The jobs that quietly eat a working week are almost never the ones anybody would name if you asked them. Here is a way to find yours in an afternoon — and to work out which of them are genuinely worth handing over.
Ask anybody what takes up their week and you will get the big things: the work itself, the meetings, the difficult bit that needed thinking about. Those are the memorable hours. They are rarely the expensive ones.
The expensive hours are the ones that leave no impression at all. Twenty minutes finding a document. Ten minutes writing an email you have written forty times before. The bit where you type something into one system that already exists in another. None of it feels like work worth remembering, which is exactly why it never gets counted — and why it can quietly take a day a week without anybody noticing.
Before anyone talks to you about AI, it is worth finding out where yours actually go. You can do this yourself, and it costs nothing.
For one ordinary week, keep a note beside you. Every time you do something you have clearly done before, write it down — two or three words is plenty. Not the interesting work. The repeats.
At the end of the week you will have a list, and it will be longer than you expected. Now put each line into one of four piles.
Those four piles cover most of what disappears. They also happen to be the four things current AI is genuinely reliable at, which is not a coincidence — they are all jobs where the answer already exists somewhere and the work is fetching, sorting or restating it.
The jobs worth handing over are the ones where the answer already exists and somebody is just moving it around.
Not all of them. A useful filter is to ask three things about each line on the list.
How often does it happen? Something that takes forty minutes once a quarter is not worth building for. Something that takes four minutes and happens twenty times a week is, and it will not feel like it, because four minutes never feels like anything.
Is the answer already written down somewhere? If the information exists — in your files, your emails, your system, a folder somebody set up years ago — then this is a fetching problem, and fetching is exactly what to hand over. If the answer only exists in somebody’s head and depends on judgement, leave it there.
What happens if it is wrong? If a mistake is obvious and easily fixed, it is a good candidate. If a mistake would be quiet and expensive, it can still be a good candidate — but built so a person checks it before it counts, which is how we build everything anyway.
Run the list through those three questions and it usually collapses to two or three jobs that are worth doing something about. That is the right number. Businesses that try to change eight things at once tend to end up changing none of them.
A fair bit, and we would rather say so.
Some of the chasing pile is a calendar reminder and a saved email template. Some of the re-typing pile disappears the moment somebody turns on a feature in software you already pay for — and that happens far more often than people expect, because most business software has grown a great deal in the last two years and almost nobody has gone back to look.
What is left after that — the jobs that repeat constantly, draw on your own documents, and need to come out right every time — is where building something actually pays for itself. That is the part worth a conversation.
Once you have your two or three, the arithmetic is simple enough to do on the back of an envelope. How many times a week, how many minutes each time, multiplied out over a year. You do not need a consultant for that, and you should be suspicious of anybody who makes it sound complicated.
What is harder to judge on your own is the other half: what each one would actually take to build, and whether it would hold up on a bad week rather than a demonstration. That is the half we put a number against during an audit, so the two sit side by side and the decision makes itself.
Once you have your list, what to automate first and which AI tool actually fits the job pick up from here — both are part of what we do.
Every business is different and none of this is a substitute for looking at yours properly. If you want a hand doing that, the audit is exactly this exercise done thoroughly, with a number against each answer.
Twenty minutes on the phone, free, no obligation. Bring your list and we will tell you honestly which lines are worth doing something about — and which ones to leave exactly as they are.