There is currently no good way to understand or categorize the valuable feedback customers provide in NPS Survey Comments. While Response Tags exist, they must be manually applied which is not viable or scalable. The issue is also complicated by comments being in many languages (as we run NPS surveys for ~19 languages).
To get around this now, in HubSpot, I have a Contact view filtered to NPS Detractors with a non-null NPS Comment. I export this to a file and then use ChatGPT or Claude to:
- Translate all comments to English
- Analyze the comment content and map to tags I have defined
- Add the tags, per comment, to a new column in the file.
- Send me the file.
- Use Excel to analyze and visualize the tag data.
It would be soooooooooooo much better if HubSpot could just do all this for me.
Note that being be able to categorize Tickets in the same way, and map to the same tags as the NPS tags, would also be great as we are doing that manually as well.
Point being: we need a scalable AI driven way to map incoming customer input (Tickets and Survey comments) to a standard Tag space (ie durable named issues in our business domain).