What CRC 2026 revealed about the future of market research

In his opening keynote address at CRC, Rohit Bhargava used an allegory that stuck with me. He told his audience that in 1968, Dick Fosbury won Olympic gold by going over the high jump bar backward. It wasn’t his technique that won him the title, but rather, it was noticing that foam had replaced the then-typical sawdust as cushioning in the landing pit. It made a new way of jumping feel all the more possible. While his competitors were staring at the bar as a means to an end, Fosbury was focused squarely on his destination.

That story — and in particular, its reference to the difference innovation makes — set the tone for three days of sessions, an Insights Association benchmark of 93 corporate insights leaders, and a session I co-led with Maxine Petrosino of Storck on why better observation beats better questions. Bhargava introduced his SIFT framework: create Space, uncover Insight, find Focus, seek a Twist, and it became the lens I saw everything through for the whole conference.

Market research is having a Fosbury moment.

The bar hasn't moved, but the ground underneath it has changed, and teams who don't notice the shift will risk wasting their time and resources trying to optimize a method that the industry has already left behind.

Space: The function has no room to think

The benchmark's resource picture is stark: among teams whose size changed this year, cuts outnumbered expansions roughly two to one, hitting mid-sized functions hardest.

Budgets are telling the same story, and nearly three in four leaders say demand for research now exceeds their capacity.

The industry's answer has been to focus on efficiency. AI adoption is nearly universal (98% of research organizations approve its use), and 40% of work is now done fully in-house. But panelists warned space can be filled badly. Delta's Lizzy Martinez mentioned that our brains default to shortcuts, and AI is the ultimate shortcut. In fact, relying on it too heavily is just a cut, taking chunks out of the deep thinking where real insights actually come from. Her rule: if it's hard, don't start with AI.

There's a longer-term risk too, which is that the hands-on work that AI now absorbs is exactly the work junior researchers used to glean invaluable learnings from. Automate the apprenticeship without replacing it, and today's efficiency becomes tomorrow's talent gap.

Insight: Stop asking, start observing

The core problem with simply asking consumers what they do, prefer, or want, is the say-do gap, which has widened as much as threefold in some categories.

A faster survey can't fix a broken foundation, and neither can AI if it’s trained on flawed data. Too much automation without innovation will just deliver the wrong answer faster, which will make it easier to trust and harder to question.

This was particularly prevalent around synthetic data. eBay's Thomas Walker argued that treating synthetic respondents as a stand-in for real people undermines the discipline's foundation, since math can't model the irrational mind. Adoption is still small but growing.

The strongest work at CRC went the opposite direction, toward real behavior.

Colgate-Palmolive's packaging model drew on 300+ pack images and 30,000-plus real shopper behaviors, then validated finalists with actual consumers. At Storck, testing 18 creative assets in real social feeds showed a "better-for-you" message outperforming a nostalgic angle with the growth audience, with newer consumers gravitating toward an unexpected product line. Observation goes beyond answering the question you asked by prompting brands to ask better questions.

Focus: Attention is scarce

Insights leaders rate the quality of their work highly, but actual satisfaction with the attention that work receives is markedly lower, especially at the largest organizations.

As the IA benchmark put it: quality isn't the problem, attention is. One team reported producing over a hundred reports a year with little visibility into how most of them are actually used.

Part of the cause is distribution: roughly two-thirds of UX, media, and advertising insight work that happening now happens outside the core insights function. When everyone generates data, the function's value has to shift from producing a signal to curating it. Does this kind of research process have a decision attached, or does it just produce content?

Focus also means tracking what leadership actually measures, like NPS, growth, and revenue influence. And yet, most functions still don't formally track research ROI. At Storck, linking activation tests to real purchase journeys gives a concrete answer. And mature organizations have moved from "which AI tool?" to "how do we govern it?”. The advice is now: pilot small, test against traditional methods, scale what proves out.

Twist — Find your option C

Bhargava's favorite example of a Twist is the futon — when the choice appears to be between “couch” or “bed”, imagine Option C… thus, futon.

The industry faces its own false binary: slow, rigorous traditional research versus fast, undertrusted AI output. The Option C at CRC was observed behavior at scale: real people, real environments, real decisions, actually measured rather than asked outright.

The Twist extends to positioning, too. Walker said he trusts agency partners more when they know what they want, especially when they're willing to turn down work that isn't a fit. Colgate's team framed their AI collaboration as a jugalbandi, the Indian classical duet where musicians trade phrases and push each other forward. Storytelling actually tops the training wish list at a rate of 75%, ahead of business fluency at just 37%, a major gap that SAP's Michelle Grant flagged: stories win attention, but fluency turns it into decisions.

Conclusion

CRC 2026 made one point from a dozen directions. Insights teams have the quality, tools, and appetite for influence.

What's missing is the method. The benchmark shows the quality is there, the tools are there, and the appetite for influence is there. Create space by using AI to buy back time for thinking, not to replace it. Find insight by observing what people do rather than trusting what they say.

As Bhargava closed, the only future we can make is one we're able to imagine. Industries, like people, tend to narrow their willingness to be open to new ideas over time, and the research industry still has plenty to learn.