Find the friction pattern across 5 user interviews

User interviews produce data that's overwhelmingly qualitative and almost impossible to summarize honestly. Five stories, five contexts, five sets of complaints — and the temptation is to either cherry-pick the quote that fits your hypothesis, or write a 12-page synthesis no one reads. This use case finds the pattern without either failure mode.

The input

Five interview transcripts. Doesn't matter the format:

• Otter / Fathom / Granola exports
• Loom transcripts
• Hand-typed notes
• Mix of all three

Each one ideally tagged with the participant's role and team size.

The flow

1. Bulk-paste into Evidence Board. Open the Evidence tab and use the paste flow — drop all five transcripts in. Each one becomes a parent evidence item with child quotes extracted.

2. Let the AI extract candidate friction points. From each transcript, Pathmode pulls the moments where the user expressed frustration, confusion, workaround behavior, or unmet expectation. These show up as child evidence items, each with a quote and a type.

3. Generate clusters. Switch the Evidence Board to the Clusters layout and click Generate clusters. The AI groups items by underlying pattern — two users can complain about totally different features but be hitting the same friction ("the system requires me to know my own role before it'll help me"), and the cluster catches that.

4. Review the cluster summaries. Each cluster gets a short summary of the pattern and the items grouped under it. You can expand to see the underlying quotes, with their original participant attribution intact.

5. Synthesize an intent from the strongest cluster. Select the items in the cluster that matters most and click Synthesize Intent. The resulting IntentSpec is anchored to specific quotes from specific participants, so the conversation doesn't lose grounding when you bring it to engineering.

The output

A research synthesis that:

• Names the pattern in one sentence, not 12 pages
• Cites the evidence with participant attribution
• Distinguishes signal from noise by surfacing outliers as outliers
• Hands off to a spec without losing the qualitative texture

Five interviews become one shareable artifact your team can actually act on.

Why this beats traditional synthesis

Traditional research synthesis lives in a Notion doc that nobody reads after the readout. The friction is dis-anchored from the original quotes by the time it reaches engineering. By the time it's a spec, the original interviews might as well not exist.

The Evidence Board keeps the chain intact: every spec outcome can be traced back to the specific quote from the specific participant. The research investment compounds instead of evaporating.

Try it yourself

1. Gather 5 recent interview transcripts (or 3 — the flow works at small N)
2. Open Pathmode → product → Evidence tab
3. Paste each transcript as a separate evidence item
4. Switch to the Clusters layout, click Generate clusters
5. Select items from the strongest cluster and click Synthesize Intent

Related

• Playbook: Interview to Evidence
• Use case: Turn 30 support tickets into a prioritized spec
• Use case: Anchor every outcome to user evidence