The Customer Interview Synthesizer

Pulls a week of customer call transcripts into a single themed synthesis with verbatim quotes, ranked by frequency and product impact.

Saves 5 hours/week Submitted by Maya Patel, Founder/CEO at a B2B SaaS

What it does

Once a week, this agent ingests every customer interview, sales discovery call, and support escalation from the prior 7 days. It clusters the verbatim quotes by theme using semantic similarity, ranks themes by frequency and revenue impact, and outputs a one-page synthesis a founder can scan in 90 seconds. Stops you from confusing the loudest voice with the most common signal.

The Manus prompt

Copy this prompt and paste it into Manus AI to run the agent.

Every Friday at 4pm, pull all customer interview recordings, sales discovery call transcripts, and high-priority support tickets from the last 7 days. Cluster verbatim customer quotes by theme using semantic similarity. For each theme: count occurrences, estimate ARR impact based on the customer tier of speakers, and pull the three most representative quotes verbatim. Output as a one-page Google Doc: top 5 themes ranked by impact, three quotes per theme, and a one-line "what would change about the product if this is true" note per theme.

Tools required

NotionSlackGoogle Workspace

Best for

OperationsProduct