Every agency owner has been told to "post value" and "build in public." That advice is not wrong, it is just slow, and it does nothing in the month where you need two clients to make payroll.
This is what actually brings in clients, ordered by how fast it works.
Fastest: people who already know you
Before any outreach, work the list you already have. Former colleagues, past clients, people you did a favour for, the client who left eighteen months ago because their budget was cut.
Two things make this work rather than feel awkward. Be specific about what you want, and give a reason for contacting them now. "I've taken on two new clients in the dental space and have room for one more, do you know anyone?" beats "let me know if you hear of anything."
Past clients are the single most underworked source in the industry. Somebody who paid you before, and stopped for a reason that has since changed, is warmer than any cold prospect will ever be.
Fast: targeted outbound
Outbound has a reputation problem because most of it is bad. Done properly it is the most predictable channel available, because it does not depend on anyone finding you.
Build a list worth contacting
Do not buy a list. Build one from businesses that are visibly failing at the thing you fix. Businesses without a website are the clearest example, but the principle generalises: a visible gap gives you something to say.
For the mechanics, free lead generation tools for agencies covers the stack, Google Maps is the highest-yield source for local, and the local business lead finder pulls names and numbers directly if you would rather skip the setup.
Thirty well-chosen names beat five hundred scraped ones, because you will actually contact thirty.
Pick the channel the buyer uses
For local businesses, cold calling still works, largely because almost nobody does it any more. The number on their Google listing is the number they answer, because it is the number customers use.
For everyone else, cold email is the default. The two things that decide the result are the subject line and what you send when they do not reply, which matters more than the first message.
Lead with the leak, not the service
The message that fails describes your agency. The message that works describes something specific about their business that is costing them money.
"I noticed you rank on page two for [the term their competitor owns]" gives them a reason to reply. "We're a full-service digital marketing agency" does not.
Medium: partnerships and referral sources
The highest-quality leads most agencies get come from people who sell to the same customer without competing.
A web designer needs someone to run ads. An accountant hears about business problems before anyone else. A PR firm cannot do paid social. Each of those is a relationship worth building, and one good referral partner outperforms most marketing channels.
Make it easy for them: tell them exactly what a good referral looks like, and pay or reciprocate reliably.
Medium: paid ads
Ads work for agencies, but they are usually the wrong first channel because they cost money before they make money and they punish a weak offer.
If you go this route, using Facebook ads to get agency clients covers realistic costs and what to expect. Have your close process working first; ads amplify whatever is already happening.
Slow: content and inbound
Content works. It just does not work this quarter.
It compounds, it is cheap, and eventually it produces leads while you sleep. It is the right investment when you have enough clients to survive the wait, and the wrong one when you need revenue this month. Do both: outbound pays the bills while content builds.
The part that decides everything
Most agencies do not lose deals. They lose track of them.
You contact forty businesses. Six say "not right now, try me after summer." Those six are the most valuable names you have, and in practice almost nobody calls them back, because the spreadsheet has no memory.
Every prospect needs a state and a next action date: contacted, interested, follow up in eight weeks, dead and why. That "why" is the most useful data you will collect. Three "too expensive" answers in a week is a pricing signal. Three "we already have someone" answers is a targeting signal.
Once the list is bigger than you can hold in your head, that is a pipeline, and it belongs in a CRM built for agencies rather than a spreadsheet, for the simple reason that the follow-ups then actually happen.
When they say yes
Do not improvise the proposal. The agency proposal template covers the six sections that matter and, more usefully, the follow-up sequence for proposals that go quiet, which is where most "lost" deals actually sit.
A realistic first month
Week 1. List everyone who already knows you. Contact all of them with a specific ask.
Week 2. Build one list of fifty prospects in one niche you understand. Not five niches, one.
Week 3. Contact all fifty. Record every outcome and a next action date, including the nos.
Week 4. Follow up everyone who did not reply, and book two conversations with potential referral partners.
Then repeat with a second niche. The agencies that make this work are not the ones with the best script. They are the ones who are still contacting the "call me in three months" people three months later.
Frequently asked questions
How do I find clients for my agency?
In order of speed: people who already know you, targeted outbound to businesses with a visible problem you fix, referral partnerships with people selling to the same customer without competing, paid ads, then content. Most agencies invert this and start with content, which works but pays out in quarters rather than weeks. Work the warm list first because it costs nothing and converts fastest.
What is the fastest way to get clients?
Contacting people who already know you, with a specific ask and a reason you are contacting them now. Past clients are the most underworked source in the industry: someone who paid you before, and stopped for a reason that has since changed, is warmer than any cold prospect. After that, targeted outbound to a list you built yourself from businesses visibly failing at the thing you fix.
How does a digital marketer find clients?
The reliable channels are outbound to a self-built list, referral partners who sell adjacent services, and inbound content once it has had time to compound. The differentiator is rarely the channel and almost always the follow-up: most deals close on the third to sixth contact, which is precisely where an untracked spreadsheet fails.
How do you get your first ten clients?
Pick one niche you genuinely understand, build a list of fifty businesses in it, contact all fifty properly, and record every outcome with a next action date. Then repeat with a second niche. Specialising early feels limiting and is the opposite: it makes your second conversation better than your first, and lets you eventually say "I work with dentists," which sells far better than "I do marketing."
Is cold email or cold calling better for agencies?
It depends on the buyer. For local businesses, calling still works well, largely because almost nobody does it any more, and the number on their Google listing is the one they answer. For larger or remote businesses, email scales better. In both cases the follow-up sequence matters more than the opening message.
How long does it take to get clients for a new agency?
Measure your own rather than trusting an average, because it varies enormously by niche, offer and channel. Track conversations to closes from your first week, and after around thirty real conversations you will have a conversion rate that is actually yours. That number is far more useful for planning than any benchmark.
