The Hidden Cost of Offloading to AI in Strategy and Research

Can AI rapidly speed up your ability to collect and analyse data? Yes.

Can using it to do these things also create unintended negative consequences? Yes, definitely.

Where AI can add value in user research

My two cents on using AI in strategy and research work is that it’s great at doing some of the laborious stuff, like analysing screeds of transcript data or identifying patterns and trends. We all know that.

But if you’re not careful, offloading critical—but sometimes ‘mundane’ or rote—tasks can result in a poor overall understanding of the very thing you’re trying to understand.

Cognitive offloading in User Research

Let’s think of this in the context of user or market research for a second. You could get AI to do an analysis for you and spit out key themes or takeaways, but they aren’t your key themes or takeaways. When you have to defend something you had no stake in creating, it’s not really yours. You can’t ever truly understand it, because you didn’t do the work to get there.

The devil is in the details

And that matters. Because the real value of research and strategy work often lies in the sense-making. It’s in sitting with ambiguity, wrestling with competing insights, and slowly forming a point of view. That’s not something AI can—or should—replace. It’s the difference between seeing the output and understanding the process that got you there.

Replace the user research analogy with pretty much anything—customer journey mapping, business model design, competitive analysis—and the same principle applies. The negative effects of cognitive offloading are real. You lose the depth. You miss the nuance. And most importantly, you risk making decisions based on insights you don’t fully grasp.

A time and a place for AI

That’s not to say AI doesn’t have a role to play. It absolutely does. It can take the edge off the tedious parts of the process, surface patterns faster, and help you explore more in less time. But it should augment your thinking—not replace it.

If you want to produce work you can stand behind—work you own—you need to stay close to the material. Do the heavy lifting, make the connections, challenge the output. That’s where the insight lives. That’s where the value is.

If you’d be interested in chatting about how you can use AI to help, rather than hinder your research efforts, give me a shout.