AI has proved to be just beyond good when it comes to processing large mounds of feedback.
GPTs may be good to get a high-level view to gauge whether the feedback is skewed (positive, negative, hovering around average).
If its a mass market the product is catering to, then its' possible that one is dealing with voluminous feedback, which could get tedious to process individually. One could classify verbatim under individual headers using GPT & then work their way through those bearing a [rating < 3*] which is the likely friction / pain point.
But, most orgs fitting into that bracket [1K+ verbatim per day] could have in-house tools / dedicated personnel to deal with the feedback.
Great article!
What are your thoughts on processing feedback via AI?
Do you have a method of when to strike a balance between reading summaries vs the actual feedback?
I find that GPT may glance over certain nuances.
Maybe the product size matters? But even then, have you come across a product where it’s tedious to go through the individual reviews?
For eg. App Store reviews are a good source and some apps can garner 1000+ per day.
What do these PMs do in that case?
Thank you Shawn, I'm glad it resonated with you.
AI has proved to be just beyond good when it comes to processing large mounds of feedback.
GPTs may be good to get a high-level view to gauge whether the feedback is skewed (positive, negative, hovering around average).
If its a mass market the product is catering to, then its' possible that one is dealing with voluminous feedback, which could get tedious to process individually. One could classify verbatim under individual headers using GPT & then work their way through those bearing a [rating < 3*] which is the likely friction / pain point.
But, most orgs fitting into that bracket [1K+ verbatim per day] could have in-house tools / dedicated personnel to deal with the feedback.