Interview: Zooming into the consulting sector’s market research programs with IncQuery CEO Felipe Ochoa

This Q&A originally appeared on Boardroom Insight Leaderboard. Read the original article.

‍

The advice that consulting firms provide to clients is often based on buyer surveys. The way those surveys are carried out has changed significantly over the past few decades. Boardroom Insight recently caught up with Felipe Ochoa, the CEO of IncQuery, to get a look at what additional changes the market can expect in the near-term.

IncQuery is a Chicago-based software and services firm that specializes in market research. Consultancies and other companies use its AI platform to run survey-based feedback collection programs. Earlier this month, the software maker published a survey of its own that looked into how organizations’ research practices are evolving. It polled staffers at consultancies and private equity firms.

The first topic we asked Ochoa about is AI adoption. IncQuery’s study found that 88% of the surveyed market research teams are interested in the technology. They’re hoping to speed up more than a half dozen tasks ranging from data collection to survey response visualization. We were curious about one use case in particular: synthetic data generation. That’s the practice of using LLMs to generate simulated survey responses.

Ochoa also gave us a primer on competitor intelligence, which the survey respondents cited as the top focus of their research programs. Plus, we got insight into how consulting firms make sure their market research data is accurate in the era of AI hallucinations.

‍

Boardroom Insight: There are AI tools that can generate “synthetic” buyer feedback. Does synthetic data play a role in consulting firms’ primary research projects?

Felipe Ochoa: Yes, but mainly as a starting point.

Synthetic feedback can help a consulting team develop hypotheses, explore possible scenarios, or identify questions worth investigating. What it cannot do is replace evidence from a real buyer who is evaluating a product, considering a competitor, or making a purchasing decision today.

For questions such as whether customers would switch providers, accept a higher price, or value a particular feature, teams still need to speak directly with the relevant people in the market.

The way I think about it is that AI can help you reach a hypothesis faster. Primary research helps you determine whether that hypothesis is actually true. When the decision is important, a plausible answer is not enough. The team needs an answer it can defend.

‍

Boardroom Insight: When companies perform competitor research, what information do they look for exactly and where do they find it?

Felipe Ochoa: Companies usually begin with the visible parts of a competitor’s business: its products, pricing, positioning, target customers, partnerships, and go-to-market strategy.

Much of that can be found on company websites, in financial disclosures, product materials, customer reviews, databases, and industry reports. AI has made collecting and summarizing this information considerably faster.

But the most valuable questions are often regarding what buyers actually think. Which competitors do they seriously consider? Why do they choose one provider over another? Does a company’s claimed differentiation really matter? What might cause a customer to switch?

Those answers generally come from customer surveys, buyer interviews, and expert conversations. Public information can show you what a competitor says and what has already happened. Primary research helps explain why it happened and what customers may do next.

‍

Boardroom Insight: How do consulting professionals make sure their primary research data is accurate?

Felipe Ochoa: It’s difficult to guarantee that every data point is perfectly accurate. The goal is to create research that is reliable, transparent, and strong enough to support the decision being made.

That starts with the questionnaire. The team needs to be clear about the business question and make sure the audience, screening criteria, questions, and survey logic are designed to answer it. A problem with the survey or sample cannot always be fixed later during analysis.

Teams also need to verify that they are reaching the right respondents and monitor the data while the research is live. That includes checking for duplicate participation, unusually fast completion times, contradictory answers, and low-effort responses. Automated checks can help, but they should support human judgment rather than replace it.

Ultimately, the test is whether the team can explain who responded, how the research was conducted, and why the evidence supports the conclusion. Clean charts are helpful, but the underlying evidence still has to hold up when someone challenges it.