Are we headed for a recession in the second half of 2022? How about in 2023? Are you ready with a bullet-proof marketing plan?

Suppose you’re a marketer at ACME Travel Corp. with a large media budget to spend this year and big objectives to meet. Your customers are traveling again, and you’re ready to make up for lost time. How’s the marketing mix model you ran back in January going to hold up now that it looks like the economy might enter a downturn?

If you’re looking for answers in the flurry of conflicting news coming from economic research bureaus, corporate earnings reports, investment banks, or media pundits, good luck. Goldman Sachs’ ex-CEO Lloyd Blankfein warned in May that a recession this year was all but assured, only to dial back that prognosis two weeks later—the same day that news leaked out that Elon Musk had a “super bad feeling” about the economy and was ordering job cuts. Until he wasn’t anymore. Getting whiplash much?

Recession or not. Which is it?

There are many signs pointing to a looming recession: At 8.6%, inflation in the US is as high as it’s been in 40 years, and showing no signs of abetting; people’s personal savings rate just took a nosedive; and the country’s GDP is down for the first time since the onset of the pandemic. Many companies are now entering a hiring freeze, especially in tech. And according to a recent YouGov survey, most Americans think we’re already in a recession.

On the other hand, unemployment has dropped back to pre-pandemic levels. A new report by TransUnion shows that since the beginning of the year, household income has increased or stayed the same for 81% of the population, and 90% expected it to continue to rise over the next 12 months. Consumers buoyed by rising pay and savings amassed during the pandemic continue to spend at record levels. Further, says the Business Insider, the May Bureau of Labor Statistics jobs report showed exactly the right numbers—not too hot, not too cold—to actually dispel fears of an immediate economic downturn.

Taking matters into your own hands

Surely, consumers won’t be able to keep spending forever by dipping into their savings. Are we headed for the cliff, or is the Federal Reserve going to be able to orchestrate a soft landing? At ACME Travel Corp., you have some experience with jumping off a cliff, so you and your team are buckling up and getting ready for the ride.

Not so fast. You don’t have to live by the word of the marketing mix models (MMM) you ran three, six, or (gasp) 12 months ago, and wait anxiously for the world to line up to your forecast. MMM gets a bad rap because it’s often complicated, expensive, and time-consuming. Too often, its results are obsolete by the time reality catches up to expectations.

It’s time we looked at it differently.

Data never stops working. Neither should your models

MMM has been left for dead a few times, most recently with the rise of marketing attribution. But it’s been around for more than 50 years, and it’s still an essential tool in the marketing toolbox. Its statistical principles are solid, and most of the top consumer goods companies in the world use it.

But too many top executives still believe that a model with data going back two to three years (to account for seasonality) should be good for at least a few quarters going forward. They don’t see the value of refreshing their models more frequently. Even worse, they fail to invest in the data infrastructure necessary to make more frequent modeling possible.

Today, data never stops flowing. Cross-channel media impressions and customer transactions can be captured in real time. Most macroeconomic indicators get published at least once a month. There’s no reason to wait for the end of the quarter to run a new model. MMM and other tools, like marketing attribution, can be combined into a virtuous circle where MMM insights (like offline activity or market conditions) inform the attribution analysis, and where attribution performance data is fed back into the MMM analysis.

McKinsey calls this approach a full-funnel marketing strategy. At Neustar, we call it unified analytics: An always-on solution where marketers use the best data available to continuously optimize how they allocate their media budgets, online and offline. It’s the future of measurement.

Lessons learned during the pandemic

In uncertain times, good planning is at a premium. Anyone can predict the future when nothing changes. But I learned a few lessons helping retail clients work through the market uncertainties brought on by the Covid-19 pandemic over the past two years. Lockdowns, stockouts, and fundamental changes in how people shop, work, and consume media have a way of testing even the most robust of marketing plans. Those lessons are still very relevant today as we’re teetering on the edge of a recession.

Don’t just refresh your data, but your variables too

One of the key lessons from the pandemic was to be constantly on the lookout for new variables to explain new behavior. For instance, early in the pandemic, we turned to the Google Community Mobility Reports to pinpoint where and when exactly foot traffic was slowing down (based on phone location data), and quickly integrate those findings into our clients’ predictive models. We also ended up using official figures for Covid-19 cases in some models, but found them to be lagging variables for consumer behavior. Google searches like ‘what is coronavirus’ were better—and more timely—predictors.

Automatic, but Not Thoughtless: The Future of Creative Marketing

Prioritize local data over global data

It seems like it should go without saying, but there’s simply never enough local data in any model. Remember that many companies are starting their marketing planning process from a place that tends to favor market stability: budgeting. When an alarming report comes out, like the recent worldwide stagflation forecast from the World Bank, the temptation is often to apply a dampening factor to all models across the board. But during the pandemic, smart marketers were able to finetune their media planning on the fly based on not just local lockdowns, but local shifts in shopping behavior too.

For example, in Greece, the online shopping population jumped from 33% to 59% in 2020—a 77% increase, by far the strongest in Europe. Imagine the missed opportunities if that data isn’t in your predictive model.

Always assume the world is changing

So, the question of whether or not a recession is coming should be irrelevant to your marketing planning. You should always assume that the world is going to change, and set up a data and analytics ecosystem that can keep up with those changes. And you should endeavor to develop a corporate culture where predictive modeling isn’t a way to insulate your organization from every twist and turn on the road, but rather make the most of those in-the-moment opportunities.

I’m not saying it will be easy. But if you can spot the cliff, you can hit the brakes and navigate around it.

Recommended: Customers Want More From Customer Experience – Here’s How Technology Can (And Can’t) Help

[To share your insights with us, please write to sghosh@martechseries.com]