Portfolio sites for design-led studios: sell the work, not the process

If you design gardens, houses, extensions or interiors, or photograph them for the people who do, your website has one job and it is not explaining how you work. It is to make somebody want the specific thing you made, and then tell them enough to work out whether they can afford it. Most studio sites do neither. Ninety photographs get squeezed into a four-across thumbnail grid, cropped square by a gallery plugin, and underneath sits four hundred words about a collaborative, human-centred, iterative process.

Nobody has ever hired a garden designer because of their process. They hire because they saw a walled garden with a rusted steel rill running down the middle of it and thought that, in my garden. Everything worth doing to a design studio's website follows from that: get the reader to that moment as fast as the connection allows, then answer the two questions they immediately have, which are "would they take my project" and "roughly what would this cost me".

Image handling is not a detail. It is the entire job.

You have probably paid a photographer somewhere between £700 and £1,800 for a day's shoot. If the images then land in a plugin that crops everything square at 240 pixels, you have thrown that money away and made your work look like everyone else's.

Two things go wrong, and they are separate problems. The first is crop. One aspect ratio for every image is the default in almost every off-the-shelf gallery, and it is wrong for this kind of work. A wide establishing shot of a garden or a room needs to stay wide, squaring it removes both ends, which is where the walls are, which is the whole point. A detail shot of a joint or a plant combination wants to be tall on a phone. The fix is art direction: two or three crops per photograph, chosen deliberately, served with <picture> and <source media> so the phone gets the portrait crop and the desktop gets the landscape one.

The second is weight. A Lightroom export at full quality is commonly 5 to 8 MB. On a typical UK 4G connection that single hero photograph takes eight to twelve seconds to arrive, and roughly a quarter of people will have gone before it does. You lose them before the work has appeared. That is the most expensive failure on this list because it happens to visitors who were already interested.

Those numbers hold up on a laptop screen. AVIF at quality 50 is visually indistinguishable from the original at these sizes and lands about 30% below WebP; keep WebP as the fallback and you are covered everywhere. Squoosh does it free in a browser tab for a handful of images. For eighteen gardens at fifteen images each, use sharp from a script or let Cloudinary do the derivatives on upload.

Then get lazy-loading right, because half the studios who do it break their own page. Never lazy-load the first image on a project page. It is the largest element in view and delaying it is the one thing you are measured on. Everything below the fold gets loading="lazy", and every single image gets width and height attributes so the browser reserves the space before the file lands. Without them the page reflows fifteen times as photographs arrive and the reader loses their place. I have written the full version of this in the website speed guide.

One project, one page, one story

A grid proves the work exists. A page makes someone want that specific project. The difference in behaviour is not marginal. Individual project pages hold attention for minutes; nobody spends four minutes in a thumbnail grid.

The structure that works is the same one you use verbally when a friend asks about a job. The site: where it is, what was there before, who lives there. The constraint: the north-facing slope, the Article 4 direction, the ceiling you could not move, the £180k budget. The decision: what you chose to do about it and what you rejected. The outcome: how it works now, and how it has settled after two summers.

Twelve to eighteen images in a deliberate order, captioned with real materials and plant names, will out-sell ninety images in a grid every time.

Caption properly. "Corten planters, Danish oak bench, Hakonechloa macra along the shaded edge" tells a certain kind of reader that you know exactly what you are doing, and that reader is the one with the budget. Vague captions read as someone hiding a lack of specificity.

The unprofitable bit: if you have three finished projects, build three excellent pages, not fourteen thin ones. A studio two years old with three strong stories beats a fifteen-year-old practice with a scrolling wall of undifferentiated thumbnails, and I would rather you spend the budget on photography than on pages you have to pad.

Qualify, don't close

A website cannot sell a £40,000 garden or a £28,000 architectural fee. It can decide who picks up the phone, and that is worth more.

Start by splitting the service properly. Design-only and design-and-build are different products with different buyers, and describing both in one paragraph means nobody knows what they are enquiring about. Separate pages, separate explanations of what happens in what order, separate fee structures. Done well, that single change splits incoming enquiries between build and design-only before they reach you, which removes an entire round of email from the front of every job.

