Where Your 10+ Hours Are Going (And How to Get Them Back)
A practical way to audit your own week and find the repetitive work worth handing off — including the tasks that look expensive but are not, and the ones that look small but are not.
It is Sunday evening and you are catching up on admin. Again. Not because the week was unusually bad, but because this is what the week is now.
You did not start your business to spend it moving information between things that should already know about each other. But when we ask owners where their hours go, almost nobody can answer with any precision. They know it is "admin". They cannot tell you which admin, or how much.
That imprecision is the actual problem. You cannot hand off work you cannot describe, and you cannot decide whether fixing something is worth £800 if you do not know it is costing you eleven hours a month.
So here is how to find out. It takes about a week of very light effort and it is genuinely worth doing whether or not you ever hire anyone to fix it.
The two-column exercise
For one week, keep a note — paper is fine — with two columns.
Left column: things you did more than once. Not projects. Not client work. The small repeated actions. "Copied enquiry into CRM." "Chased Hendricks invoice." "Confirmed Thursday appointment." One line each, every time it happens.
Right column: roughly how long it took. Round to the nearest minute. Precision does not matter; consistency does.
Most people find this tedious for two days and then start finding it interesting, because the same six lines keep reappearing and they had genuinely not noticed.
At the end of the week, group identical lines together, multiply by frequency, and annualise. That is your list, in order of what it actually costs you.
What the list usually looks like
We have now seen a lot of these. The pattern is remarkably consistent.
The tasks that look small and are not
Re-entering the same information. A customer's details typed into the booking system, then the invoice, then the spreadsheet. Two minutes each time, forty times a month. That is sixteen hours a year on typing that a connector would do in the background — and every re-entry is a chance to introduce a typo you will find out about later, expensively.
Confirming and reminding. Individually trivial. In aggregate, one of the largest line items on most lists, and one of the easiest to hand over completely.
Checking whether something happened. Opening a tab to see if a payment landed, if a form came in, if a job was marked done. This one barely registers as work, which is why it never gets fixed, and it is often the single most frequent line in the whole list.
The tasks that look big and often are not worth automating
The monthly report. It feels like a burden because it arrives all at once, but three hours twelve times a year is thirty-six hours — less than the re-entry above, and usually much harder to automate well. Worth doing eventually. Rarely worth doing first.
The annual anything. If it happens once or twice a year, automating it will cost more than it saves and the automation will have rotted by the time you next need it.
The judgement calls. Deciding which quote to send, how to handle an unhappy customer, whether to take a job. If it needs you to know the person, it needs you.
The number that decides it
Once you have the list, the calculation is simple and you should do it yourself rather than let anyone do it for you.
Take the total hours per year for the tasks you would genuinely hand over. Multiply by what an hour of your time is worth — and be honest, which usually means "what you would pay someone competent to do this", not your billable rate.
That annual figure is the number worth carrying into any conversation about fixing it. For the sort of work we do, automation projects start at £800 and go up with how many processes you are connecting; most clients are live in two to four weeks.
If your list comes to sixty hours a year, the maths is marginal on almost any quote and you should probably wait. If it comes to three hundred, it stopped being marginal a while ago. Either way you now know, which is the point — and you will be able to tell instantly whether a quote is worth taking seriously.
Why "just use Zapier" is sometimes the right answer
We are not going to pretend otherwise: if your list has three items on it and they are all two-step connections between mainstream tools, you can build that yourself in an afternoon and you should.
Where it stops being a DIY job is when the connections have logic in them — when the follow-up depends on what the customer did, when the invoice depends on which service was delivered, when something needs to happen only if something else did not. That is the point at which people either give up or accidentally become their business's unpaid systems administrator.
The honest dividing line is not complexity for its own sake. It is whether maintaining it will end up on your list next year.
What good looks like afterwards
The thing owners describe most often is not the time. It is that the low-level background hum of "have I forgotten something" gets quieter.
Concretely, after this kind of work you should expect: enquiries answered while the person is still on your website rather than four hours later; no manual re-entry of anything a system already knows; invoices chased on a schedule you set once; and a small number of alerts telling you when something needs a human, instead of you checking whether anything does.
Most of our clients get back somewhere between five and twenty hours a week. That range is wide because it depends entirely on how much of this work you are doing today — which is exactly why the two-column exercise comes first.
Where to start
Do the week. Write the list. Rank it by annual cost.
Then take the top item and ask whether you would still be doing it by hand in two years. If the answer is obviously no, that is your starting point.
If you want a second opinion on the list — including which items we would tell you not to bother with — book a free 30-minute call. We don't guess. We diagnose, and we will say when the honest answer is that you do not need us yet.
More on how we approach workflow automation, or read why this matters more in 2026.