Determining how to best allocate marketing budgets is one of the most important strategic decisions advertisers make. The importance of these decisions increases when unexpected events occur, such as the COVID-19 pandemic and its ongoing impact on global markets. In times of uncertainty, it’s critical to focus spend on channels and tactics that are more likely to deliver measurable results.
CONTENTS
Marketing mix models, or MMMs, are a proven tool for developing more well-informed strategic plans. However, in order to continue to access useful planning insights, advertisers must adjust their models to account for COVID-19’s impact. Igor Skokan, Facebook Marketing Science, explains: “We advise advertisers to reevaluate their existing MMMs to determine the best way forward. Most existing MMMs are built without specific consideration for dealing with unexpected events and may not be able to control and accurately account for many of these changes. By updating their models now, advertisers can better understand the pandemic’s impact and get useful insights for their planning.“
To help advertisers continue to access the strategic insights they need to make important budget decisions, we’re launching a new guide: “Adapting Marketing Mix Models to Major Shifts.” Advertisers will get an overview of why traditional MMMs fall short in today’s economic conditions and how to adjust models in order to continue to provide strategic insights and guidance from trusted partners and industry leaders from Deloitte, Nielsen, Ekimetrics, Neustar, and IRI.
How unexpected events impact marketing mix models
Marketing mix models (MMMs) synthesize historical data into patterns and trends that can be used to make predictions about channel performance. But the COVID-19 outbreak has changed people’s behaviors and disrupted marketing across the globe, and even relatively recent data often can’t fully account for this impact.
In addition, impact differs greatly across the world. Consumer behaviors vary by region, country, etc., and in turn, policies and restrictions vary as well. Most MMMs are not granular enough to capture these local differences.
Moreover, many advertisers have significantly changed their creative in response, using new messaging and tones to appropriately connect with their audiences. Traditional MMMs often don’t measure creative impact, and so it’s difficult to assess the effects of creative changes without adapting the models.
Accounting for disruptive events in MMMs
Rather than abandon MMMs altogether, advertisers should seek to adapt their models to account for disruptive events. By following the five key steps outlined below, advertisers can update their models and continue to access useful insights to help aid with their media planning.
1. Account for changes in consumer behavior. Running MMMs more frequently allows advertisers to include more recent data, helping ensure the output of their models reflects rapidly shifting digital environments. Mark Vermut, Vice President of Marketing at Neustar, notes: “Advertisers should consider the lasting implications of consumer behavior. Retail online sales, for example, may remain higher than prepandemic conditions if there is a more permanent shift in how consumers are purchasing.” Advertisers can use recency variables, which allow models to calculate when a purchase was made, and custom adstocks for ads according to target audience, demographic, media channel, format and placement, but also optimization goals, target audience and other campaign-specific characteristics.
2. Include more granular data. Using more precise, granular data—such as data focused on specific geographies—delivers more targeted insights, especially in unusual times. “Solutions should focus on how COVID-19 has affected specific markets—country-state, country-province or country-region combinations—by the brand’s category and vertical,” says John Puhl, VP of Global Marketing Analytics Programs at Nielsen. By using granular data, such as daily, campaign-level, geo-region, direct marketing area (DMA) level data, marketers can better understand the impact of evolving behaviors. When modeling by geography to account for COVID-19, advertisers should keep in mind that the timing of both impact and response has varied. Puhl explains: “When collecting data relevant to the pandemic, it will be important to understand when the virus started to impact each market or when governments began to implement mitigation efforts.”
3. Consider how creative quality and media type impact performance. Ad campaign quality can vary, so not all impressions are of equal value. However, MMMs don’t typically account for creative quality and media type. This issue is especially key today, as many businesses have shifted their creative strategies—updating copy, using different imagery, etc.—in response to COVID-19. By comparing campaign performance according to adherence to best practices, advertisers can better contextualize the value of impressions and incorporate these assigned values into their MMMs.
4. Calibrate and validate MMMs. Running experiments helps advertisers check assumptions, choose among models, test against known outcomes and tweak MMMs to make them more accurate. When there is an increased scrutiny on marketing budgets or less relevant historical data to reference, calibrating marketing mix models with experiments is an effective way to ensure models are as accurate as possible. Riddhi Shah, Marketing Science Research Program Lead at Facebook, explains: “For a few years now, we have been encouraging clients to think of unifying their analytical tools. MMMs, though popular because of their ease and help in decision making, fall short in key measurement activities: measuring creative impact and incorporating lift results. Both of these become crucial during events like COVID-19, when brands are actively tweaking their creative to be sensitive to public sentiment. MMMs calibrated with experiments tend to be more accurate and in times when lift studies are not possible, even historical calibrations can be a game changer.”
5. Adjust optimization and scenario planning. Current audience behavioral patterns may not indicate long-term trends. Advertisers should evaluate findings from high-impact periods of the pandemic to determine if these should be included in planning. Adam Rodgers, Managing Director/GM at Ekimetrics, notes: “The key thing to remember about optimizations in this period is that there is no [one] solution. Brands should optimize under a different set of assumptions and constraints and also optimize to a wider set of outcomes.”
What it means for marketers
- While marketing mix models are a powerful tool for gaining strategic insights, many may need to be adjusted to account for disruptions and unexpected events, such as the COVID-19 pandemic.
- Accounting for changes in consumer behavior, including more granular data, calibrating marketing mix models with experiments and adjusting scenario planning are key to adjusting models to account for the pandemic’s impact.
Resource:
- app keyword installs
- buy app installs for android
