Ask HN: How do you understand what users want when your interface is a chat?

[shubhamintech]

Building a conversational AI product and struggling with a specific problem: traditional analytics (clicks, funnels, heatmaps) are useless when there's no UI to track.

Curious how other founders handle this, how do you know what your users are actually trying to do? What intents are coming up most? Where are they getting frustrated?

Are you reading conversations manually? Building something internal? Just flying blind?

[Outboundex]

I see you're facing challenges in understanding user intents for your conversational AI product, especially without traditional analytics tools. Many founders have found success by leveraging AI to analyze chat interactions and identify pain points directly from user conversations. Have you considered using advanced tools that can automate this process and provide you with actionable insights? I'd love to hear more about your approach and see if I can offer some solutions that could help streamline your understanding of user behavior.

https://www.xmi-x.com/

[Bender]

Community Forum is one way, at the risk of letting customers talk to one another. Companies with high integrity can do this and the end result will often be customers helping one another find solutions. Similar concept to crowd sourcing.