| We did something interesting recently, but not sure how to proceed. Ideas are welcome. • Context: We (an AI for Support startup) recently onboarded a support team that had no formal KB/HelpCenter. Their “knowledge” was scattered across tickets and Slack threads. • What we did:
– Ran an AI pipeline to convert ~6 months of support tickets into 1700 FaQ/KB-style entries.
— The typical entry was under 200 words, and the average of 150 words.
– These ended up as Q&A/FAQ snippets rather than polished, full-length KB articles.
— Such information was not available elsewhere, but they cannot be called KB articles, as they don’t look and feel like KB articles; they are more like FAQs.
— The customer asked if we could auto-tag these articles; that's the next step. We have not done it yet. • Insights:
– Volume ≠ structure: Writing 1,700 full articles manually is impossible. AI gave us breadth, but these entries aren’t organized like a typical KB (no headings, related topics, etc.).
– Search-ability wins: Even though they’re rough around the edges, the Q&A format indexes really well—agents find answers fast via search, which was the primary goal.
– User experience gap: Navigating dozens of standalone Q&A entries in a HelpCenter is clunky. There’s no clear hierarchy or grouping by topic. So this is not good UX for navigation.
• Outcome: The customer is happy—agents can now “search and serve” answers instead of digging through old tickets. That immediate boost in efficiency seems valuable. • Next steps (brainstorm):
– Should we layer a lightweight organization (e.g., tags or categories) on top of these Q&A entries?
– Can we automatically cluster related Q&As into mini-articles or topic pages?
– Is there merit in a hybrid approach: AI-generate bulk Q&As, then enlist human editors to refine high-impact sections? I’m convinced there is something here. not sure what it is. Thoughts? |