The work that was never worth doing

Somewhere in your business right now there is an invoice that is fifty quid too high

Somewhere in your business right now there is an invoice that is fifty quid too high. A carrier has rounded something up, or mapped a standard order to express, or added a surcharge that doesn’t apply. Nobody will ever catch it. Not because nobody could. Because checking would take an hour, and the error is only fifty quid.

So you pay it. You pay it every month. You have always paid it.

This is not a finance problem. It is a maths problem. Every business runs an unspoken calculation on every task: is the time worth the result? Most of the time the honest answer is no. The juice is not worth the squeeze. So a whole category of work sits just below the line. Worth doing in theory. Never done in practice.

That line just moved.

Take carrier invoices. Give Claude your order data as a CSV and the rate card you actually agreed with DHL. Ask it whether the invoice matches. It goes line by line, the way nobody has the patience to, and tells you where the numbers don’t add up.

Here is the part that makes it useful rather than just clever. When the invoice comes back too high, it is often not the carrier’s fault. Someone shipped a week of standard orders as express. The warehouse used the wrong labels. Claude flags the gap, you find the cause. Sometimes it’s DHL. Often it’s you.

The same trick works on 3PL invoices. They charge per unit in, per unit out, plus a list of extras nobody reads. Give Claude the contract and the invoice and let it tell you if they line up.

Global-e payout files are the best example, because they are genuinely horrible. Cross-border charges, odd payment terms, surcharges that move with the price of fuel, every brand on a slightly different deal. Give Claude the rules you signed up to and it will tell you whether the payout is what you were owed, instead of you taking it on faith because the file is too long to argue with.

Review contracts before the need for legal

Then the contracts. The eight grand deal that lands in your inbox with twelve pages of terms. Too small to bother the legal team with. Too boring to read yourself. So you sign it. Claude reads it in a minute and pulls out the bits that matter.

None of this is new work. You always knew you should check the invoices. You always knew the contract had a clause in it somewhere. The job has not changed. The cost of doing it has.

That is the bit worth sitting with. AI has not handed you a superpower. It has lowered the price of attention. And when the price of attention drops, a whole pile of work that was never worth your time becomes worth it. Twenty minutes to save fifty quid used to be a bad trade. Now it is twenty minutes you did not spend, because Claude did.

So go and look at the jobs you have been ignoring because they are too small. That is where the easy wins are. They were always there. You just could not afford to look.