How to find clients: start with the people already asking for what you sell
How to find clients as a freelancer, agency or small business: the people asking for what you sell this week come first, then referrals, then outreach.
One of the quickest ways to find clients is to find the people who are already asking for what you sell. Every day, somewhere public, someone writes "can anyone recommend a bookkeeper who understands restaurants?" or "need a web designer, budget around $3k". Those posts are potential clients describing the job. A lot of businesses never see them, and the ones that do often answer too late or with too much of a sales pitch.
If you need work soon, a sensible order is: answer people publicly asking for what you do this week, then ask your recent clients who else they know, then contact businesses where something has just happened that creates your kind of problem. Marketplaces, content and ads come after that. The first three can turn into conversations within days; the others tend to take months, or money.
Ways to find clients, by how fast they tend to work
Most advice on how to get clients lists channels as though they were interchangeable: social media, networking, SEO, cold email, referrals. They aren't. They differ most in how long they take to produce a first client, and if you need work this month, that's the difference that matters.
This table is our rough judgement rather than measured data, and your market will shift it a little. But the broad shape is consistent.
| Channel | Time to a first client | What it costs | Limit |
|---|---|---|---|
| People publicly asking for it | Days | An hour or so a day | How often people ask |
| Your past clients | Days | A few emails | How many clients you've had |
| Businesses next to yours | Weeks | Coffee, and sending work their way | Grows as the relationships do |
| Outreach triggered by events | Weeks | Research time | High, if the research is good |
| Marketplaces | Days to weeks | Fees and price pressure | High, but at lower rates |
| Content and SEO | Months | Writing time | Very high, very slow |
| Paid ads | Days | Money from day one | Your budget |
Spend the first month on the top of the table, and invest in the bottom once the top is working.
1. Answer people who are asking this week
People ask for recommendations in public all the time: in small-business subreddits, local Facebook and neighbourhood groups, on LinkedIn and X, in trade association communities, and in Slack and Discord groups. These aren't leads you need to warm up. They have a problem, a rough budget and often a deadline, and they've asked.
To find them, search the way they'd write, not the way you'd describe yourself.
| If you're a... | Search for posts containing |
|---|---|
| Bookkeeper | "recommend a bookkeeper", "switching from QuickBooks", "behind on my books" |
| Web designer | "need a website for my", "who built your website", "Squarespace or hire someone" |
| Commercial cleaner | "cleaning company for our office", "recommend a janitorial" |
| Marketing consultant | "how do I get more customers for my", "agency or freelancer" |
| IT support | "need IT help for a small office", "managed IT recommendations" |
| Accountant | "need a CPA who knows", "accountant for small business" |
On Google, add site:reddit.com to the phrase and set the date filter to the past week. On Facebook and LinkedIn, search the same phrases and sort by most recent. It helps to check the same places at the same time each day, because these threads move quickly and the earliest useful answers are the ones people read.
When you reply, answer the question first. If someone asks how to choose a bookkeeper, tell them how to choose one: what to ask, what it should cost, and what a red flag looks like. Mention that you do this in one line at the end, and say plainly that it's you. Many communities remove self-promotion that doesn't come with a genuinely useful answer, and the rules vary enough that they're worth reading first.
One thing we've noticed from running agents that look for these conversations: platforms differ a lot. When ours searched Reddit between July and September 2026, 44% of runs came back empty, compared with 83% on X. And on LinkedIn, people posting about a problem were very often selling a fix for it, so the buyers were more likely to be in the comments. There's more on finding the right places here.
2. Ask your past clients a specific question
Your recent clients know other people with the same problem. They just won't think of them unless you ask the right way.
"Do you know anyone who might need my services?" usually gets a polite "I'll keep you in mind". Ask something they can answer by picturing one person: "Who else do you know who runs a restaurant with more than one location?" The more specific the question, the more likely you'll get a name.
Send it as a short personal email, not a newsletter. A handful of those tend to do more than a referral programme with a discount code.
3. Get to know the businesses next to yours
Every business gets asked for things it doesn't do. A web designer gets asked about copywriting, an accountant about bookkeeping, a physio about personal trainers. Those neighbouring businesses are already hearing your future clients' questions and sending them somewhere.
List five businesses that serve your clients just before or just after you do, and offer to send them work first. These relationships take a while to get going, and once they do, they're often the most reliable source you have.
4. Contact businesses where something has just happened
Outreach works best when there's a reason for it this week. A restaurant that just opened a second location may need a bookkeeper who understands two sites. An architecture firm that just won an award has a portfolio page suddenly getting traffic. A clinic with several new reviews complaining that nobody answers the phone probably has a booking problem.
Those events are public: job posts, local news, new listings, review sites, planning applications. It's worth being strict about which events really count as evidence, and your message should open with the event, not with you. These graded examples show what that looks like.
5. Marketplaces, content and ads
Marketplaces such as Upwork, Fiverr, Thumbtack or Clutch can bring clients quickly, and the price competition on them is real. They're useful for filling gaps and collecting your first reviews. Just don't let one become your only source.
Content is the slowest channel and the longest-lasting. One genuinely useful answer to a common question can keep bringing people in for years. It's a poor choice if you need a client this month, and a good one for next year.
Ads work from the first day and stop the day you stop paying. They make most sense once you know exactly who your clients are; before that, you're paying to find out.
How to find clients, by type of business
Freelancers do best starting with recommendation requests and past clients, and by describing themselves specifically. "Bookkeeper for restaurants" is far easier for someone to remember and pass on than just "bookkeeper".
Agencies usually win on a specific result for a specific kind of client, so lead with the sector where your best work is, and put time into neighbouring businesses and event-triggered outreach.
Local businesses should watch neighbourhood groups, local subreddits and review sites. People ask for recommendations in their town constantly, and a helpful answer from a local owner carries more weight than an ad.
B2B companies will lean more on events and public questions. This overview of sales prospecting techniques covers the trade-offs in more detail.
A first week, written out
Here's what a first week could look like for a freelance bookkeeper who works with restaurants. Swap in the phrases for your own trade.
- Monday. List ten places restaurant owners ask questions: a couple of subreddits, a few local Facebook groups, a hospitality Slack, LinkedIn. Save a search for each phrase in the table above.
- Tuesday to Friday, thirty minutes each morning. Check each saved search for posts from the last day. Answer the ones you can answer well, and skip the rest.
- Wednesday. Email your last ten clients the specific question: who else do you know running more than one location?
- Thursday. Find two restaurant groups that opened a new site in the last month, and write to them about the second-location problem, with the evidence in the first line.
- Friday. Note which places led to conversations, and drop the ones that didn't.
By the end of the week you'll know more about where your clients actually talk, which is worth more than any single lead. Working out which of those places are worth your time is a job in its own right.
What won't help
Posting about yourself and waiting. Announcing that you're "open for new clients" mostly reaches people who already know you, while the people who need you now are asking a question somewhere else.
Buying a list and emailing all of it. It feels like progress and mostly produces unsubscribes and spam reports. Prospecting is the part where you throw most of the list away, and skipping it just moves that filtering into the recipient's inbox.
Being everywhere at once. Seven channels at a tenth of your effort each rarely produce much. Pick two from the top of the table and do them daily for a month.
Revtive's agents handle the looking part of this. They watch the places your buyers talk for posts asking for what you sell, and bring each one to you with the source link and a drafted reply. Nothing gets posted for you; you decide what to say.