Before every external meeting, an agent assembles CRM history, recent emails, LinkedIn, and news into a 30-second briefing.
Account managers and salespeople walk into half their meetings under-prepared because the prep work — reading the last 5 emails, checking the CRM notes, glancing at LinkedIn for a job change — takes 10-15 minutes per meeting. Multiplied across the day, it's either skipped or it eats an hour. A meeting prep agent runs automatically the night before (or 30 minutes before) and drops a one-page briefing in Slack or email.
Subscribe to the rep's Google Calendar webhook. For each event with an external attendee, trigger the prep workflow N hours before start time.
Parse attendee email → extract domain → look up the company + the specific contact in your CRM. Bail gracefully if not found (still prep with public data).
(a) Last 5 emails between the rep and the contact (Gmail API). (b) CRM activity log: deal stage, last note, open tasks. (c) LinkedIn profile + recent posts (via Phantombuster or manual cache). (d) Company news search (Perplexity API or Tavily).
Prompt the model: 'Write a 6-bullet briefing: who they are, current deal context, last touchpoint, recent signal worth referencing, one question to open with, one risk to watch for.'
Slack DM with the brief + a link to the calendar event. Optional: email digest at 7 AM with all of today's meetings stacked.
After the meeting, send a one-tap Slack prompt: 'Was this brief useful?' (👍 / 👎). Use signal to tune which signals to include and which to drop.
Use these templates as-is or customize for your business.
You are creating a 6-bullet pre-meeting briefing for a salesperson. Inputs: CRM data {{crm_json}}, recent emails {{email_thread}}, LinkedIn profile {{linkedin_json}}, company news {{news_results}}. Output exactly these 6 bullets:
• Who they are (1 line)
• Current deal/relationship context
• Last touchpoint and what was promised
• One recent signal worth referencing in the meeting
• One question to open with
• One risk or concern to watch for
Keep it factual. Cite sources inline like [email 2024-04-12] or [LinkedIn] so the rep can verify.📋 *Meeting in 30 min: {{meeting_title}}*
👥 With: {{contact_name}} ({{contact_title}}, {{company}})
{{briefing_bullets}}
📅 [Open calendar event]({{cal_link}}) • 💬 [CRM record]({{crm_link}})After meeting end (poll Calendar 30 min after start time), DM the rep: 'Was today's briefing for {{contact_name}} useful? 👍 / 👎 / Mostly'. Log responses to tune which signals get weight in next iterations.Get a new AI workflow every week. Prompts, tool stacks, and ROI math included.
Multiple specialized agents collaborate: a router/orchestrator delegates to sub-agents (researcher, writer, classifier). Higher capability, more failure surface — invest in observability before scaling.
Learn the agentic glossary →Where this workflow tends to break in production — and what to put in place before you ship it.
Stale CRM data produces wrong context
Mitigation: Surface 'last updated' timestamps in the brief; flag if > 30 days.
Hallucinated quotes from emails
Mitigation: Force the model to quote verbatim with [email date] citations; reject paraphrases.
Skip if your team meets primarily with the same 5-10 contacts repeatedly (briefings become noise). Skip if data sources aren't connected — a partial brief is worse than no brief.
A phased approach to get this workflow running and delivering ROI.
Days 1–30
Foundation
Days 31–60
Optimization
Days 61–90
Scale
AI agents and traditional automation tools like Zapier solve different problems. Here is a clear framework for when each one is the right choice.
There are 500+ AI tools marketed to small businesses. These are the 8 that actually drive revenue for most SMBs — plus what to skip.
Three of the most popular AI agent platforms, but they serve very different users. Here is an honest comparison of pricing, ease of use, and when to pick each one.
One practical AI workflow per week. No fluff.
Get the full guide with step-by-step setup, workflow templates, and copy-paste assets.