiser makes budget allocation decisions on a weekly basis, then weekly MMM recommendations would be most beneficial. For strategic planning and longer-term decisions, monthly or even quarterly updates might be more appropriate.

The gaming industry is subject to swift changes due to new game releases, competitor actions, and shifts in consumer preferences. An MMM model that refreshes its recommendations frequently can help advertisers stay agile and respond to shifts promptly. The availability of quality data is also a key factor, as the model can be updated only as frequently as the flow of data allows.

16. What’s an MMM model’s error rate?

As the webinar panelists stated upfront, every model, including marketing mix modeling, has an inherent error rate. However, a well-designed MMM model will have a relatively low error rate and provide confidence intervals to help understand the precision of its predictions.

The error rate is a measure of the model’s accuracy in forecasting outcomes based on input the data it receives. This is often expressed through confidence intervals that give a range within which the true value of the model’s estimates is likely to fall. Narrower confidence intervals indicate higher precision, while wider intervals suggest greater uncertainty.

A robust MMM model aims to minimize error by using advanced statistical methods and high-quality data, accounting for any and all factors that could influence the outcome, such as seasonality, market trends, and external events. The model is calibrated and validated using historical data to ensure that its predictions remain as accurate as possible.

Webinar panelists emphasized the importance of transparency in MMM results, where error rates and confidence intervals—as well as insights into how these figures are calculated—are clearly revealed. This helps marketers understand the limitations of the model and make informed decisions based on its outputs. Note that the error rate of an MMM model can be put into context by comparing it to the error rates of other marketing measurement methods such as last-click attribution, which may not capture the full complexity of the consumer journey.

Advertisers are encouraged to use MMM in conjunction with other measurement tools to cross-validate findings and gain a more comprehensive understanding of marketing performance. This will help them leverage the strengths of each method and mitigate the impact of any individual model’s error rate. The goal is to reduce error to a level where the insights provided by the MMM model are actionable and drive marketing efficiency and positive results.

17. Why isn’t MMM more common? Why is it hard to find a tool that provides usable results?

Marketing mix modeling has indeed been used by traditional advertisers for a good long while. Despite this, adopting MMM has traditionally come with challenges due to its inherent complexity, significant resource requirements, and, in recent decades, the notion that it is an outdated approach compared to real-time attribution models. However, as webinar panelists emphasized, MMM methodology has evolved significantly in the last few years, with next-generation providers developing more advanced, automated models better suited to dynamic industries like gaming. In fact, it is highly likely that personnel at your company are already deriving valuable insights from MMM, even if they are not explicitly aware of it.

Historically, implementing MMM has been seen as a complex and resource-intensive process. It required significant data collection, processing, and a team of skilled data scientists to build, interpret, and maintain the models. This level of complexity and resource requirement made MMM less accessible, especially for smaller companies or those without the necessary in-house expertise. Too, perception may linger that MMM is a one-off model run annually or semiannually that provides static, PDF-type outputs not actionable in real-time. This has made MMM appear outdated and less relevant, especially when compared to more immediate and responsive attribution models crucial for industries like gaming, where the market is dynamic and decisions need to be made quickly.

All this has changed. Cutting-edge MMM providers such as AIM by Kochava have developed more advanced models adapted for dynamic industries like gaming. These next-generation MMM solutions are automated, frequently updated, and provide user-friendly, actionable insights, making them highly relevant and useful for today’s market.

With the evolution of MMM technology amid the evolving privacy landscape, there is a marked shift toward adopting more sophisticated, real-time, and privacy-compliant MMM solutions that provide a holistic view of marketing effectiveness and drive optimal, future-forward decision making.