Photographer Websites: Beautiful, Slow, Invisible

A thirty-image grid on a photographer's homepage is not a design decision. It's a payload. Thirty JPEGs at 2,500 pixels wide, exported at the quality setting a photographer would actually sign off on, land somewhere around 800KB to 1.5MB each. Call it 1.1MB average and the homepage is asking a phone on suburban 4G to pull roughly 33MB before it stops moving. That figure is arithmetic you can run on your own site in about a minute, not a statistic we're borrowing from anyone.

The instinct behind it is correct in every context except this one. A photographer's whole professional identity is built on not compromising the image, and the web is the one place where refusing to compromise the file produces a worse image — because the visitor closes the tab before it renders. The picture that never loads has a quality setting of zero.

Two things go wrong on photography websites, and neither of them is photography. The first is the image pipeline: the chain between the file on your hard drive and the pixels on a stranger's phone, which on most template platforms is entirely out of your hands. The second is that the pages ranking for wedding photographer Melbourne are, almost without exception, unglamorous text-heavy pages — and the best portfolio in the city can sit on page four of Google indefinitely, admired by nobody, because there's nothing on it for a search engine to read.

The bytes are only half of it

Total page weight is the number everyone reaches for, and it understates the problem.

Every image a browser downloads also has to be decoded — turned from a compressed file into raw pixels in memory. A 2,500 by 1,667 pixel JPEG decodes to roughly 16MB of uncompressed bitmap, and thirty of those is memory a three-year-old mid-range Android does not have spare. The decoding happens on the same main thread meant to be responding to the visitor's scroll. This is why heavy galleries don't just load slowly, they feel broken: the scroll stutters, the menu tap doesn't register, the lightbox opens half a second after the finger lifted.

Google's Core Web Vitals put the threshold for Largest Contentful Paint at 2.5 seconds. On a photographer's site the largest contentful element is nearly always the hero image, which means your LCP score is literally a measurement of how long your best photograph takes to appear. Most portfolio sites we audit fail that metric badly, and they fail it on mobile specifically — desktop on office wifi hides everything. We've written separately about why a site can be fine on a laptop and hopeless on a phone, and photography sites are the most extreme version of that gap we see. Google also indexes mobile-first, so the version it forms an opinion about is the version your visitors are bouncing off.

Squarespace is the default, and its image pipeline is the weak point

Most Australian photographers are on Squarespace for good reasons: presentable out of the box, templates designed by people with visual taste, and you can add a shoot without ringing a developer. None of that is the problem.

The problem is what happens at upload. Squarespace re-encodes every image you give it and serves it back from a fixed set of standard widths. Its own guidance asks for uploads in the 1,500 to 2,500 pixel range and ideally under 500KB, which tells you where the ceiling sits. You don't choose the compression quality, the format, or which crop the phone gets versus the desktop. You hand over the file and accept the output.

Three consequences follow, and they compound.

You're mostly still serving JPEG. Google's own figures for WebP put lossy WebP at 25–34% smaller than a comparable JPEG. AVIF, supported across every major browser since early 2024, typically lands smaller again at matching visual quality. On a page that is 90% imagery, format choice alone is the difference between a 33MB homepage and one under 10MB, with no visible loss. That decision isn't available to you on a template platform.

The breakpoints aren't yours. A phone at 390 pixels wide, at 3x pixel density, wants roughly a 1,170 pixel image. If the platform's nearest served width is 1,500, you've shipped about 60% more pixels than the screen can display. Multiply by thirty images.

Gallery blocks load everything. Many template gallery layouts request the whole set on page load rather than as the visitor scrolls, and lightboxes often fetch the full-size file the instant the grid renders rather than on click. That's the mechanism behind the 33MB homepage: not thirty images the visitor looked at, thirty images the visitor paid for and saw four of.

Squarespace is not a bad product. It's a product whose image pipeline is tuned for sites where images are decoration, being used by an industry where images are the entire inventory.

Where quality has to be perfect, and where nobody can tell

The useful move is to stop treating every image on the site as equally precious. They're doing different jobs, and the compression that ruins one is invisible on another.

That split is the whole technical argument. Quality where it's perceived, savings where it isn't, and a hard boundary between the site that sells the work and the system that delivers it.

Gallery architecture: fewer images, loaded later

The two structural fixes are unrelated to formats and worth more than all of them.

Cut the number of images. A homepage showing thirty frames isn't showing a visitor thirty times more than one showing eight. It's showing them you can't tell which eight are your best, which is a quiet argument against hiring you. Curation is a performance fix and a sales fix at once, and it's free — it also removes the situation where a prospective client's lasting impression is your fourth-best wedding.

