Job Photo Galleries That Sell The Work Instead Of Filling Space

Every trade business has thousands of job photos on phones, and almost none of them are on the website. The ones that make it are usually dumped into a gallery page with no captions, no order and no explanation, where they sit as decoration rather than evidence. That is a waste of the single most persuasive material a trade business owns.

Why a gallery page on its own does not work

The standard arrangement on a trade website is a navigation item called Gallery, leading to a grid of images with no text. It is a reasonable-looking page and it converts almost nobody, for a simple reason: it is separated from the decision.

Somebody reading your bathroom remodel page is, at that moment, deciding whether you can do their bathroom. Proof at that moment is worth a great deal. The same proof two clicks away on a page they may never open is worth very little, because the visitor has to interrupt their own train of thought to go looking for it.

So the primary home for job photos is the service pages. Three to six jobs on each, chosen because they are the jobs that page describes. The combined gallery page can still exist as a secondary destination, for the visitor who wants to browse and for the crew lead who wants a link to send somebody, but it is not where the work happens.

The caption is the product

This is the part that separates a gallery that sells from a folder of images, and it takes under a minute per job.

A caption needs to carry three things:

1. What the situation was. Not “before”, which the photo already says. The actual condition: what had failed, what was wrong, what the customer had been living with.
2. What made it awkward. The constraint is where the credibility lives. Access through a finished room, a shut-off that was seized, a slab, a tenant who could not be without water overnight, a deadline.
3. What was done and how long it took. The fix, in plain words, and the duration.

Here is the difference in practice. A photo captioned “Bathroom remodel” tells a reader nothing they could not see. The same photo captioned “1950s bathroom with a cast iron tub over a joist that had rotted. We rebuilt the floor before anything else went in, which added a day, and the customer had a working shower again on day four” tells a reader that you find problems before they become disasters and that you will tell them the truth about timelines. That is the whole sale.

Nobody needs to write these at a desk. The person who did the job can say all three sentences into a phone in thirty seconds while standing in front of the finished work.

What to photograph

Consistency matters more than equipment.

Shoot the before, always. The universal complaint from every trade is that nobody remembers the before photo until the job is finished. Make it the first thing that happens after the tools come out of the van. It takes ten seconds and it doubles the value of the after.

Same position, same angle. A before and after pair only reads as a pair if the camera was in roughly the same place. Stand in the doorway. Stand at the end of the yard. Pick a spot and use it twice.

Tidy the frame, not the job. Move the tool bag, the coffee cup and the dust sheet that are not part of the story. Do not stage the work itself; the honest version is what is persuasive.

Get the detail shot. One wide shot to show the scale and one close shot of the part that took the skill. The close shot is the one that convinces somebody in the trade, and a surprising number of your readers are related to somebody in the trade.

Light. Turn the lights on. Open the blinds. If it is outside, avoid pointing the camera at the sun. That is the entire photography course.

What not to photograph

Anything that identifies where somebody lives: house numbers, street signs, mail, vehicle plates, the view from the window that a neighbour would recognise. Frame tightly on the work.

People, unless they are your own crew and they have agreed. A customer in the background of a photo on a business website is a problem waiting to happen.

Permission, kept properly

Ask, and keep a record of the answer.

The ask is easier than most operators expect when it is framed correctly: you are not asking to photograph somebody’s home, you are asking to show the work you did. Tightly framed photos of a panel, a heater, a patio or a finished bathroom rarely feel intrusive.

Keep the answer somewhere retrievable. A yes recorded in a job record is worth having two years later when somebody asks where a photo came from. Where you have no record, do not publish.

For commercial and property work, permission may sit with a manager rather than an occupier, and some sites cannot be shown at all. Ask early, because a bid packet full of photos you cannot use is a wasted afternoon.

Organising the gallery

Two organising principles beat everything else: by job type, and by recency within that.

By job type because that is how the visitor is thinking. A person planning a patio wants patios. A property manager assessing whether you can handle a nightly contract wants commercial buildings. A grid of everything in date order forces every visitor to do the filtering themselves, and most will not.

By recency inside each type because a gallery of work from six years ago reads as a business that has slowed down, even when it has not.

Avoid two organising schemes that look sensible and are not. Organising by town produces a set of thin, near-identical collections. Organising by crew or technician is interesting internally and meaningless to a customer.

Before and after, done honestly

The before and after pair is the most persuasive format in trade marketing and the easiest to discredit.

Rules that keep it honest:

• Same angle, same framing, same distance.
• No filters or colour changes on one that are not on the other.
• Lighting as close as circumstances allow. A dark before and a bright after with the blinds opened is a real effect and a slightly dishonest one.
• Do not stage the before by making it worse.

A reader who suspects the pair has been manipulated does not just discount that image; they discount the whole site. The honest version is persuasive enough, and it has the advantage of being true.

An interactive slider comparing the two is a nice piece of interface when it is done well, but it should never be the only way to see both images. Some people will not use it, and some assistive technology will not expose it. Show both images and let the slider be an enhancement.

Making images work technically

Photos are the heaviest thing on most trade websites, and a gallery is where a fast site usually goes to die. Three things matter.

Serve the right size. A gallery grid needs a thumbnail, not a twelve megapixel original resized in the browser. Responsive image techniques exist precisely for this, and MDN’s guide to responsive images covers the mechanics. Delivering appropriately sized images is also the single biggest lever on page weight, which is why image delivery services exist and why web.dev’s material on image CDNs treats it as a first-order optimisation.

