Let me start with a number that should make any business leader uncomfortable: According to McKinsey's 2025 State of AI report, 88% of organizations are now using AI in at least one business function. Sounds impressive. But when you dig a layer deeper, the majority of those companies are still in the experimenting or piloting phase. Only about one third have started to scale anything meaningful. And in customer insights specifically, the gap between what is possible and what most companies are actually doing is enormous.
This is not a technology problem. The tools exist. They are accessible, they work, and in many cases they are not even that expensive. This is an urgency problem. And the companies that figure that out first will have a significant advantage over everyone else.
What AI actually does for customer insights
There is a lot of noise around AI in business right now. So let me be specific about what it changes in the context of customer research and insights, because that is where I have spent the last decade.
Traditionally, insight work has been slow. You design a survey, collect responses, export the data, clean it, analyze it, produce a report, present it to stakeholders, and then wait for someone to decide what to do about it. By the time action happens, the customer reality you captured may already have shifted. You end up managing the past instead of shaping the future.
AI compresses that entire cycle. At Feedbackly, we built text analytics powered by AI that can process open-ended customer feedback at scale, find patterns, cluster sentiment, and surface actionable signals in real time. What used to take a team of analysts days now happens continuously in the background. The insight function stops being a reporting machine and becomes something much more valuable: an early warning system and a growth engine at the same time.
The specific areas where AI creates the biggest leverage in customer insights are: automated text and sentiment analysis across large feedback volumes, pattern detection across journey touchpoints, predictive modelling of churn and conversion, and rapid qualitative research synthesis. Each of these on their own would justify the investment. Together, they fundamentally change what an insight team can deliver.
The speed problem that nobody talks about enough
Here is what I have seen happen in organisation after organisation. A customer insight team produces genuinely good work. The analysis is solid, the recommendations are clear, the data tells a real story. But by the time it reaches the people who can act on it, three things have happened.
First, it is two months old. Second, the business has already moved on to a different priority. Third, nobody has connected the insight to a commercial outcome in a language that a CFO or a VP of Sales actually responds to.
AI does not just make insights faster. It makes them more alive. When sentiment shifts in your customer feedback, you can know about it the same week it starts happening, not the same quarter. When a specific touchpoint in your customer journey starts generating more negative emotional responses than usual, you can catch it before it becomes a retention problem. That is a fundamentally different kind of insight function.
According to IBM research, analysing customer interactions with AI helps companies improve responses and reduce costs by 23.5%. That figure alone should be enough to end most internal debates about whether the investment makes sense.
Why most companies are still too slow
If all of this is true, and the tools exist, and the ROI is clear, why are so many companies still running their insight functions largely the same way they did five years ago?
From what I have seen, there are three real reasons.
The first is that insight teams are often measured on output volume rather than business impact. Reports produced, surveys launched, dashboards updated. When your KPIs reward activity, there is less internal pull toward transformation.
The second is that AI adoption in insights requires someone who understands both the technology and the research craft. That profile is rarer than people expect, and most organisations have not invested in building it. According to Deloitte's AI research, lack of technical expertise is consistently one of the top barriers to meaningful adoption. You can buy the tools, but if nobody knows how to ask them the right questions, the value stays locked.
The third reason is cultural. Insight work often lives inside a function that has historically been asked to inform decisions rather than drive them. Shifting that dynamic requires leadership willing to give the insight function a sharper commercial mandate and a seat closer to where strategy actually gets made.
What fast looks like in practice
I will give you a concrete example. One of the implementations I worked on at Feedbackly involved connecting emotional experience data from customer surveys directly to transaction data. The goal was to understand which emotional states at the pre-purchase stage predicted higher basket values and repeat purchases.
The analysis that would have taken weeks of manual work was automated. The signal was clear within days. Customers who felt trust and excitement at the product discovery stage converted at significantly higher rates and spent more. Customers who felt confusion or indifference at the same stage churned. That is a finding with direct implications for product, marketing, and customer service, and it came from combining AI-powered text analysis with emotional measurement rather than relying on NPS alone.
That is what fast looks like. Not fast for the sake of it, but fast in a way that connects the customer reality to a business decision with a short enough gap that something can actually change.
The insight function of 2026 and beyond
The companies that will lead in customer understanding over the next few years will not be the ones with the most data. They will be the ones with the shortest distance between a customer signal and a business response.
AI makes that possible. But it requires insight leaders who are willing to redesign how their function works, advocate for the right tools, build the right capabilities in their teams, and translate findings into commercial language that moves people to act.
The technology is ready. The question is whether the organisation is. Most are not there yet. Which, if you think about it, is actually good news for the ones who decide to move now.
Jaakko Männistö is a CX professional, growth leader and the developer of EVI® (Emotional Value Index). He has worked with 450+ brands globally on customer experience strategy and measurement. He is the author of "The Journey – How to Create the Happiest Customers in the World."