bots.new_
Sign in
USE CASEWorkModerate

“Turn ten years of calendar meetings into a Notion CRM, automatically”

Gaurav Munjal pointed Grok Bot at ten years of his calendar and told it to build a Notion database of every person he ever met - each with their LinkedIn, what they're doing now, and tags. It ran 2.5 hours straight and logged ~1,000 people from the first three years alone. A back-office job no human ever gets to.

Grok·0.0·0 saves·11 uses·via @gauravmunjal·Tweet by @gauravmunjal
Open @gauravmunjal on X

How it runs

5 steps. Do them in order.

  1. 01**Connect your calendar and Notion.** The bot needs read access to your calendar history and write access to a Notion database (it can create the schema for you).
  2. 02**Paste the prompt.** It works backwards through your meetings, pulling out every real person (skipping blocks, holds, and yourself).
  3. 03**For each person it builds a row:** name, how you met (which meeting, when), their LinkedIn, a short line on what they're doing now, and tags you define - industry, relationship, warmth, wherever you last left off.
  4. 04**It runs on its own.** A decade of calendar is a long backlog; the bot chews through it in the background and checkpoints its progress so a pause doesn't lose work.
  5. 05**You review the enrichment.** LinkedIn matching and "current status" are the parts most likely to be wrong, so the bot flags low-confidence matches instead of guessing, and you approve before it's trusted.

What you get

The value here isn't the database - it's that the database was never going to get built by a human. Everyone has a decade of relationships buried in their calendar and nobody has the afternoon to reconstruct it row by row. Handing that to an agent that works patiently in the background is exactly the shape of task these tools are best at: high-volume, low-judgment, tedious, and valuable only once it's complete. The design details are what keep it honest - working chronologically so "how we met" is accurate, checkpointing so a long run survives interruption, and treating LinkedIn enrichment as a flagged best-effort rather than confident truth. It's the back-office job of a chief of staff you never hired, and it doubles as a template: swap Notion for your CRM and "people I met" for "companies I researched" and the same skeleton rebuilds any relationship graph you've been meaning to.

Prompt

Copy it. Change the names. Keep the job.

Prompt
You are Rolodex, my relationship archivist. Build me a CRM in Notion from my calendar history.

SETUP:
1. Create (or use) a Notion database called "People" with columns: Name, How we met, First met (date), Last met (date), LinkedIn, Current role / what they're up to, Tags, Confidence.
2. Read my calendar starting from the most recent and working backwards. For each event, extract the real people who attended - skip focus blocks, holds, reminders, and me.

FOR EACH PERSON:
3. Add or update their row. "How we met" = the earliest meeting I have with them and its title. Update first/last met dates as you find more events.
4. Find their LinkedIn and a one-line summary of what they're doing now. If you are not confident it's the right person, put your best guess in the field, set Confidence to "low", and flag it for me - never present a guess as fact.
5. Tag them: industry, how I know them (colleague / investor / customer / friend / etc.), and anything else useful.

RULES: work in batches and checkpoint your progress so nothing is lost if you stop. Never invent a person, a company, or a LinkedIn. This is my private network data - do not share, post, or send it anywhere. When you finish a batch, tell me how many people you added and how many need my review.
Use withGrok

via @gauravmunjal · Tweet by @gauravmunjal

Tips

  • ·Connect your calendar and Notion, then paste the prompt. It works chronologically through your history, so a decade takes hours - let it run. LinkedIn enrichment is best-effort: it flags matches it isn't sure about rather than guessing, and you review before trusting any 'current status' field.
  • ·adhoc
  • ·acts-with-approval

Related setups

Same category. Different job.

More in Work →

Tomorrow’s meetings, one phone-readable page each

Krista’s daily meeting-prep routine pulls Salesforce, Gmail, Slack, Granola, and Gong into a short, skimmable brief she can read on her phone. New meetings get light research; existing ones get last-touch and the ask.

Work·Advanced·0 saves
Setup

Dump a messy ask, get plain-language results back

Kun Chen’s Firstmate setup is a single Grok Bot you talk to for everything. It creates, delegates, and juggles the other bots, and for bigger work it spins Cursor cloud agents, then brings the result back in plain language so you never have to chat the specialists.

Work·Moderate·0 saves
Setup