Step01
Trigger Research from a Calendar Event
Create a Zap triggered by a new or updated calendar event. Filter for the customer meeting type and schedule the research far enough ahead to finish before the call.
IntermediateProduct
Create an Automated AI Meeting Prep Assistant with Zapier
Prepare for customer interviews automatically by triggering on the right calendar events, gathering product and CRM context before the meeting, and placing a concise source-linked brief directly in the notes page.
Source episode · 20:00
How I AI: Reid Robinson's Zapier Workflows for CRM Automation, Meeting Prep, and Feedback Loopswith Claire Vo
Reid triggers research for upcoming meetings, looks up company usage and sales history, converts the returned HTML into a file Gemini can summarize, and appends the brief to the customer’s Coda page.
Before you start
What you need
- A calendar with identifiable customer meetings
- Customer email addresses
- Authorized product usage and CRM lookup tools
- Coda meeting pages and Gemini access
What you’ll make
A pre-meeting brief in the correct Coda page with usage context, prior interactions, open questions, and source links.
Tools used
- ZapierVisitZapier
Workflow automation platform
- GeminiVisitGemini
Google AI assistant
Step by step
The workflow
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
4 steps
Step02
Fetch and Convert Customer Data
Use the external attendee email to retrieve the authorized account, product usage, and prior interaction data. Convert returned HTML or PDF material into a file and retain source URLs and timestamps.
Step03
Summarize Data with Gemini
Send the collected file to Gemini with a fixed brief structure. Require factual context, explicit dates, source references, and separate suggested questions.
Example prompt
Prepare a concise brief for this customer interview. Use only the attached account and interaction data. Include company and attendee, dated product usage signals, previous conversations, open commitments, risks or unknowns, and three suggested discussion questions. Cite the source or date for each factual section and label suggestions as suggestions.
Use a model like Google Gemini that is specifically optimized for handling file-based inputs (like PDFs, audio, or converted HTML) for more reliable and efficient processing.
Step04
Deliver the Brief to Your Notes
Find the Coda page for the exact meeting and append or replace a named “AI meeting brief” section. Record the generation time and source links, and alert the owner if identity resolution or research failed.
What good looks like
- Only intended customer meetings trigger the workflow.
- The lookup resolves the correct company and contact.
- The brief distinguishes observed data from suggested discussion topics.
- The Coda page is updated before the meeting without duplicating content.
Build your next product with ChatPRD
Turn an idea into a PRD, user stories, and a plan.
After the steps
Runbook notes
How to recover when the loop fails and where human judgment helps.
Recover
If it goes sideways
- The Zap runs for internal or unrelated meetings
- Filter on meeting type, title, attendees, or a dedicated calendar and require at least one external participant.
- The attendee email maps to the wrong account
- Use CRM identifiers and company domain, surface ambiguous matches, and do not merge data across accounts.
- The brief uses stale or unsupported facts
- Include data timestamps and source links, and label model-generated discussion suggestions separately.
- Slow research finishes after the meeting starts
- Trigger far enough in advance, cache slow lookups, and post a clear incomplete status instead of silently failing.
Keep building
Related workflows
intermediateSalesWarpFigma+1
How to Build an AI-Powered CEO Toolkit for Sales, Marketing, and Design
Automate and accelerate executive-level tasks by using AI agents integrated with your business tools. Learn how to use prompts to redesign Figma slides, analyze sales calls for insights, and generate lead lists from your email.
Open
advancedEngineeringWarp
How to Measure and Self-Improve Your AI Software Development Factory
Go beyond simple automation by creating a system that measures its own performance and improves itself. Learn to track metrics, score AI agent runs to find failures, and use observer agents to automatically fix your factory's code.
Open
advancedEngineeringGitHub Copilot
How to Build an Automated Software Factory from Slack to GitHub
Learn how to automate your software development lifecycle, turning a simple Slack request into a fully implemented, tested, and QA-verified pull request in GitHub, reflecting a true software factory model.
Open
Start shipping
better products.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.
Free to start
No credit card
SOC 2 certified
Enterprise ready