2 Comments
author

@Artur: I like your use to group 5 years of customer chat for feature roadmap ideas. Getting a grip on qual data is one of the things AI is good for. I am less happy with trying AI for heuristic evaluation (as the originator of this method!) because that doesn't seem to be a current strength of the LLM.

A great case study that's similar to your analysis of customer chat is the way GE use AI to understand and track support tickets: https://jakobnielsenphd.substack.com/p/ai-qualitative-data-at-scale

What I particularly like about this case study is that they turn Qual data into Quant by having the AI count the number of support requests in different categories. This allows them to track over time if they are improving the design and whether new hot problems emerge that need fast attention.

For more ideas on how to use AI in UX, see article "Getting Started with AI for UX" https://jakobnielsenphd.substack.com/p/get-started-ai-for-ux

Expand full comment

Great write up!

One big question for me is, what has been some of the best use-cases today for AI in the UX practice?

As for myself and my startup: https://VisualSitemaps.com ( which automatically generates visual info architecture of public websites ) - I've used it to analyze and group 5 years-worth of our customer live chat messages for feature roadmap ideas. I've also used it to transcribe and synthesis salient points in long-form customer discovery video interviews.

I've even seen nascent attempts at AI performing UX heuristic analysis ( with varying degrees of success ). We live in exciting times.

Expand full comment