A visitor who sees your eight best photographs thinks you are excellent. A visitor who sees ninety thinks you are available.

Load them when they're needed. Native lazy loading — the loading="lazy" attribute — is supported by every current browser and means images below the fold don't download until the visitor scrolls toward them. Pair it with explicit width and height attributes so the page doesn't lurch as each image arrives — that lurch is layout shift, another Core Web Vitals metric photographers routinely fail. For a long gallery, paginate or load in batches rather than dumping three hundred frames into one DOM.

Done properly, a sixteen-image portfolio page can be interactive in under two seconds on a phone and still show work at a standard the photographer signs off on. That combination isn't reachable by uploading better files to a platform that re-encodes them — it's a build decision, which is most of why we end up rebuilding these sites rather than tuning them. Our custom web design work treats the image pipeline as the first architectural decision on a portfolio project, not a post-launch optimisation pass.

Why wedding photographer Melbourne is won by boring pages

Here's the part that stings. Ranking for the searches that produce enquiries has almost nothing to do with how good your photographs are, because Google cannot see them.

Look at what occupies the top results for a competitive photography search in any Australian capital. Text-heavy pages: real price ranges, a written FAQ, pages about specific venues, posts documenting individual weddings with the venue and suburb named in the copy, an about page longer than a paragraph. Unglamorous, and they rank.

Now the typical high-end portfolio: a full-screen slideshow, a nav bar, a contact form. Perhaps two hundred words of text on the entire site, no headings with any subject matter in them, and image filenames like DSC_4471.jpg. Nothing there to match a query against. Beautiful, and to a search engine close to blank.

The fixes are unfashionable and they work.

• Write about individual shoots by name. A post about a wedding at a named Yarra Valley winery targets the exact search a couple who have just booked that venue will make. Do twenty and you have twenty entry points.
• Put words on the portfolio pages. A paragraph of context above each gallery — where, what, what the couple wanted — gives Google something to index and gives visitors a reason to slow down.
• Give the images real filenames and real alt text. Descriptive, accurate, in plain language. This is accessibility work first and search work second, and both matter.
• Claim and complete the Google Business Profile. For local intent it does more work than most of the site.

None of this is a promise about rankings — nobody credible can make one. It's a description of the surface area you currently don't have. A page with no text cannot rank for text queries, and that's the whole of the mechanism.

The enquiry page photographers refuse to build

Assume the site now loads and ranks. The last leak is the contact page, and photography is the industry most committed to making it hard.

The default enquiry form asks for a name, an email, and a message box. A visitor who has just decided they like your work now has to compose a paragraph from scratch about a wedding they may not have fully planned. Plenty don't bother. Swap the blank box for a date field, an event type, a location and one optional message field: it turns an essay into a form.

Then pricing. Photographers hide it harder than any industry we look at, and the argument — every wedding is different, I want the conversation — isn't wrong, just expensive. A visitor with no price signal assumes one of two things: you're out of reach, or you're evasive. One honest line, collections start from $4,800, loses the enquiries that were never going to convert and keeps the ones that were. Same logic applies to us: we publish that custom builds start at $8,000 AUD rather than making people ask, and the enquiries are better for it.

Finally, response time. These enquiries are shopped in parallel, so the photographer who replies within the hour is competing against three who reply in three days. And check that the form actually delivers — a form that silently fails is far more common than anyone believes.

Where to start

If you're reading this with your own site open in another tab, the order of work is roughly fixed.

Measure before guessing. Run your homepage through our free audit and look at three things: total page weight, LCP on mobile, and how much indexable text the page actually contains. Those numbers tell you whether you have an image problem, a content problem, or both. Most photography sites have both, and the image problem is the one making the content problem invisible.

Then cut images before you optimise them. Going from thirty frames to ten is a bigger performance win than any format change and costs an afternoon of honest editing. Move client delivery off the marketing site. Add written context to the portfolio pages, and start publishing named-venue shoot posts — one a fortnight, indefinitely.

If the platform is still the ceiling after all that, that's the point where a custom build starts paying for itself. The same tension turns up in adjacent visual industries: we've covered it for architecture studios and for tourism operators shipping heavy photography to regional 4G, and the resolution never changes. Keep the quality where a human can perceive it. Spend nothing anywhere else.

Custom builds with us start at $8,000 AUD. To work out whether that's your situation or whether you're two weekends of curation away from fixing it yourself, book 20 minutes or call us on 0421 933 907 — we're happy for the answer to be the free one.