How to Estimate a Fixed-Price Project Without Losing Money
A step-by-step method to estimate a fixed-price project: task ranges, project management time, a costed contingency and a day-rate check, with a worked example.

On a fixed-price project, every hour beyond your estimate comes out of your margin. Most of that loss is decided before the work starts, in the estimate itself. This guide gives you a repeatable method in six steps, a full worked example and the assumptions to write into your quote.
If you are still deciding whether a fixed price is the right model for this client, read Hourly, Fixed Price, Retainer or Value-Based: How to Choose first.
Why single-number estimates run short
People tend to underestimate how long their own tasks will take, even when similar tasks ran late before. Psychologists call this the planning fallacy. In one study, students estimated on average that their honors thesis would take 33.9 days; it took 55.5 days, and only 29.7% finished by their own estimate. Even their "worst case" guess, 48.6 days, was too short (Buehler, Griffin and Ross, Journal of Personality and Social Psychology, 1994).
The lesson for a quote: one number per task, given from memory, is likely to be low. Ranges, explicit overhead and a costed contingency correct for that.
Step 1: Break the work into tasks you can picture
List every deliverable, then split each one into tasks small enough that you can picture yourself doing them. A useful rule of thumb is that no task should be longer than about two days; if it is, split it again.
Include the work clients don't see:
- kickoff call, brief and questions;
- internal reviews and revision rounds;
- testing, corrections and handover;
- travel, set-up and access to the client's tools;
- admin: contract, invoicing, payment follow-up.
If part of the work is outside your skills, get a written quote from a subcontractor before you price, and add your coordination time. You can find consultants available for missions on Sendbiz. Our guide on subcontracting work to freelancers covers how to keep quality under control.
Step 2: Estimate each task as a range
For each task, write three numbers in hours:
- Optimistic (O): everything goes right the first time.
- Most likely (M): what usually happens.
- Pessimistic (P): a realistic bad case, such as an extra revision round or a slow client. Base it on a real past project, not on a disaster.
Then combine them. The PERT formula weights the most likely value: (O + 4 × M + P) / 6. A simple average, (O + M + P) / 3, gives more weight to the extremes (3-point estimating, Nulab). Both are approximations; what matters is that the pessimistic case is written down and counted.
Step 3: Add project management and communication time
Calls, emails, status updates, scheduling and coordination are real work, and they grow with the number of people on the client side. Add them as a separate line rather than hoping they fit inside the tasks.
Set the percentage from your own timesheets: compare the hours you logged on coordination with the hours on delivery for two or three past projects. If you don't track time yet, start with the next project. In the example below, we assume 15%; that is an assumption for the example, not a standard.
Step 4: Set a contingency from named risks
A contingency is not a random percentage added at the end. List the risks you can name, then for each one estimate a probability and an impact in hours. The expected cost of a risk is probability × impact, and the sum for all listed risks gives your contingency (A model to develop and use risk contingency reserve, PMI, 2014).
The same paper separates two reserves. The contingency reserve covers "known unknowns", the risks on your list. A management reserve covers "unknown unknowns", the risks nobody identified. In a small firm, that second buffer is usually your margin and the way you round the price.
Step 5: Price it and check it against your day rate
Add tasks, project management and contingency, then multiply by your hourly rate. Then run three checks:
- Day-rate check: divide the total hours by your billable hours per day. Is the price in line with your day rate for that many days? If you need to set or review your rate, use the day rate calculator and the guide on how to calculate a consulting day rate.
- Exposure check: price the pessimistic total. That is roughly what you would work if most things went wrong. If the gap is too large to absorb, reduce the scope, split the project into phases or add payment milestones.
- Capacity check: spread the hours over the calendar. If the project blocks other paid work for weeks, include that in your decision.