Load them lazily, below the fold. Images further down the page should not compete with the page’s first paint. This is a one-attribute change and it matters most on the pages with the most photos.

Write real alternative text. Not “bathroom”, and not the file name. Describe what is shown, because that text is what a blind visitor gets and it is one of the signals image search uses. The W3C’s images tutorial is the reference for how to write it well, and Google’s image guidance covers the search side. In practice, the caption you already wrote is usually most of the way there.

Do not put text in the image. A price, an offer or a caption baked into a JPEG cannot be read by a screen reader, cannot be translated, cannot be searched and cannot be edited without going back to the original file.

Worked examples, by trade

The shape of a good gallery entry is the same everywhere. What changes is what counts as the interesting part.

Plumbing. The interesting part is nearly always what was behind the wall. A finished bathroom looks like every other finished bathroom; the photo of the corroded run that came out of it is the one that makes a reader trust you. Pair the close shot of the old pipe with the wide shot of the reinstated room, and say in the caption what you found that was not in the original quote and how you handled it.

Electrical. The interesting part is order. A photograph of a tidy panel next to the photograph of the one it replaced does more work than any amount of copy about standards. Add the constraint: whether the utility had to be involved, whether the house stayed live, how long the power was off.

Heating and cooling. The interesting part is the install detail and the site conditions. A cramped loft, an awkward condensate run, a unit that had to be craned in. Replacement buyers are worried about mess and disruption more than about equipment, so a caption that says what the house was like during the work answers the question they are holding.

Landscaping. The interesting part is time. Landscaping is the one trade where an after photo taken the day the job finished undersells the work, because nothing has grown in yet. Go back. A photo from the following season is the most valuable image a landscaping business can own, and almost nobody collects it.

Cleaning. The interesting part is the specific surface, not the room. A whole clean kitchen is unconvincing because it is hard to read. One oven door, one grout line, one carpet with a boundary halfway across it is legible at a glance and impossible to argue with.

Handyman and remodeling. The interesting part is the list. Punch list work is many small jobs, so one gallery entry can be a set of four small photos with a line each, which also communicates the thing that page is selling: that a lot of small things got done in one visit.

Video, and when it earns its place

A short clip does one thing a photograph cannot, which is show a process. Water draining after a line is cleared. A system starting up. A door that used to stick, closing.

That is a narrow use and it is worth doing when it applies. What is not worth doing is a two-minute walkthrough with commentary, which almost nobody watches to the end, or an autoplaying background video, which costs a great deal of page weight to say nothing.

If you use video, keep it under fifteen seconds, let the visitor start it, and never let it be the only place important information appears. And treat it like any other heavy asset: it should not be loading while the rest of the page is still trying to paint.

How many is enough

Fewer than you think, chosen better.

A visitor does not browse a gallery the way they browse a photo album. They check it. They want enough evidence to believe you have done this before, and then they want to get on with deciding. Three to six well-captioned jobs on a service page does that. Twelve to twenty across a combined gallery page is plenty for most trades.

The instinct to upload everything comes from a good place, which is that each job took real work and deserves to be seen. Resist it anyway. Two hundred unlabelled images say less than twelve labelled ones, because the twelve are readable and the two hundred are a wall.

Keeping it alive

A gallery is only impressive if it is current. The failure pattern is always the same: a burst of photos at launch, then nothing for two years.

What fixes it is making the handover trivial. Whoever runs the site should be able to receive photos and a voice note and turn them into a gallery entry without a meeting. On a managed plan that is a revision request. On a self-managed site it needs to be a form somebody can fill in from a phone in a van.

Set a rhythm rather than a target. One job a month, properly captioned, is twelve a year and a gallery that always looks like a business currently doing work. A push to add fifty at once will not happen twice.

The gallery a commercial buyer needs

Residential proof and commercial proof are not the same thing, and a single gallery rarely serves both.

A homeowner is looking for reassurance that the finished result will look right. A facilities manager or a general contractor is looking for evidence that you can work inside constraints: after hours, around occupants, to a programme, with the paperwork done. The photographs that prove that are different photographs.

For commercial work, show scale and context rather than finish. A corridor mid-strip, a plant room, a rooftop unit, a site set up with barriers and signage. Caption them with the things that buyer is assessing: how many people were on site, what hours the work ran, how the building stayed operational, whether the programme was met.

That material is also what makes a bid packet, so the work of assembling it is not extra. It is the same work, published once.

Getting the material off the phones

The last obstacle is logistics, and it defeats more galleries than any design decision.

Photos sit on crew phones because there is no obvious place to send them. An email address that somebody has to remember is not an obvious place. A shared folder that requires an app login is not an obvious place. Whatever the route is, it has to be one tap from the camera roll and it has to accept a voice note alongside the picture, because the caption is the part that matters and nobody is going to type it in a van.

Decide the route, write it on a card in every vehicle, and ask for it once a week for a month until it is habit. After that it runs itself, and the gallery stops being a project and becomes a by-product of doing the work.

What this is worth

Job photos are the only marketing asset a trade business produces every single day at no additional cost, and the only one that a competitor cannot copy. A national brand can outspend a local operator on advertising. It cannot show the panel it upgraded on your street last Tuesday.

The gap between a business that quietly accumulates that material and one that publishes it properly is not a matter of budget. It is a two-minute routine at the end of a job and somebody on the other end who turns it into a page.