How to find local clients for your agency — a step-by-step method
Most agencies and freelancers prospect the same way: open Google Maps, type "hairdresser Kraków", scroll, copy phone numbers into a spreadsheet. Two hours later you have 40 businesses you know nothing about — and no idea which one is worth contacting.
This article describes a process that flips the order: signal first, contact second.
Why lists of businesses don't work
A list of businesses is not a list of leads. A lead is a business where you know three things:
- what's wrong (or what could be improved),
- why now is a reasonable moment to reach out,
- which argument to open with.
Without those three, you send the same thing to everyone. Reply rate: 1–2%. With them, you actually start conversations, because the first sentence is about that specific business.
Step 1: define the signal, not the industry
Start with one question: what has to be true about a business for my service to make sense?
| What you sell | Signal you're looking for |
|---|---|
| Websites | no site, or only a Facebook page |
| Local SEO | site exists but few reviews and weak visibility |
| Google Business Profile | incomplete listing, no replies to reviews |
| Reputation management | rating below 4.3 with a high review count |
| Google Ads | good site, good reviews, but low traffic |
The signal matters more than the industry. A hairdresser with no website and a mechanic with no website are the same lead — and get the same message.
Step 2: look where micro-businesses actually are
Big B2B databases (registries, LinkedIn, commercial lists) contain companies with accountants and HR departments. The local hairdresser, garage or beauty studio has no profile there — and those are the ones who buy websites, SEO and ads.
The one place they always are: Google Maps. Any service business that wants to be found has a listing. And that listing immediately reveals the signal: whether it has a site, how many reviews, what rating.
Step 3: score leads before you write
Don't contact the whole list. Split it:
- A — strong signal + reachable (no website, phone and email available)
- B — signal present but weaker (site exists but looks abandoned)
- C — no clear signal, business is doing fine
- D — not a fit
Contact only A and B. That's usually 30–40% of the list, and it accounts for practically all replies. The rest just burns your time and your domain reputation.
Step 4: the first message is about them, not you
A bad message starts with "We're an agency specialising in…". A good one starts with an observation.
Hi — Studio Aga has great Google reviews (4.9 from 87 ratings), but I couldn't find a website, only a Facebook page. People searching for a hairdresser on Google often don't click through to Facebook. We build simple sites for salons, about a week to launch. Want me to send two examples?
Three things this message does:
- proves someone actually looked at this business,
- names a specific problem without judging,
- ends with a small ask, not a pitch.
Step 5: rhythm beats bursts
The most common mistake: 200 messages over one weekend, then three weeks of silence. What works better is 15–20 contacts a day, five days a week. At a 20% reply rate and 10% conversion to a call, that's roughly 8–10 conversations a month — enough to close 1–3 clients.
How long it takes
Manually: about 3 hours to build a list of 40 businesses with context and write the messages. Per week, that's half a working day you don't invoice.
That's exactly the part worth automating. Lead Radary does steps 1–4 in minutes: finds the businesses, detects the signal, scores them A/B/C/D and writes the first message in the prospect's own language. You're left with step 5 — the actual selling.
Summary
- Search for a signal, not an industry.
- Micro-businesses live on Google Maps, not in B2B databases.
- Contact only A and B.
- The first sentence must be about them.
- Consistency beats volume.