Site documentation: what to photograph before the structure is covered

A finished building is a black box

You cannot see from a finished building how it was built. Once the board is up and the tiles are grouted, what lies behind exists only in the form it was photographed. That is why, in construction, documentation is a cost every day and a return rarely, but all at once: on the day someone asks what is inside the structure, whether the damage was there before the work, or who asked for the change.

This article covers what to photograph, how a photo becomes evidence, whose material it is, and how to start documenting if nothing has been done so far.

Four situations to always photograph

It is not worth photographing everything, because material you cannot find anything in is as useless as no material at all. Four situations cover most of the need.

Before covering. Everything that stays inside the structure: pipes, cables, insulation, fixings, waterproofing and penetrations. This is the most important category, because the information cannot be recovered later except by opening the structure.

The starting condition. The state of the site before you began, including adjacent rooms and the neighbour's walls if you are demolishing. This protects you against claims that damage was caused by your work.

Deviations. Surprises that were not in the plans: rotten structures, mould, faulty earlier work. They are the basis for extra work, and that basis is needed as a photo before the problem is fixed.

Finished work. Especially spots that will soon be out of sight, and spots the client is likely to ask about.

In wet rooms, an inspection record is usually completed for the waterproofing, and photos are its best attachment. A photo does not prove membrane thickness or adhesion, but it shows whether corner reinforcements, the drain connection and penetration seals were done, and it narrows down where the structure needs to be opened later if necessary.

A photo becomes evidence when it knows where and when

A photo with no link to the site is close to worthless. The timestamp is usually automatic, but the location is missing. A photo of the inside of a wall in a phone's gallery does not say which site, which floor or which wall it is, and three months later nobody remembers.

So what matters is whether the photo is created in the project or in the device's camera roll. A photo attached to the project and placed on the floor plan answers "where is this" by itself. A photo left in the gallery disappears when the device changes.

Three practical guidelines on quantity and quality:

• An overview and a close-up. Two or three photos are usually enough for a structure about to be covered: an overview that shows where you are, and a close-up that shows the detail. One overview of a room proves nothing about a single structure, and two hundred unsorted photos are in practice the same as none.
• Include a measure. A tape measure in the photo shows a scale you otherwise cannot see, such as insulation thickness or floor fall.
• Same angle before and after. A pair of photos from the same angle is stronger evidence than two good photos from different directions.

A phone camera is enough. More important than resolution is that the photo is linked to the site and the location.

Three situations where a photo decides

Water damage two years after handover. The client reports that the bathroom wall is damp. The question is whether the waterproofing was done correctly. Without photos the structure is opened by guesswork, and the cost of the investigation and the liability are argued over without evidence. Photos of the waterproofing with corner reinforcements and the drain connection before tiling show what was done, and often also where the fault is more likely to be.

A neighbour's damage claim. After demolition, the neighbour reports a crack in their wall. Photos of the starting condition show whether the crack was already there. Without them the matter is handled with the insurer without evidence either way.

Extra work the client does not remember ordering. Rotten floor structure is found under the demolition, it is repaired and the invoice goes out. A dated photo of the rotten structure linked to the site is the basis. A verbal description is a claim. Extra work should still always be agreed in writing before it is done, and the photo is an attachment to that agreement. More on this in the article on managing extra and variation work.

In all three, the decisive moment is far from the moment the photo should have been taken. It cannot be done afterwards, which is why photographing has to be routine rather than a judgement call.

Who takes the photos

The most common model is that the site manager takes photos on their visits. The material is then thorough on the days they visit and empty on the others. Structures being covered do not wait for the site manager. They are covered when the installer is done.

A model that works better is that the person doing the work takes the photo. That needs two things: photographing has to be quick, and the worker has to know which situations to photograph. The second is the four-point list above, not a manual.

Whose material it is

The question usually comes up at the wrong moment: when someone leaves the company, a phone breaks, or photos are needed two years after handover when the crew has moved on to other sites.

Material created at work belongs to the employer, but if it exists only on a worker's phone, it is gone in practice the moment the phone or the worker is. The photo has to be created in the company's material at the moment it is taken. Moving it afterwards usually does not happen.

Data protection is part of this too. Photos from customers' homes can contain personal data, and the company is responsible for handling it. The company should be able to say where the photos are and who can access them, which is hard if they are on the phones of ten current and three former employees or in a WhatsApp group. Why a group chat is not an archive is covered in the article on site communication.

Retention and long liability

The warranty period in construction contracts is typically two years, but liability for hidden defects lasts considerably longer. A photo of a structure may therefore be needed ten years later, and then what matters is not whether the photo was taken but whether it can be found.

Three questions you should be able to answer about your own material: where are the photos when the phone has been replaced twice? Who can access them when the site manager has left? Can a single photo be found without someone remembering when it was taken? If any of these stays open, the material exists but cannot be used.

Handover documents are not the same as your own documentation

The two are easily mixed up, and then one of them does not get done.

Handover documents go to the client: operating and maintenance instructions, inspection records, warranty certificates and the necessary drawings. Their content is agreed in the contract, and they are part of completing the contract. Missing handover documents can delay acceptance and the payment instalment with it.

Your own documentation stays with the contractor: photos of covered structures, the site diary and deviations. Its job is to exist when you need to show something.

Both are worth collecting as they arise. Putting the handover documents together in the handover week is the most common reason acceptance of a finished site drags on.

Where to start if there is nothing

If no documentation has been done at all, the whole thing looks too big. A three-week path works for most construction companies:

1. Week 1: one site, one rule. A photo of every structure before it is covered. Nothing else. At the end of the week, look at what built up.
2. Week 2: add the starting condition. A photo of every new site before anything is done. One visit per site, and the most protection for the effort.
3. Week 3: add deviations. A photo and one sentence whenever something unexpected turns up.

After three weeks the routine covers the situations where documentation is actually needed. If you have to pick one, it is the first: a structure being covered is the only situation where the information disappears completely.

How this works in FiSAS

In FiSAS, photographing is built around exactly the questions a camera roll cannot answer:

• The photo is saved straight to the project. A worker takes the photo on their phone, and it is immediately under the project for the whole team to see. A new phone or a worker leaving does not take it away.
• Location on the floor plan. Defects, deviations and structures being covered are marked on the floor plan with an exact location and backed up with photos. Nobody has to remember which wall it was.
• Owner and status. Every marking has a person responsible and a status, so an observation that needs fixing does not get left hanging.
• Videos and attachments. Videos, checklists and other attachments are stored in the same place.
• Site diary. Entries can be typed or dictated, and AI turns speech into text and suggests a title and category. Weather data is saved on each entry automatically based on the site's location, and the diary is available as a PDF report.

Documentation is included in every plan, including the single-user Maker plan.

Test your own situation

Pick a site you handed over a year ago and look for a photo of any structure before it was covered. If you find it within a minute, your routine works. If you cannot find it at all, you know what it would cost if the question came up.

FiSAS can be tried free for 14 days. You can photograph the next site's covered structures straight into the project and onto the floor plan, and after two weeks you will see whether a photo is quicker to find than in the camera roll.