Strategy & Planning

How to synthesize customer research interviews as Chief of Staff and Founder's Office

Strategy & PlanningFor Chief of Staff and Founder's Office2 apps10 steps~20 min to set up

You've just wrapped eight customer discovery calls ahead of a pricing strategy session. The recordings live in Zoom, the rough notes are split across three Notion pages and a Google Doc someone shared in Slack, and two interviewees replied with follow-up emails in Gmail. Synthesizing all of it into actionable themes falls to you — because the CEO is traveling, the product lead wants the summary by Thursday, and there's no researcher on payroll. You spend four hours copy-pasting quotes, tagging themes by hand, and writing a synthesis doc that half the exec team won't read before the offsite anyway.

Build this on Starch → See the apps used

Strategy & PlanningFor Chief of Staff and Founder's Office2 apps10 steps~20 min to set up

Outcome

What you'll set up

A searchable interview archive that pulls transcripts, email follow-ups, and Notion notes into one place, so anyone on the exec team can find the exact quote from three months ago without pinging you

An AI synthesis app that tags themes, surfaces contradictions, and drafts a structured findings brief — ready to hand to the CEO or drop into the board deck

An automation that files new interview notes into the archive and Slacks you a one-paragraph summary each time a new session is logged, so the backlog never builds up again

The Starch recipe

Apps, data, and prompts

The combination of Starch apps, the data sources they pull from, and the prompts you use to drive them.

Apps used

Meeting Notes Knowledge Management

Data sources & config

Starch syncs your Notion data on a schedule (pages and databases) and your Gmail on a schedule (messages and threads), so interview follow-up emails and async notes are all queryable. Starch syncs your Slack channels on a schedule so the summary automation can post directly to a channel. Meeting Notes captures and transcribes calls in real time and writes structured summaries. Knowledge Management organizes the growing interview archive and powers search across all of it. Zoom recordings and any web-based transcription services are reachable through browser automation if a formal integration isn't configured.

Prompts to copy

Transcribe today's customer interview with [Name / Company]. After the call, pull out: the three biggest pain points they mentioned, any exact quotes worth preserving, and any open questions they raised. Save the summary to my Customer Research database in Notion.

I have 8 customer interview summaries stored in my Notion database tagged 'Q2 Pricing Research'. Synthesize them into a structured findings doc: top 3 themes with supporting quotes, any contradictions across interviews, and a recommended framing for our pricing conversation at the exec offsite.

Every time a new note is added to my 'Customer Interviews' Notion database, send me a Slack message summarizing the key takeaway in two sentences and flagging any theme that contradicts something we heard before.

Run these in Starch → or paste them into your favorite agent

Walkthrough

Step-by-step

1 Connect Notion and Slack — Starch syncs both on a schedule, so your existing interview notes and any databases you've been keeping in Notion are immediately queryable, and Starch can post to Slack channels without you manually exporting anything.

2 Connect Gmail — Starch syncs your inbox on a schedule, which means interviewee follow-up emails and clarifications that arrived after the call are part of the same dataset as your Notion notes.

3 Install the Meeting Notes app from the Starch App Store. Before your next interview, open it and tell it who you're talking to and what the session is about — it will transcribe in real time and generate a structured summary with key decisions, quotes, and open questions when the call ends.

4 Tell Starch where finished summaries should live: 'After each interview, save a structured summary to my Customer Research database in Notion, tagged with the interviewee's name, company, and the research round (e.g., Q2 Pricing).' Starch writes directly to Notion so your archive stays in one place.

5 Install the Knowledge Management app. Point it at your Customer Research Notion database. It will index everything already there and make the full archive searchable — so when you or the CEO needs to find 'what did enterprise buyers say about annual contracts,' you get an answer in seconds instead of scrolling through eight separate docs.

6 Once you have five or more interviews logged, open a Starch app and type: 'I have [N] customer interview summaries in my Notion database tagged Q2 Pricing Research. Synthesize them into a structured findings brief: top 3 themes with supporting quotes, contradictions across interviews, and a recommended framing for our pricing conversation.' Starch queries Notion live and drafts the document.

7 Review the synthesis draft. Ask Starch to go deeper on any section — for example: 'Pull every quote related to price sensitivity from SMB interviewees specifically' or 'Show me the two interviews where the feedback on annual contracts was most different from the others.'

8 Set up a recurring automation: 'Every time a new page is added to my Customer Interviews Notion database, Slack me in #founders-office with a two-sentence summary and flag any theme that contradicts something heard in previous interviews.' This keeps the synthesis current without any manual effort between research rounds.

9 When you're preparing for the exec offsite or a board update, tell Starch: 'Using my customer interview archive from Q1 and Q2, write a one-page customer insight section for Thursday's pricing discussion — include the top jobs-to-be-done, top objections, and a direct quote for each.' Export or paste directly into the relevant deck or Notion doc.

10 After the offsite, use Knowledge Management's search to answer exec follow-up questions on the fly — 'did any mid-market customer mention implementation time as a blocker?' — without pulling up six different interview docs or remembering which call that came from.

