Some phrases are not aimed at you, but they stay anyway.

I once heard a manager say: “Do you know which team goes first when there is a crisis? Marketing.”

And as often happens with certain phrases, it kept turning in my head because of how naturally he said it. As if it were a law. Apparently, there is an almost obvious order of sacrifice: first marketing, then everything else.

Something similar happens with R&D. When a company enters a crisis, it cuts innovation. The paradox is that, many times, the only thing that could get it out of the crisis is precisely innovation. So I wondered whether the same was true for marketing, and I went looking for evidence.

The first thing I found was uncomfortable for managerial common sense. The strongest academic evidence does not support the idea that cutting marketing during a recession is a smart decision.

A critical review of more than 40 studies on advertising during crises concludes that cutting during a recession usually damages sales both during and after the crisis. The same review shows that companies that maintained or increased advertising during recessions achieved better sales, higher share or stronger results.

That does not mean every dollar spent on marketing is sacred, or that every cut is a mistake. It means the automatic decision to “remove marketing because it does not sell tomorrow” usually confuses visible expense with real cost. A Journal of Marketing study analyzing 10,580 observations of public companies in the United States across seven recessions shows there is no single recipe. Effects depend on the type of company, the market it competes in, its share, leverage and whether it sells goods or services to consumers or other businesses. In other words: the problem is not adjusting; the problem is cutting by reflex.

With R&D, the story is even more revealing. The literature shows that companies tend to cut innovation in crises, especially when credit tightens. Several studies describe that procyclicality: when recession arrives, innovative investment falls, especially in financially constrained firms. But at the same time, evidence shows that companies able to better sustain intangible investments tend to navigate the recovery better. A review of the global financial crisis and its aftermath notes that firms that managed to maintain R&D and innovation showed better survival odds, more growth and greater profitability. And a study of U.S. companies during the Great Recession found that companies that did not reduce R&D achieved better operating performance in later years than those that did.

The real picture, then, is not “marketing is extra” or “R&D will have to wait.” The real picture is this: when a crisis arrives, many organizations stop reading those areas as investments in intangible assets and start seeing them again as discretionary spending. It is understandable. Their returns rarely appear fully in the quarterly close.

That is where the heart of the problem appears. For years, a formula was repeated as if it were rational: “data beats story.” But organizational experience and evidence on human behavior suggest something less epic and more realistic: very often, data does not beat any story. Very often, the story kills the data.

The problem is different: people do not process evidence neutrally. Psychology has shown for decades that we reason in motivated ways. We tend to reach the conclusion we want to reach, as long as we can dress it in reasonableness. We also tend to search for and value more strongly the information that confirms previous decisions.

Confirmation bias can operate almost unnoticed. Management under pressure needs to show control; control becomes scissors; scissors need a narrative; and then the narrative goes looking for data that makes it look responsible. The order matters: first the story is chosen, then the evidence is selected. In that movement, what looked like a pragmatic decision can become a strategic amputation.

Maybe that is why the phrase I heard stayed with me. Not because it was “strong,” but because it was sincere. It said out loud what many organizations do quietly: when crisis arrives, we do not always follow the data. We follow the story that best fits our fears, our urgencies and our incentives. Then, yes, we look for the spreadsheet that legitimizes it.

So no, maybe we do not live in an age where “data beats story.”

Maybe we live in an age where we idolize data, speak of data as the new oil, fill presentations with metrics… and still obey stories, especially when those stories allow us to do what we already wanted to do.

Because when marketing is cut by reflex, when R&D is sacrificed to “organize cash flow,” when long-term evidence loses to short-term anxiety, what remains standing is not a data-driven culture. What remains is something else.