Workings · 04 · the fee
What £1,200 actually buys, and what it doesn't.
Nobody likes asking what's in a fixed price. So here it is written down, including the parts that aren't in it and the case for not buying it at all.
The audit is the second step and the first one you pay for. It's £1,200, it takes two weeks, and the fee comes off the first build if you go ahead. That's the whole commercial shape of it. Every price on this page excludes VAT.
What follows is the sequence. It doesn't vary much, because the point of a fixed price is that the scope is tight enough to know how long it takes — and because I've run the same process on my own business, repeatedly, which is where all the evidence on this site comes from.
-
Half a day watching the real thing
Over screen-share, with whoever actually does the job — not the manager's description of the job. I want to see the spreadsheet with the tab nobody talks about, and the bit where someone copies a figure across because the two systems have never spoken.
-
Every repetitive job costed
Hours a week, times a loaded hourly rate, times a working year. The same arithmetic as the ledger on the front page, run on your numbers instead of illustrative ones. Most of the value is here, and most of it is uncomfortable.
-
The list ranked, worst first
Not by how interesting it is to build. By what it costs you now, against what it would take to fix. Some expensive jobs are hard to automate and some cheap ones are trivial — the ranking is the honest version of that trade, not the flattering one.
-
A build plan with a price against each line
So you can buy one item, three, or none. No bundle, no programme, no phase two you have to commit to now. If you take the plan to someone cheaper, it's specified well enough for them to quote against.
-
One quick win built and live
Working, in your business, before the two weeks are up. This is the part that separates an audit from a report. You should be able to watch something happen on a Tuesday morning that nobody had to do.
What it doesn't include
Worth being blunt about, because the gap between what people assume and what's written down is where every bad agency relationship starts.
- It isn't the build. One quick win goes live; the rest is priced, not built. A full build starts at £3,500 and takes three to four weeks.
- It isn't a data cleanup. I work around messy data rather than billing you to tidy it first. If your data genuinely blocks everything, I'll say so — that's a different job and probably not mine.
- It isn't new software. I don't resell anything and I take no commission from any vendor. Where the answer is a tool you already pay for, that's the answer.
- It isn't a strategy document. There's no maturity model, no roadmap with quarters on it, and no slide with four boxes on it. If that's what you need, I'm the wrong person and I'd rather say so now.
If the audit turns up nothing worth building, I'll tell you on the day. You'll have paid £1,200 to find out that your admin is roughly as good as it's going to get without spending real money — which is a genuine answer, and cheaper than discovering it eighteen months into a project.
It hasn't happened yet. I'm not going to pretend it can't.
Why the fee comes off the build
Because otherwise the incentive is wrong. An audit that gets paid for whether or not anything follows is an audit that can afford to be vague, and the industry is full of them.
Taking the fee off the first build means I'm carrying some of the risk that the plan is any good. If the findings are thin, you don't buy the build, and I've done two weeks for £1,200. That is deliberately uncomfortable for me and it's the main reason the recommendations stay short.
When not to buy it
Four cases where I'd tell you to keep your money:
- You already know the answer and just need it built. Skip the audit, describe the job, and I'll quote the build. Paying me to discover something you could have told me in ten minutes is daft.
- The problem is a person, not a process. Automation makes a working process faster. It makes a broken one break more efficiently, and it will not settle an argument about who owns what.
- You're mid-migration. If you're moving accounting system or platform in the next quarter, wait. I'd be automating something that's about to stop existing.
- Cashflow is genuinely tight this month. The first step is free and 45 minutes. Take that instead and use it properly — I'll still tell you the two or three things I'd change.
The honest comparison
The ledger on the front page puts a typical twelve-person firm's manual admin at around £26,000 a year. That figure is illustrative — yours will be different, and working out your own is what the free calculator is for.
But it sets the scale of the decision. If an audit finds one job worth a fifth of a number like that — say £5,000 a year — and the build to fix it costs £3,500, that's roughly an eight-month payback on a one-off cost, and it stays paid back every year afterwards. Good, then. Not a rout, and I'd rather show you the working than round it up. If it finds nothing, you've spent £1,200 to stop wondering.
Both of those are better than the third option, which is continuing to pay for it in hours and never seeing it on an invoice.
Prices are per project, not per day, and nothing starts without a written scope you've agreed. If you want the number for your own business before you speak to anyone, the audit calculator on the front page takes about three minutes and doesn't ask for an email address to show you the total.