Getting clients · 23 August 2026

How to Find Web Design Clients

Finding clients is a filtering problem, not a hunting problem

Almost every article on this question sends you somewhere to compete: bid on Upwork, post on Fiverr, join a Facebook group, cold-DM on LinkedIn. Those are all the same move — go where buyers already gathered, and stand in a queue with everyone else who had the idea.

The queue has a price. Fiverr takes 20% of every order and Upwork's freelancer service fee runs 0–15% per contract, but the commission is the smaller half of the cost. The larger half is that your proposal sits beside eleven others and gets compared on number first. We did that arithmetic in full in what Upwork and Fiverr actually cost you.

The alternative is not “work harder at outreach”. It is to change what you are looking for. A business that has no website is not a lead you have to qualify — the qualification already happened, in public, and you can see it from the outside.

Where the gap is biggest

The global figure hides an enormous spread. In Germany 14.0% of local businesses have no website; in Thailand it is 80.4%. If you work in a saturated market, the same pitch that fails at home lands in a market where four in five businesses have nothing.

Two honest readings of that table. If you are in the United States or Germany, the first-website pitch is a minority game — 19.2% and 14.0% respectively — and the rebuild pitch is the bigger market. If you can work remotely and bill in dollars, the second half of the table is where the volume is.

The trade you pick moves the odds more than the pitch does

Trade matters as much as geography. A bar is found by being on the right street; an estate agent is found by being compared against two other agents online. That difference shows up starkly in the data.

Nobody in the top six ever decided they did not need a website. They simply never reached the moment where its absence cost them a customer they knew about. That is a very different conversation from convincing a contractor to replace something that already works. The full ranking is in which trades are easiest to sell a website to.

What a working list actually looks like

Concretely: one city, one trade, no real website, has a phone number. In a mid-sized city that is typically 80 to 400 businesses. That is the whole list, and it is enough.

The arithmetic is unforgiving but small. At a 3% reply rate, 200 contacts is six conversations. At a 30% close rate on conversations, that is roughly two clients. If your average project is $1,200, one afternoon of filtering is worth about $2,400 — and the list does not expire, because the businesses on it are still not online next month.

The three places everyone else looks

Marketplaces. Fastest to start, worst unit economics, and you never own the client relationship. Useful for the first three portfolio pieces; a trap after that.

LinkedIn and Facebook groups. Free, and genuinely effective for a while. The problem is that everyone selling web design is in the same group posting the same offer, so response rates decay in months rather than years.

Referrals. The best clients you will ever get, and completely outside your control. Referrals are a result, not a channel — you cannot turn them up when the month is quiet.

A filtered list is the only one of the four you can dial. That is the entire argument for it.

Before you send anything

Contacting a business is not the same as contacting a person, and the rules differ by country. A generic info@ address at a limited company is treated differently from a named individual's inbox under PECR, and CAN-SPAM, GDPR and CASL each want something different from you. It takes ten minutes to read and it is the difference between outreach and a fine: is cold outreach legal?

Questions