How to synthesize customer research interviews on Starch

Customer research synthesis is the work that happens after the interviews: turning 8 to 20 conversations full of anecdotes, contradictions, and half-formed ideas into something you can actually act on — a clear picture of what customers want, what's blocking them, and which patterns are signal versus noise. Most operators know they should be doing this more rigorously. Most also know that a stack of Otter.ai transcripts sitting in a Google Drive folder is not the same thing as a synthesis.

What this looks like in practice depends on your context — a B2B SaaS founder coding job-to-be-done themes is doing something different from a consumer brand founder looking for language patterns, who is doing something different from a services operator trying to understand churn. The workflow shares the same shape, but the inputs, the outputs, and what 'done' looks like vary.

On Starch, you end up with a searchable knowledge base where every interview lives alongside the themes extracted from it, a recurring digest that surfaces new patterns as more conversations come in, and a presentation-ready synthesis you can drop into a board update or a product review — without spending a weekend manually tagging transcripts. Describe what you want: 'a running summary of customer interviews organized by pain theme, updated each time I add a new transcript.' That's what you get.