Charging for extra work is easy to agree with in theory and hard to do in the moment. The project has started, the client is happy, and then the work quietly drifts past what you agreed. By the time you notice, you have already done some of it, and bringing up money feels like changing the rules halfway through.
This is the part most advice skips. Knowing you should charge is not the problem. The problem is having a process calm enough that you actually use it while the work is moving. Here is the one I would use, broken into four steps you can run in a couple of minutes.
Why mid-project is the hardest moment
Once a project is in motion, three forces push you toward absorbing the extra work for free. Momentum: stopping to write up a price feels slower than just doing the thing. Goodwill: the relationship is good and you do not want to be the one who makes it about money. And the sunk-cost feeling: you have already done a little of it, so it seems too late to charge now.
All three are real, and all three are how a profitable project slowly turns into a break-even one. The fix is not willpower in the moment. It is a process so light that using it is easier than the silent resentment of not using it.
The four-step system
- Tag it. The moment a request lands, name it as outside the agreed scope, in one neutral sentence. You are not arguing, just labelling.
- Price it. Turn the work into a number: hours and rate, or a fixed add-on, plus any timeline impact. Numbers, not adjectives.
- Get the yes in writing. Send it as its own short message and wait for an explicit "approved" before you touch the work.
- Then start. The whole system is just putting the approval before the work instead of after it.
If you only remember one thing, remember the order: approval first, work second. Everything that goes wrong with mid-project billing comes from doing those two in the wrong order. For the exact words to use in the heat of the moment, the companion piece on the "it's just a small change" reply has the scripts.
How to actually price the extra work
Most freelancers freeze at step two, because pricing feels like it needs a formula. It does not. Pick whichever of these three shapes fits the request, and say it plainly.
Best when you cannot fully predict the size yet.
This is outside the original scope. I can take it on at my usual [$120/hr]. My rough estimate is [6 to 8 hours], and I will flag it before I cross [8] so there are no surprises on the invoice.
Best when you can see the edges of the request clearly.
Happy to add this. Since it is beyond the agreed scope, it would be a fixed [$900], and it moves delivery by about [4 business days]. If that works, I will send a short approval and slot it in.
Best when the requests are stacking up into a second project.
These additions are adding up to roughly [a week of work]. Rather than nickel-and-dime each one, I would suggest a small phase two after launch so the current deadline holds. Want me to put together a quick proposal for it?
Put it in writing, then start
A verbal yes on a call evaporates the moment there is a dispute about the invoice. You want a written record that names the added work, the price, and the new timeline. It does not need to be a contract. A short change order, or even an email the client replies to with one word, is enough.
If you want the structure ready to fill in, the change order template lays out exactly what to include, and the change order generator drafts the wording for you from the request. Either way, the rule is the same.
No out-of-scope work starts before a written yes. Not because the client is untrustworthy, but because a written approval is the only version of the agreement that survives a busy month and a fuzzy memory.
When the client pushes back on paying
Sometimes the client genuinely thought it was included. That is not a fight, it is a scope misunderstanding, and you handle it by pointing at the original agreement, not by defending yourself.
Stay on the facts: the agreed scope, the request, and the options.
I understand why it feels like part of the project. The agreed scope was [five pages with two revision rounds], and this sits outside it. I would rather flag it now than surprise you at invoice time. Happy to price it, park it for later, or swap it for something of similar size already in scope.
Offering the swap and the park options matters. It shows you are solving their problem, not just protecting your invoice, and it keeps a "no" from feeling like a wall. Most clients pick the paid option once they see you are being fair about it.
Make the next project easier than this one
Mid-project money conversations are hardest when the rules were never stated. You can remove most of them in advance by writing one sentence into your next proposal: what is included, how many revision rounds, and that anything beyond it gets a quick priced approval before work starts.
Set that expectation on day one and the change order stops feeling like a surprise. It is just the process the client already agreed to. If you want to see what those out-of-scope requests are quietly costing you today, the scope creep calculator puts a number on it in a minute.
Where ProofNod fits
All of this works with plain email and a bit of discipline. We are building ProofNod because the discipline is hardest at exactly the wrong moment: you are mid-sprint, the client is waiting, and writing a formal-enough message feels heavier than just doing the favor. ProofNod turns the four steps into a two-minute approval link: the added work, the price, the timeline impact, one click for the client, and a timestamped record you can point to later.