UX research platforms have carried methodological blind spots for years. Nielsen Norman Group's core argument: most of these tools, some now nearly two decades old, were never built by expert researchers or with meaningful input from them. That was a manageable problem when the worst outcome was a slow workflow.
The risk profile has changed. Research tools now plan studies, moderate sessions, and analyze results autonomously using AI. When the underlying methodology is flawed, the output is not just inconvenient. It is bad research delivered at scale, with algorithmic confidence attached.
The full article maps the historical landscape of UX research tooling before dissecting where AI has compounded existing failures. The value is not in the conclusion. It is in the specific breakdown of which methodological assumptions get baked into these platforms before a researcher ever touches them.
[READ ORIGINAL →]