Skip to content
Leads

Leads

Build a local prospect list with AI help

ResearchedWhat this sticker meansAdvancedSmall business

Last updated:

Quick answer

Search your trade plus town on Google Maps, export the listings, and let the AI research each one and draft a personalized first line. Fifty real prospects in an afternoon.

At a glance

All outcome figures are estimates from cited sources, not independent verification by us.

What it is allowed to do
Drafts for you

Writes drafts (emails, posts, plans, replies). You review and send. Nothing goes out without you.

Reviewed Oct 19, 2026

Time saved (estimate)
About 3 to 5 hours per 50-prospect list
Difficulty
Advanced
Tools needed
ChatGPT, Perplexity, Google Maps
Typical cost
Google Maps is free. Perplexity has a free tier with limited web searches; ChatGPT Plus or Claude Pro speed up big batches but are not required.

Basis (estimate): Estimate only: clicking through 200+ listings by hand to find 50 qualified prospects typically burns a full afternoon; no timed study found.

What does this setup do?

Build a researched prospect list of local businesses: who they are, what is missing from their online presence, and a personalized opener for each.

Who is it for?

Web designers, marketers, and service providers who pitch local businesses like landscapers, cleaners, and contractors.

How do you set it up?

Tools

  • Google Maps (free) for finding businesses by trade and town
  • A spreadsheet for the list
  • A chat AI with web access (Perplexity, or ChatGPT/Claude for the research summaries)
  • Optional: a list-export tool if you want more than manual copying

Steps

  1. Pick one trade and one town. "Landscapers in Rutland" beats "businesses in Vermont."
  2. Search Google Maps for that trade plus town. Open the listings that look active.
  3. For each business, note: name, phone, website (or "no website"), and review count. Aim for 50.
  4. Paste the list in batches plus the prompt below into the AI. It researches each website and flags what is missing.
  5. Review the flags. Delete anyone who is not actually a fit, no matter what the AI says.
  6. For your top 20, have the AI draft one personalized opener each, based on the specific thing it found.
  7. Work the list in order. Log every contact: date, channel, response. One follow-up max for non-responders.

Example output

Illustrative. Your results will differ.

  • Your input turned into a clear, prioritized plan
  • Next 3 actions with dates, in priority order
  • You stay in charge. The agent just organizes.

Where to paste this prompt

ChatGPT: ChatGPT → Project instructions, or ChatGPT Tasks for scheduled reminders (once a day on Free)

Copy-paste prompt

Adapt bracketed fields to your situation.

Research these local businesses for my outreach list. Be factual and skeptical. If you cannot verify something, say so.

My service: [WHAT YOU SELL, e.g. "websites for trades businesses"]
My filter: I only want businesses that [CRITERIA, e.g. "have no website or a clearly outdated one, and look active with recent reviews"]

Businesses (batch of 10):
1. [NAME]: [WEBSITE OR "NO WEBSITE LISTED"]: [REVIEW COUNT IF KNOWN]
[ADD MORE]

For each business, return:
- What they do, in one line (from their site or listing).
- Online presence check: website? How old does it look? Active social profiles? Recent reviews?
- Fit verdict: GOOD FIT / MAYBE / SKIP, with one-line reason.
- If GOOD FIT: one personalized opener (two sentences max) referencing the specific thing you found. Factual and verifiable only: no invented details.

Rules: never invent a website, a review count, or a business detail. Mark anything unverified as unverified. Skip chains and franchises unless I say otherwise.

Newsletter

Get the free starter pack

Free starter pack with five beginner copy-paste prompts after you confirm, plus three sourced setups in your inbox each week. Unsubscribe anytime.

What usually goes wrong?

Where agents tend to stall on this job. More patterns in Where AI gets stuck.

  • Pitching from the list without verifying. An AI-hallucinated "outdated website" in your opener destroys trust instantly.
  • Scraping at scale without checking the rules. Google Maps data has terms of service, and cold outreach has spam laws. Keep it manual-sized and honest.
  • Using only public contact info is the line. Never buy or scrape personal emails or phone numbers.
  • Targeting businesses that already have what you sell. Check their site before you claim they need one.

Which chat tools work with this prompt?

Paste the prompt into any of these assistants (free or paid tiers).

  • ChatGPT
  • Claude
  • Perplexity

Common questions

Is it legal to pitch businesses I found on Google Maps?
Contacting a business at its public business contact details is generally fine. What crosses the line: harvesting personal emails, robocalling, or ignoring opt-outs. Keep it small, relevant, and human.
How many prospects should I start with?
Fifty researched names. That is enough to learn what works and small enough to verify by hand. Scale the outreach, not the sloppiness.
Can the AI do the whole thing without me?
No. It is fast at the research draft and unreliable at the details. Your job is verification and the actual outreach. The AI does the afternoon of clicking, not the judgment.

What tips make it smoother?

  • One trade, one town, one filter. "Plumbers with no website in Rutland County" is a list. "Local businesses" is a wish.
  • The best prospects are active but visibly missing something: no website, three photos, reviews from 2022. That is a solvable problem, not a dead business.
  • Verify before you pitch. The AI is fast at research and wrong often enough that every opener needs your eyes.
  • Work in small batches. Research 50, pitch the best 20, then build the next 50.

Where do the claims come from?

Summaries use public sources linked below, not independent verification. How setups are picked.

Related setups

Best AI setups for small business (2026) →

Ready for the next one?

Next up: a Research setup on the Small business path.

Research competitors through your customer’s eyes

Build a competitor comparison from public websites, ads, and reviews.

Roughly 2–4 hours per competitor scan