Then publish a starting budget. This is the change studios resist hardest and it is the one that does the most. Not a price list, a floor and a reason. "Complete garden design and build projects typically start at £45,000." "Our full architectural service is charged at 9–12% of build cost, with most residential projects landing between £22,000 and £45,000 in fees." "Full-house interiors from £900 per room in design fees, excluding specification." Anyone whose budget is £6,000 self-selects out, and they were never going to become a client. They were going to become three emails and a polite decline.

Two honest caveats. If your diary is empty, do not publish a floor price yet, take the small jobs, build the portfolio, add the number when you can afford to turn work away. And if a lot of your work arrives through architects, developers or interior designers referring you, a public price can complicate their conversations. Ask two of them before you post it.

Sell the aftercare as a product, not a footnote

Most design studios have a second service they mention at handover and then forget: seasonal maintenance, a styling refresh, defects and snagging support, licence renewals on a set of images. It gets a sentence in the final email and generates almost nothing.

Give it a page. A newly planted garden looks its worst in year two if nobody prunes it, which is a genuinely good reason for a client to keep paying you, so say that plainly. Put the visit schedule on the page by season, put a price on it, and, this is the part that matters, include it in the design proposal rather than after the invoice. Aim to put most finished gardens on a seasonal retainer, which turns a lumpy project business into something with a predictable monthly base.

The same shape works elsewhere. Architects sell a build-stage inspection package. Interior designers sell a seasonal restyle plus trade-price sourcing. Photographers sell an annual refresh shoot to studios whose portfolios go stale. The product is not new; the page is.

Search: the town term earns, the specific term compounds

For a design studio there are really two search strategies, and only one of them pays this year.

The money term is your discipline plus a place. "Garden designer Tunbridge Wells." "Architect Lewes." "Interior designer Sevenoaks." Volumes are small, often 20 to 150 searches a month, but the intent is absolute, and one converted enquiry can be worth £25,000. Build a real page for each town you genuinely work in: projects you have completed there, the local planning quirks, the soil, the housing stock. Not a template with the place name swapped. Google has been discounting those since long before it got good at reading. The local SEO checklist covers the mechanics; the strategy is on the SEO service page.

The second angle is slower and better. Write about the thing you know that a generalist cannot fake. For a planting designer that is chalk downland, heavy Weald clay, dry north-facing walls, coastal wind. For an architect it is conservation areas, listed building consent, Article 4 directions, what actually gets approved in a specific district. These posts take four to nine months to rank and bring a small, patient trickle of people with precisely the right problem and precisely your kind of budget.

The unprofitable version: if you work nationally on £500k-plus projects and your enquiries come from referrals and publication, local SEO is close to irrelevant to you. Spend the money on photography and on getting published instead. I would rather say that than sell you twelve town pages you do not need.

Restraint in motion

Design-led sites attract design-led animation, and it is nearly always too much. A parallax hero, a cursor follower, a scroll-jacked section, a text scramble, a page-transition wipe and a hover tilt on the tiles, each defensible alone, collectively a site that fights the photography and makes a mid-range Android stutter.

Pick one gesture and do it properly. On these builds it is usually a soft image reveal: a 500 ms fade and a two-percent scale as each photograph enters the viewport, and nothing else moves. It reads as confidence. Wire it to prefers-reduced-motion so anyone who has asked their device for stillness gets stillness. Your work is the effect. Anything competing with it is noise.

The About page does more work than you think

For a £30,000 project the client is not buying a service, they are choosing somebody to be in their life for twelve to eighteen months, in their garden or their house, spending their money. They will read your About page more carefully than anything else on the site, and most About pages are unreadable, a team-of-passionate-creatives paragraph written to sound like a bigger practice.

Write it as a person. One good photograph of you, taken by the photographer who shot the work. Where you trained, how long you have done this, how many projects, what you personally do on a job and who does the rest. What you are not good at is worth a sentence: "we do not do swimming pools" saves everybody a month. If your studio is you and a part-time assistant, say so, clients at this level are often actively choosing the small practice, and pretending otherwise loses you the ones you want. The same honesty should run through the identity and tone of voice, or the site and the first meeting will feel like two different companies.

None of the above is expensive. It is a fortnight of decisions about crops, captions, structure and numbers, applied to work you have already done. If you want a straight opinion on where your current site is losing people, send me the URL and the three projects you are proudest of. The process page sets out how a build runs from first call to launch, or start a conversation here.