Why silence is the expensive option
Scope creep almost never arrives as an obvious ask. It shows up as "oh, and can you also just..." or "while you're in there, could you..." — small, reasonable-sounding requests that individually feel too minor to push back on. The problem isn't any single request. It's that saying yes silently, every time, teaches the client that scope expansion costs nothing — so it keeps happening, and by the end of the job you've done real extra work you were never paid for.
The fix isn't refusing reasonable requests. It's responding to every out-of-scope ask the same way, immediately, so the client always knows where the line is.
The script
This is close to word-for-word what experienced freelancers actually use, because it does three things at once: it says yes (you're not being difficult), it names the cost (so it's not free), and it puts the decision back on the client (so you're not the one who has to say no):
"Happy to add that — it's outside the original scope, so it'll add about [X hours / $Y] and push the timeline by [Z]. Want me to go ahead?"
That's it. No apology, no long explanation, no guilt about bringing up money. Just: yes, here's what it costs, your call.
Why this works better than the alternatives
Silently doing it costs you the time and money, and guarantees it happens again — you've just shown the client that "the original scope" is more of a suggestion than an agreement.
Refusing outright ("that's not what we agreed") is technically correct but reads as inflexible, and often isn't even necessary — most clients aren't trying to get free work, they just haven't registered that the request is a new ask rather than a clarification of the old one.
The script splits the difference: you're not blocking the work, you're pricing it. Most clients will either say yes (and now you're getting paid for it) or say "actually never mind, not that important" (and now you've avoided free work that wasn't even valued enough to pay for) — both outcomes are better than silently absorbing it.
Use it immediately, not after you've already done the work
The script only works if you say it before you do the extra work, not after. If you've already built the thing and then ask for more money, you've lost your leverage — the client already has what they wanted. Say it the moment the request lands, even mid-conversation: "Sure, that's a quick scope note — [script] — let me know and I'll get started."
Say it in the job thread, not in a call
If a scope-expanding request comes up on a call or a quick message, that's fine — but restate the script back in the job thread afterward: "Confirming from our call — adding X will add Y hours, let me know if you want me to proceed." This does double duty: it's the actual scope conversation, and it's also the record that shows what was agreed if there's ever a disagreement about what was included later.
When the request is small enough to just include
Not every extra ask needs the full script. If something takes two minutes and clearly falls under "obviously part of doing this properly" — a typo fix, a one-line copy change — just do it. The script is for anything that takes real time or wasn't part of what you scoped and quoted. Using it selectively, only when it matters, is what keeps the relationship easy instead of transactional.
The one-line version
Every out-of-scope request gets the same three-part response — yes, here's the cost, your call — said immediately and in writing, every time, no exceptions.