Worked example: a five-page website
This is an example with stated assumptions, not a benchmark. A small studio quotes a five-page website with a content management system. Its hourly rate is $85 (any currency works the same way), and it bills about 7 hours per working day.
| Task | O (h) | M (h) | P (h) | PERT (h) |
|---|---|---|---|---|
| Kickoff and brief | 4 | 6 | 10 | 6.3 |
| Wireframes, 5 pages | 8 | 12 | 20 | 12.7 |
| Visual design, 2 revision rounds | 12 | 18 | 30 | 19.0 |
| Build and CMS set-up | 20 | 30 | 50 | 31.7 |
| Content entry | 4 | 6 | 12 | 6.7 |
| Testing and fixes | 6 | 8 | 16 | 9.0 |
| Launch and handover | 2 | 3 | 6 | 3.3 |
| Total | 56 | 83 | 144 | 88.7 |
The studio then lists three risks:
| Risk | Probability | Impact (h) | Expected (h) |
|---|---|---|---|
| Client content arrives late, extra coordination | 50% | 6 | 3 |
| A required plugin doesn't fit, workaround needed | 30% | 10 | 3 |
| A new stakeholder joins the reviews | 25% | 8 | 2 |
| Contingency | 8 |
The estimate becomes:
- tasks (PERT): 88.7 hours;
- project management, assumed at 15%: 13.3 hours;
- contingency: 8 hours;
- total: 110 hours, or about 15.7 working days;
- price: 110 × $85 = $9,350, rounded up to $9,400.
Compare this with a quote built on the "most likely" column alone: 83 hours, or $7,055. That quote would have left 27 hours, about $2,300, unpaid in an ordinary project. The exposure check matters too: the pessimistic tasks (144 hours) plus 15% project management come to about 166 hours. If that happened, the studio would work around 56 hours beyond what the price covers, so it adds a milestone payment after design approval.
Step 6: Write the assumptions into the quote
A fixed price is only fixed for a defined scope. Write down what your estimate assumes, so that any change becomes a priced change instead of free work. Our consulting proposal template shows where these go.
- Deliverables: the list, with quantities (five pages, not "a website").
- Revision rounds: how many are included, and the price of an extra one.
- Client inputs: what the client provides (content, access, approvals) and by when.
- Exclusions: what is not included, in plain words.
- Change requests: how a change is estimated and approved in writing before work starts.
- Payment milestones: for example, a deposit, a payment at a key approval and the balance at delivery.
- Taxes: state whether prices include or exclude tax. Rules differ by country and by client (for example sales tax in the US, VAT in the UK and France, GST in India), so check with your accountant.
- Validity: how long the quote stands before you re-check it.
After delivery, compare your actual hours with each estimated line. That comparison is what makes your next estimate more accurate, and it gives you real data for your project management percentage and your risk list.
Common mistakes
- Estimating from the client's budget. Estimate the work first, then decide whether to adjust the scope or walk away.
- Quoting the "most likely" total. It ignores everything that usually goes a bit wrong.
- Forgetting non-delivery work. Calls, admin and handover add up across a project.
- A contingency with no risks behind it. A flat percentage is hard to explain and easy to cut in negotiation.
- Unlimited revisions. Without a number in the quote, every round is free.
- No change process. Small "quick" requests accepted by email erode the margin.
- Never comparing estimate and actuals. Without that review, the same gaps repeat on the next quote.
Frequently asked questions
How much contingency should I add to a fixed-price project?
There is no universal percentage. List the risks you can name, multiply each one's probability by its impact in hours, and add them up. Then compare with your actual hours after each project to refine it.
Should I show the contingency to the client?
You don't have to show it as a separate line; many providers include it in the price. What you should show are the assumptions behind the price, so the client understands what is and isn't included.
What if the client asks for changes after signing?
Use the change request process written in your quote: estimate the change with the same method, send the price and the impact on the timeline, and start only after written approval.
Is the PERT formula better than a simple average?
PERT gives more weight to the most likely value, while a simple average gives more weight to the extremes. Both are approximations; the key is that the pessimistic case is written down and counted.
Do taxes change my estimate?
Taxes don't change the hours, but they change the price the client pays. Rules differ by country and by client type, so state whether your prices include tax and check the rules that apply to you with an accountant.
How to Raise Your Rates With Existing Clients (Email Templates)
How to Write a Consulting Proposal (Annotated Template)
Hourly, Fixed Price, Retainer or Value-Based: How to Choose
How to Calculate Your Consulting Day Rate (With a Worked Example)