See this running on Starch

Connect your tools, describe what you want, and the agent builds it. Closed beta is free.

Try it on Starch →

Worked example

Q2 2026 Pricing Research Synthesis — 8 Interviews, Exec Offsite Prep

Sample numbers from a real run

Interviews completed	8
Notion pages with raw notes	6
Gmail threads with follow-up from interviewees	4
Hours to manual synthesis (before Starch)	5
Hours to synthesis brief (with Starch)	1

The chief of staff at a 150-person SaaS company ran 8 customer interviews in April ahead of a pricing model overhaul. Six interviews had Notion notes in three different databases; two were just raw Gmail threads with no structured summary. She connected Notion and Gmail to Starch (both sync on a schedule) and ran the Meeting Notes app on the two remaining calls that week. After all 8 summaries were indexed in Knowledge Management, she typed: 'Synthesize the 8 Q2 Pricing Research interviews in my Notion database into a structured brief for the exec offsite — top themes with quotes, contradictions, and a recommended framing.' Starch returned a five-section document in under two minutes: three dominant themes (buyers want usage-based pricing, IT buyers block annual contracts, onboarding cost is the real objection), two direct contradictions between SMB and enterprise respondents on discount expectations, and a recommended framing for Thursday's session. The CEO reviewed it on the flight. The offsite started from a shared baseline instead of six people remembering different things from calls they weren't all on.

Measurement

How you'll know it's working

Time from last interview to completed synthesis brief (target: same day, not same week)

Number of exec decisions that reference a specific customer quote (tracks whether research actually influences strategy)

Percentage of open customer questions that get answered before the relevant meeting (vs. getting tabled for follow-up)

Reduction in 'can you find that quote from the interview' Slack messages to the chief of staff

Research archive search-to-answer time (how long it takes any exec to find a specific piece of customer evidence)

Comparison

What this replaces

The other ways teams handle this today, and how the Starch version compares.

Notion + manual synthesis doc

You're already using Notion for notes, but Starch is what lets you actually query across all of them at once — Notion search finds documents, Starch answers questions about what's inside them.

Dovetail or Aurelius

Purpose-built research repositories with good tagging and highlight tools, but they're another standalone system that doesn't connect to Gmail, Slack, or your board prep workflow — you'd still be copying outputs manually.

Otter.ai or Fireflies.ai

Good transcription, but the transcript sits in yet another tool with no path to synthesized themes or connection to your Notion archive and exec communication layer.

ChatGPT or Claude with copy-paste

Works for one-off synthesis when you paste everything in manually, but doesn't connect to your actual data sources, doesn't update as new interviews come in, and produces a doc with no home in your existing workflow.

On Starch RECOMMENDED

One platform — meeting notes, knowledge management all running on connected data. Setup in plain English; numbers stay current via scheduled syncs and live agent queries.

Try it on Starch →

FAQ

Frequently asked questions

My interview notes are spread across Notion, Google Docs, and a few Gmail threads. Can Starch pull from all of them?

Yes. Starch syncs your Notion databases and Gmail inbox on a schedule, so notes and follow-up threads are all queryable from the same place. For Google Docs, connect Google Drive from Starch's integration catalog — the agent queries it live when your synthesis app runs. If a doc lives somewhere else on the web, Starch can reach it through browser automation.

Does Starch record or join my calls directly?

The Meeting Notes app captures and transcribes calls in real time. If you prefer to keep using a tool like Zoom's native recording, you can feed the transcript into Starch through Notion or a connected file — and Starch will synthesize from there. You don't have to change your recording setup to get value from the synthesis workflow.

How does Starch handle the fact that some interviews happened months ago and the notes are inconsistent?

Starch queries what's actually in your Notion databases and Gmail, so inconsistency in the source material is reflected in the output. If some notes are detailed and others are three bullet points, the synthesis will be honest about gaps. You can ask Starch to flag which interviews have thin notes and which have the most evidence — that's a useful QA step before you take the brief into an exec session.

Is my customer interview data secure? These calls are sensitive.

Starch is not SOC 2 Type II certified today, so if your company has a compliance requirement at that level, that's worth knowing upfront. Data is not shared across customers. If your security policy requires on-premises or self-hosted tools, Starch doesn't offer that option currently.

Can I share the synthesis output directly with the CEO or the board without reformatting it?

Yes — once Starch drafts the synthesis, you can copy it into Notion, paste it into a Google Doc, or feed it to the Presentation Agent to turn it into slides. You can also tell Starch to format the output specifically for an exec audience: 'Write this as a one-pager for the CEO — three findings, one quote each, no more than 300 words.'

What if some of my interviewees are customers in HubSpot? Can Starch cross-reference the deal data?

Yes. Starch syncs your HubSpot contacts, companies, and deals on a schedule, so you can ask Starch to pull deal size, segment, or lifecycle stage for the companies you interviewed and layer that into the synthesis — for example, 'Show me how the pricing objections differ between customers on contracts above $50k and those below it.'