Fix scope creep on fixed-fee agency projects by writing a tight statement of work (SOW) with explicit exclusions, logging every out-of-scope request, and triggering a formal change order before doing the extra work. Most margin loss comes from "quick favors" that never get billed — so make tracking and change orders non-negotiable, not optional.
Scope creep is the silent margin killer on fixed-fee work. You quote a number, the client keeps adding "small" requests, and three months later you've burned 40% more hours than you priced. The fix isn't saying no to everything — it's building friction into the moment scope expands so it gets caught, priced, and billed.
Why fixed-fee projects bleed margin
Fixed-fee pricing transfers all overage risk to the agency. If you estimate 200 hours and the project takes 300, you eat the 100 hours. Scope creep is what drives that overage, and it usually arrives in tiny increments nobody flags:
- "Can you just tweak the homepage too?"
- "While you're in there, add a second language."
- "We showed it to the CEO and he wants a different direction."
Each request feels too small to fight over. Stacked together, they destroy profitability. The root problem is almost always a vague SOW plus a team that doesn't track or escalate out-of-scope work in real time.
Write a tight SOW with explicit exclusions
The SOW is your first and best defense. Vague scope language is what lets clients reasonably assume extras are included. A strong fixed-fee SOW should specify:
- Deliverables by name and quantity — "3 landing page designs," not "website design work."
- Revision limits — e.g., "two rounds of revisions per deliverable; additional rounds billed at $X/hour."
- An explicit exclusions list — name the things clients commonly assume are included but aren't (copywriting, hosting, third-party integrations, post-launch support).
- Assumptions — "Client provides final copy by [date]" so delays don't become your problem.
This is the same discovery discipline that makes a good sales discovery call valuable — surface the real requirements before you commit a number. The Project Management Institute's research on project scope management reinforces that documented, agreed scope is the single biggest predictor of on-budget delivery.
Make change orders mandatory, not awkward
Most agencies have a change-order process on paper that nobody actually uses because it feels confrontational. The fix is to make it routine and lightweight.
When a request falls outside the SOW, respond with a standard line: "Happy to do that — it's outside the current scope, so I'll send a quick change order with the cost and timeline impact." That framing positions the change order as a normal step, not a fight. Keep a template ready so issuing one takes two minutes:
CHANGE ORDER #CO-007
Project: Acme Website Redesign
Requested: Add Spanish-language version of all pages
Impact: +28 hours, +$4,200
New timeline: +6 business days
Approval: ____________________ Date: ________
The key rule: no out-of-scope work starts until the change order is signed. That one policy stops the slow bleed cold.
Track hours against scope in real time
You can't manage what you don't measure. Track actual hours against the budgeted estimate per deliverable, not just per project. When a single deliverable hits 80% of its budget, that's the alarm to investigate scope before it overruns.
Set burn-rate thresholds
- 50% of budget: routine check-in, confirm you're on track.
- 80% of budget: flag to the project lead, review what's left.
- 100% with work remaining: stop and assess — is this scope creep or estimation error? If it's creep, issue a change order.
This distinction matters. If you blew the estimate on in-scope work, that's a pricing problem to fix next quote. If extra deliverables crept in, that's billable.
Price the risk into the model
Some scope drift is inevitable, so price for it. A few options:
- Contingency buffer: add 10-20% to fixed-fee estimates to absorb minor creep without renegotiating.
- Retainer or hybrid models: move recurring or fuzzy-scope work to time-and-materials or a monthly retainer, reserving fixed-fee for tightly defined projects only.
- Phased contracts: quote discovery as a separate paid phase, then price the build once requirements are actually known.
Many agencies move ambiguous, ongoing work off fixed-fee entirely — the same logic teams use when weighing the tradeoffs of in-house versus outsourced capacity: match the commitment model to how predictable the work really is.
Train the team to spot creep early
Scope creep usually slips past at the individual-contributor level — a designer or developer just does the favor to keep the client happy. Fix this by making it explicit that catching out-of-scope requests is part of everyone's job, not snitching. Give the team a simple script: "That's a great idea — let me check with the project lead since it might be outside our current scope." That deflects the request without saying no and routes it to someone who can issue a change order.
Key takeaways
- Tight SOW first: name deliverables, cap revisions, list exclusions, and state assumptions.
- Change orders are routine: no out-of-scope work starts until one is signed.
- Track per deliverable: use 50/80/100% burn thresholds to catch overruns before they happen.
- Price the risk: add contingency buffers or move fuzzy work to retainer/T&M models.
- Empower the team: give frontline staff a script to route requests instead of absorbing them.
The agencies that protect margin don't have fewer scope-change requests — they just have a fast, normalized process that turns every request into either a documented exclusion or a billed change order. Build that friction in, and fixed-fee projects stop eating your profit.