Getting job records off paper and into something you can search

Paper job records work perfectly right up until the moment you need one. Then you are standing in front of a drawer trying to remember the customer's surname, roughly which month it was, and whether the sheet ever made it back off the van.

This is not an argument for going paperless on principle. Paper is quick to fill in, it works with wet hands and no signal, and no clipboard has ever run out of battery. It is an argument for the record ending up somewhere you can search, because almost all the value of a job record is in being able to find it later.

The cost is not the filing. It is the handful of things that quietly become impossible.

The usual mistake is designing a job record that captures everything, because thoroughness feels responsible. What you get is a form nobody completes properly, and a system full of half filled records is worse than paper. You trust it, and then it lets you down in front of a customer.

Work backwards instead. Write down the questions you actually get asked, in the words people ask them. What did we fit? When were we last there? Who signed for it? What did it look like before we started? Then make sure the record answers exactly those, and nothing else, to begin with.

In our experience five fields plus photographs covers most of it: customer, address, date, who attended, and what was done and what was used. Anything more can wait until you catch yourself wanting it, and then you will know it is real rather than theoretical.

Photographs are the most valuable thing in a job record and the thing most often lost. They get taken on somebody's personal phone, they sit in a camera roll with four thousand other pictures, and when that person leaves, so do they.

The rule that matters is simple. A photo has to be attached to the job at the moment it is taken, not sorted out later. Later does not happen. That means the app the photo is taken in is the app the job lives in, or the photos go up against the job before the van leaves the street.

Two habits are worth insisting on. Take a before photo of anything you might later be blamed for, every time, including things that look completely fine. And photograph the label: the model number, the serial, the meter reading, the settings. A picture of a rating plate has saved more return visits than any amount of note taking.

Think about where those photos live, particularly when they include people, vehicles or the inside of somebody's home. A business account with access you control is a very different thing, legally and practically, from thirty personal phones.

Most job notes are written for the person who wrote them. They are shorthand, they assume context, and they make sense for about a week. Then the person leaves and you discover the notes were never really records at all.

The test is blunt: could somebody who has never been to this site read the note and do the next visit? If not, it is a memory aid, not a record.

A few things make notes survive their author.

You almost certainly should not scan all of it. Scanning years of historical paperwork is expensive, tedious, and most of it will never be looked at again by anyone.

The approach that tends to work is a line in the sand. From a date you choose, everything new goes in digitally. The old paper stays where it is, and a folder gets scanned only when you have had to open it anyway. Within a year or two the drawer you actually reach for is empty, and the rest is dead weight you can box and archive.

There are exceptions worth pulling out on purpose: anything with a warranty still running, anything on a contract that renews, certificates and test records you are required to keep, and anything tied to a live dispute. Those are worth an afternoon. The rest is not.

Check what you are legally required to keep, and for how long, before anything goes in the shredder. Records tied to tax, to health and safety, or to a regulated certification carry their own retention periods, and your accountant or your trade body will know them better than we do.

Putting scans into a folder on a shared drive is storage, not searching. If the file is called scan0047.pdf, nothing has been solved. The filing cabinet has just been moved somewhere with a worse index.

Two things make a record findable. The first is that it is attached to a customer and a site, so you can start from the address rather than from a date nobody remembers. The second is that the text inside it can be searched, which for scans means the scanner or the system runs text recognition rather than saving a flat image.

The practical test is the one that counts. Can somebody who is not you find last March's visit to a given address, on a phone, in under a minute, while a customer is talking? If not, there is still work to do.

Pick one job type, the one you do most of. Put its record into whatever you already have, even a shared folder with a strict naming rule, and run it for a month. You will learn what people actually fill in when they are cold, wet and in a hurry, which is never quite what they said they would.

Then widen it. If the sticking point turns out to be that the thing has to work on a phone with no signal in a basement, that is much better discovered before you buy anything than after.

One more thing worth saying plainly. Moving records out of an old system, or off paper, is its own job and it is usually underestimated. Decide early what you are bringing across and what you are deliberately leaving behind, because trying to carry everything is how these projects stall.

If you want help deciding whether the systems you already pay for can do this, or whether you genuinely need something built around your jobs, tell us what a job record looks like in your business today and we will tell you honestly.