How to get web design clients with one page

A high-ticket client hires on evidence of judgement, and a grid of twelve thumbnails contains none. It shows that you finished twelve things. It does not show that a single decision inside any of them was made on purpose, which is the only question the buyer is actually asking.

The verdict: stop optimising where you post and rebuild what lands. One case page, one project, shown completely, with the design decisions written as values. That page closes work that a hundred cold messages and a redesigned portfolio grid will not.

The channels do matter, and they matter second. Every channel is a way of getting somebody to look at something, and if the something is a thumbnail grid then every channel performs equally badly.

What a high ticket client looks at first#

Watch a studio principal open a designer's site. They scroll the work, stop on the one project closest to their own, and look for two things: whether it resembles their industry, and whether anybody made a decision they can name. Thirty seconds, and the grid loses on both counts because a thumbnail carries neither.

What survives that scroll is a single project told at length. A buyer trusts a designer who can explain one choice more than a designer who can display twenty outcomes, because the explanation transfers to their project and the outcomes do not.

Three things they are checking, in order:

1. Have you worked on something shaped like mine? Not the same industry necessarily, the same problem: a lot of photography, a long service list, a booking flow, a multi-location footer.
2. Did you decide, or did a template? Naming the type ratio answers this in one line.
3. What happens after launch? A photograph of a handover document answers it better than a paragraph promising support.

Not one of those three is answered by a logo grid, a services list or a line about your process. All three are answered by a single project told properly, which is why the thing to rebuild is a case page and not a home page. The home page has one job, which is getting somebody into the case page quickly.

The case page, block by block#

Eight blocks, in this order, on one page per project. Length limits are part of the spec, because a case page that runs to two thousand words gets skimmed into nothing.

Block four is the one nobody writes and the one that wins the work. Write it as values: "headings run a 1.25 scale from 16px, so 16 / 20 / 25 / 31 / 39 / 49 / 61 and nothing between them, at 1.05 leading with -0.02em tracking above 31px." A sentence like that cannot be written by somebody who did not make the decision, which is precisely why it functions as proof.

Block five is the one that needs a recording rather than a screenshot. Motion is half of how an expensive site feels and a static image carries none of it, so show the entrance at cubic-bezier(0.16, 1, 0.3, 1) over 560ms actually happening.

Three projects, one niche#

Three complete case pages in one niche beats twelve across nine. A buyer scanning for a specialist reads breadth as availability, and availability is not a selling point at $12,000.

If you do not have three in one niche, build the missing ones on spec. Pick a real named business, use their own photography, and say plainly on the page that it was unsolicited. Nobody has ever been offended by this, and a clinic owner looking at their own treated photographs is evaluating your judgement rather than imagining it.

How to pick the target for a spec build:

• Good photography, badly used. Their best asset is already paid for.
• One decision maker, named on their own site.
• A service with a case value above $2,000, so your fee is small against one sale rather than large against a month.

The spec build takes an afternoon when the direction already exists. That is the part Thanor covers: art direction per design in exact values, for client work included, so the page you build to win a clinic is a real page rather than a placeholder. Which niche to aim at is a separate question with a clear answer.

The approach that follows the page#

Forty named businesses in one niche, researched, with one observation each that could only apply to them. Not a pitch, an observation: their case photographs are good and the page crops them square at aspect-[1/1], or their headings and body text are both set at 1.5 leading so the page has no hierarchy at all.

Then one link, to the case page closest to their situation. The observation earns the click and the case page does the selling, which is the opposite of the usual arrangement where a long message does the selling and the site does nothing.

• Email or a physical letter for clinics, studios and developments, because the decision maker reads their own mail.
• Existing clients for referrals, asked once per project at handover, naming the kind of business you want next.
• Partner trades for hospitality and property: photographers, brand designers, fit-out contractors, all of whom get asked for a web recommendation monthly.

Send all forty inside one week rather than over three months. Forty in a week gives you a readable answer about whether the case page works. A slow trickle teaches you nothing, because by the time the tenth reply arrives you have already changed the page twice and cannot tell which version earned it.

Forty approaches with a real case page behind them converts better than four hundred without one, and it takes less time. Follow it with a proposal that names the design commitments and the pricing conversation becomes an order form.

The verdict on where clients come from#

They come from the page, not the channel. Build one case page properly, with the decisions written as values and the signature moment recorded, and every channel starts working at once because the thing the channel points at finally answers the buyer's question.

The order of work, which is short enough to start today:

1. One case page, eight blocks, the decisions written as values.
2. Two more in the same niche, built on spec if you do not have the clients yet.
3. Forty named approaches in one week, one specific observation each.
4. A proposal with a design commitments page inside it.

If a project has no result number yet, say so and show the decision instead. "Bookings are not measurable yet, the site went live three weeks ago, here is the type scale and here is why the case photographs are cropped at 4 / 5" is stronger than an invented percentage, and a buyer who has commissioned work before knows immediately which of the two is real.

Then hold a price that reflects what is in the file rather than what is in the calendar. The four band model is here when you get to the number.

Browse the Thanor library and build your spec page from one design's direction. $89 for three months, $189 a year, $299 once, and a free account opens the designs marked free.