In Monday’s product meeting, your team decides an onboarding problem is an edge case. Over the next week, another twenty customers contacted support about it.
In an engineering review, you discuss a fix. After it ships, customers keep reporting the same issue.
In a leadership meeting, you make simplifying setup a priority. A month later, you want to know whether customers are finding it easier.
Your meetings capture decisions, assumptions, and commitments. Customer conversations provide evidence of how those decisions play out.
Today, we’re introducing Rippit’s Granola integration to connect the two.
Sync Granola meeting notes and transcripts into Rippit, analyze them alongside conversations from Zendesk and Intercom, and save the findings as structured data you can revisit. Then turn the analysis into an Agent App that runs on a schedule and delivers the results to Slack or email.
Meeting notes and decisions, read‑only
Why add Rippit if you already use Granola?
Granola helps you capture meetings and make use of what was discussed. Rippit brings those meetings into a broader analytical workflow across your conversation data.
| What Rippit adds | Why it matters |
|---|---|
| Broader conversation coverage | Analyze synced Granola meetings alongside Zendesk tickets and Intercom conversations to connect internal discussions, customer calls, and support feedback. |
| Structured, reusable analysis | Classify conversations into saved columns, then calculate counts, percentages, and breakdowns you can filter, chart, and revisit. |
| Managed recurring work | Turn the analysis into an Agent App with a defined data scope, output, delivery destination, and schedule. |
For example, start with what your team decided about onboarding. Then investigate how often the underlying problems appear in customer conversations, which workarounds succeed, and whether the situation improves after a change.
Bring Granola meetings into your analysis
Rippit’s Granola data integration imports meeting content on a schedule. Each synced Granola note becomes a conversation in Rippit, containing:
- The AI summary from Granola.
- Private notes, when present, stored as a separate internal comment.
- The full available transcript, organized into speaker turns.
- Meeting context, including the title, owner, attendees, folders, and a link to the original Granola note.
That gives you a dataset you can analyze repeatedly. You can investigate what was said across the synced meetings, save classifications, and compare those findings with customer support data already in Rippit.
Start with a question that needs both sources
Imagine your product team has just reviewed next quarter’s roadmap. The priorities and reasoning are captured in Granola. Meanwhile, thousands of support conversations describe the problems customers encounter every day.
You want to know:
“Compare the priorities discussed in our synced product-planning meetings with the issues customers raised in support last month. Which problems are we addressing, and where should we take another look?”
Here is how the workflow comes together.
1. Read the decision and the customer evidence
The meeting might explain why your team deferred an onboarding improvement: the problem appeared uncommon, a workaround seemed sufficient, or another issue was more urgent.
The support conversations let you examine those assumptions. Are customers using the workaround successfully? Is the issue becoming more frequent? Are they describing a different problem from the one discussed internally?
With meeting transcripts available in Rippit, you can investigate the discussion behind a decision as well as its summary. Source links let your team return to the original Granola note.
2. Turn conversations into reusable analysis
Ask Rippit to add the classifications your question needs. For an onboarding investigation, those might include:
- The problem discussed in each conversation.
- The onboarding step involved.
- Whether a workaround resolved the issue.
- Any commitment or proposed fix mentioned.
- The severity and supporting evidence.
Rippit can apply classifications across every conversation in the selected dataset and save the results in structured columns. Then calculate how often each issue appears, compare periods, and break the findings down by relevant fields.
The analysis remains available for follow-up questions, reports, and charts. You can reuse the classifications or add new ones as the investigation develops.
3. Make the investigation a standing job
Once the analysis is useful, turn it into an Agent App.
For example:
“Every Friday morning, analyze this week’s onboarding support conversations alongside our synced product and engineering meetings from the past 30 days. Highlight growing issues, reports related to documented fixes, and recurring complaints with no matching discussion in the meetings checked. Send a brief with evidence links to the product Slack channel. If nothing needs attention, send a short update saying so.”
Rippit builds a flow you can inspect and adjust: which records it reads, what it analyzes, how it formats the findings, where it sends them, and when it runs.
Test the flow, review the evidence, and activate it. Scheduled runs bring fresh customer feedback back into the discussion without someone having to remember to ask. Tests can send real messages, so use yourself or a test channel as the delivery destination while refining the app.
Agents that scan, analyze, and act
The same approach supports several recurring jobs:
| Agent | Scan | Analyze | Act |
|---|---|---|---|
| Priority alignment | Synced planning meetings and customer support conversations | Compare stated priorities with recurring customer problems | Send a weekly brief with evidence and questions for the next planning meeting |
| Fix follow-through | Meeting discussions of fixes and subsequent support reports | Identify problems still being reported after a confirmed release date | Send examples for the team to investigate |
| Customer commitments | Synced customer meetings and related support conversations | Identify promises mentioned in calls and evidence of unresolved issues | Deliver a follow-up brief to the account team |
Two ways to connect Granola
Rippit supports both a data integration and a Granola MCP connection. Choose the connection that fits the work, or use both.
| Granola data integration | Granola MCP connection | |
|---|---|---|
| Purpose | Sync meeting content into Rippit for analysis, reporting, and Agent Apps | Let Rippit’s agent search and read meeting content directly in Granola during an investigation |
| Setup | A Granola API key, available on Business or Enterprise; a Rippit workspace admin sets up the integration | Browser sign-in to Granola; a Rippit Full Admin creates the connection |
| Data access | Imports accessible notes, summaries, private notes, and transcripts; initial backfill up to 90 days | Reads content available to the signed-in user through Granola MCP, subject to plan and admin settings |
| Updates | New and edited notes sync on a schedule | The agent fetches content when it uses the enabled tools |
| Changes to Granola | Read-only | Read-only |
For the MCP connection, transcript access requires a paid Granola plan. On Granola Enterprise, an admin must enable MCP access. An MCP connection used by an Agent App must be shared organization-wide; its users access the connecting account’s available meeting content.
For the data integration, review the authorizing account’s access before connecting: private notes are included in the imported content. Disconnecting stops future syncs but does not remove data already imported, and the Granola API key remains valid until revoked in Granola.
See the data integration guide and the MCP connection guide for setup and access details.
Get started
Have a Rippit workspace admin connect Granola with an API key from your Granola Business or Enterprise plan. Once your meetings have synced, start with one decision you want to evaluate:
“What did we decide about onboarding in our synced meetings, and what are customers reporting about it in support now?”
Review the evidence, refine the classifications, and turn the version you trust into an Agent App.
Your team keeps making decisions. Your customers keep giving you feedback. Put an agent to work connecting the two.
Documentation reviewed October 1, 2026:Granola data integration, How data integrations work, Supported integrations, Ask Rippit questions, Agent Apps, Granola MCP connection, How MCP servers work, and Rippit MCP server.
.webp)
