Strawberry

How to automate lead enrichment with AI

Two hundred bare domains go in. Filled, sourced fields come back, matched into your CRM, with every cell that could not be confirmed left blank and flagged.

A 200-row list comes back enriched: employee range, HQ city and country, industry, last funding round, CEO, and the right marketing contact, each with the source URL beside it. A companion does not slow down at row 140.

By hand that is two to three minutes a row, so the list from the conference is a full day nobody volunteers for. The five prompts below are the whole method. Paste them into Strawberry, swap in your sheet and field names, and check the first twenty rows.

What should each enriched row contain?

Enrichment fails at the edges: what counts as “employee count”, which contact is “the right one”. Decide the columns once and the companion holds them for every future run.

Prompt: “We’re going to enrich lead lists together, so remember this format: for each company I want employee range, HQ city and country, industry, last funding round with date, CEO name, and the LinkedIn URL of their head of marketing. Every fact needs a source URL in a column next to it. If you can’t confirm something from two places, leave the cell blank and flag it. Never guess.”

How small should the first batch be?

Twenty rows. Enough to catch a misread column before it multiplies by 200.

Prompt: “Open the sheet ‘Webinar signups (March)’ in my Google Sheets. Enrich the first 20 rows using the format you remembered. Use Apollo where it has the company, the open web where it doesn’t. When you’re done, tell me which cells you left blank and why.”

The companion opens the tabs, cross-checks sources and writes into the sheet. Check those twenty rows the way you would check a new hire’s first batch.

Correct it once, then run the rest

Say what it got wrong in plain language and it carries the correction forward. Then release the other 180 rows.

Prompt: “Good, but two fixes: use the parent company when the domain is a subsidiary, and for funding, treat anything older than 24 months as ‘none recent’. Now do the remaining 180 rows the same way.”

Get the fields into your CRM

A finished sheet is not finished work.

The fields need to land where the reps look. A companion matches each row against existing HubSpot records, and does the same in Salesforce, Attio or Pipedrive through your logged-in session.

Prompt: “Match each enriched row against HubSpot. Update existing company records with the new fields, create records for the ones that don’t exist, and give me a list of possible duplicates instead of merging anything yourself.”

Merging is destructive, so it stays your call. The companion stages, you approve, which is the pattern across CRM work in Strawberry.

How do you put it on a schedule?

Routines run a saved pass on a schedule. Describe the loop once and you stop having an unenriched list.

Prompt: “Every weekday at 7am, check the ‘Inbound’ sheet for rows added since your last run, enrich them in our format, sync them to HubSpot, and Slack me a two-line summary: how many rows, and anything you flagged.”

New signups are enriched, sourced and in the CRM by the time you open your laptop.

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide

Frequently asked questions

Yes. A Strawberry companion opens each company’s site, LinkedIn and sources like Apollo, cross-checks the facts, and writes them into your sheet or CRM with a source URL per field. Anything it cannot confirm is flagged rather than guessed.

Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide