Fixed Price vs Time and Materials

Fixed price feels safer. Often it is not. It moves risk to the supplier, and the supplier prices that risk into the number — then protects it in ways that work against you.

Neither model is dishonest. Both have a shape, and understanding the shape tells you which to use and what to watch.

What each model actually does

Fixed price

You agree a scope and a number. The supplier delivers it for that number regardless of effort.

Risk sits with the supplier. They absorb an underestimate.

What follows from that:

• The number includes a risk premium. Anyone quoting fixed price on uncertain scope adds 20–40% for the unknown. If the project runs smoothly, you paid that for nothing.
• Scope becomes contested. Once the price is fixed, the only lever the supplier has is what counts as “in scope”. Expect that boundary to be defended closely.
• Change requests become the margin. Some firms deliberately quote low and recover through changes. This is a known and legal commercial strategy.
• Cutting corners becomes rational. When effort rises and price does not, the pressure falls on testing, documentation and edge cases — the parts you cannot see at handover.

Fixed price works when the scope is genuinely well understood, discovery has already happened, and the work is similar to things the supplier has built before.

Time and materials

You pay for time at an agreed rate. Scope can evolve.

Risk sits with you. You absorb an overrun.

What follows:

• No risk premium. You pay for the work that happens.
• Scope can change without a negotiation. Learning something mid-project is normal rather than contractual.
• Quality pressure is lower. Doing it properly does not cost the supplier margin.
• The budget is open-ended unless you cap it. This is the real risk, and it is manageable.
• It requires trust, or oversight. You need visibility into what is being done.

T&M works when scope is genuinely uncertain, you want to adapt as you learn, or the relationship is ongoing.

Hybrid models worth knowing about

Beyond the two main options, three structures that appear in real contracts.

Target cost with shared pain and gain. An agreed target. Come in under it and the saving is split; go over and the overrun is split to a cap. Aligns both sides toward efficiency and requires enough trust to share the books. More common in construction than software, and it works here.

Fixed price per sprint. A fixed amount per two-week block, with the scope of each block agreed at the start of it. You get budget predictability per block and flexibility across the project. Effectively capped T&M with tidier accounting.

Fixed price with a change pool. The fixed price includes a pre-agreed allowance for changes — say 15% of the contract value — drawn down against agreed changes. Removes the friction of negotiating each one while keeping a boundary. Underused and genuinely good for projects where change is expected.

Milestone-based with retention. A percentage of each milestone — typically 10% — held until final acceptance. Common in larger contracts and it gives you leverage at the point where leverage matters most.

What to avoid: any structure where your payment obligation is disconnected from an accepted deliverable. Paying for elapsed time with nothing to point at is how projects drift.

Side by side

The structure that works better than either

Fixed-price discovery, then fixed-price phases.

1. Pay a fixed price for discovery. £3,000–£8,000 typically. Small, bounded, and the output is a requirements document, a technical approach and a costed plan you own.

2. Then fix the price per phase, not for the whole project. Each phase is two to six weeks with a defined deliverable. The scope is close enough to see clearly, so the risk premium is small and honest.

3. Re-estimate between phases using what you have learned.

This gives you budget certainty in the units that matter, avoids paying a large premium on distant uncertainty, and lets you stop after any phase with working software and no penalty.

The important property: you can leave. Phase-based pricing means walking away after phase two costs you phase one and two, not a cancellation fee on a twelve-month contract.

How the models change supplier behaviour

Not a moral point — an incentive one. Both models create pressures, and knowing which pressure you are creating helps you manage it.

Under fixed price, the supplier’s margin improves when effort falls. That is not sinister; it is arithmetic. The pressures it creates:

• Scope gets defended closely, because every concession is margin
• Change requests become commercially attractive
• The invisible work — testing, documentation, edge cases — is where corners get cut, because you cannot see it at handover
• Finishing early is rewarded, which is genuinely good for you

Under time and materials, the supplier is paid for time. The pressures:

• No incentive to finish early
• No incentive to refuse scope growth
• Quality is cheap to deliver, because doing it properly costs the supplier nothing
• Investigation and care are rewarded rather than penalised

The practical conclusion: under fixed price, specify the invisible work explicitly — testing, documentation, handover contents — as deliverables, because those are what the pressure falls on. Under time and materials, cap it and review it, because those are what the pressure falls on.

Neither model is more honest. They put the pressure in different places, and you protect the place where it lands.

Why “fixed price before discovery” is a warning sign

If a supplier quotes a firm fixed price for a system they have not investigated, one of three things is true:

1. They have padded heavily. You are paying for uncertainty that may not materialise.
2. They intend to recover through change requests. The low number wins the work; the margin arrives later.
3. They have not understood the problem and will discover that mid-build. This ends in a difficult conversation or a quietly poor result.

The Panorama 2026 ERP study found that among projects exceeding budget, 54.9% needed additional technology discovered after the project began and 51% saw scope grow beyond the original plan. Those are not rare events. A fixed price quoted before anyone has looked properly is a number quoted in ignorance of the most likely cause of overrun.

Discovery is what makes a fixed price meaningful. Without it, the certainty is illusory.

What a change request should look like

The mechanism that keeps either model honest. Agree it before work starts, not when the first change arrives.

The process:

1. You request it in writing. Email is fine. Verbal changes are the source of most disputes.
2. They quote within an agreed period — two to three working days is reasonable — with cost, timeline impact, and any knock-on effects.
3. You approve in writing before any work begins.
4. It is logged, so there is a record of what changed and why.

What a quote should contain: the work, the cost, the effect on the delivery date, and whether anything already built needs changing as a result. That last one is where the real cost often sits.

The distinction that matters, and should be written into the contract:

The warning sign: a supplier who classifies everything as a change request. The other warning sign: a client who classifies everything as a defect. A functioning project has some of each, and the contract should name the test rather than leave it to whoever argues harder.

Protecting yourself under either model

Under fixed price

• Get the scope written in detail, including what is explicitly excluded
• Agree a change process up front — how changes are priced, how long a quote takes, who approves
• Define acceptance criteria before work starts. “Done” must be testable, not a matter of opinion
• Hold a meaningful final payment until acceptance. 30% is reasonable
• Specify testing and documentation as deliverables, so they are not the first thing cut under pressure
• Ask what happens if they underestimated. A supplier who says “that’s our risk” and means it is telling you something useful

Under time and materials

• Cap it. “Not to exceed £X without written approval.” This removes almost all the downside
• Agree reporting — what was done, by whom, how long. Weekly
• Define the rate and who is on the project. Being billed a senior rate for junior work is a known pattern
• Agree a review point — typically every two to four weeks, with the option to stop
• Get estimates before work, not invoices after

Retainers, and when they beat both

The third model, for work that continues rather than completes.

How it works: a monthly fee buys a defined allowance — response times, monitoring, and an amount of change work.

Where it fits: after launch, when you have a working system and a steady trickle of small changes. Quoting each one individually costs more in administration than the work.

What a good retainer specifies:

• Response times by severity, with definitions. “Critical” must mean something checkable.
• What is included — monitoring, security patching, dependency updates, backups, and a change allowance in hours or points.
• What is not — new features above a size, which get quoted separately.
• Whether unused allowance rolls over. Usually not, and that should be stated rather than discovered.
• Notice period, both ways. Thirty days is normal.
• That you keep everything if you leave. The code and the infrastructure are yours regardless.

The warning sign: a retainer that is required in order to keep access to your own system. That is not a support arrangement, it is a licence with a different name.

When a retainer is wrong: the system is stable and genuinely changes once or twice a year. Pay ad-hoc and keep the money.

What we do, and why

Discovery is fixed price. £3,000–£8,000, bounded, and you own the output including the right to take it elsewhere.

Builds are fixed price per phase, after discovery. We can quote honestly because we have looked. The risk premium is small because the uncertainty is small.

Ongoing work is retainer or T&M with a cap.

When we are wrong on a fixed price, that is our problem. You pay the quoted figure. That is the point of the model and it is why we only offer it after discovery.

When something is genuinely new, you get a revised estimate before the work happens, with the option of doing less. Never an invoice afterwards.

Full pricing → · How we work →

Capped time and materials, in detail

The model that is frequently the best value and the least often offered, because it requires more trust in both directions.

How it works: you pay for time at an agreed rate, with a written cap. Work stops at the cap unless you approve more.

Why it often costs less than fixed price: you are not paying the 20–40% risk premium. If the work goes smoothly, you keep that money rather than the supplier.

What makes it safe — all five, not three of them:

1. A written cap. “Not to exceed £X without written approval.” This removes almost all the downside.

2. Weekly reporting. What was done, by whom, how many hours. Not a summary — a list. You should be able to see where the money went.

3. Named people at agreed rates. Being billed a senior rate for junior work is a known pattern, and a rate card by role prevents it.

4. A review point. Every two to four weeks, with an explicit option to stop. A cap without a review point means you discover a problem at 90% of the budget.

5. Estimates before work, not invoices after. Any piece of work above a trivial size gets an estimate first.

When capped T&M beats fixed price: scope is genuinely uncertain, you want to adapt as you learn, the relationship is ongoing, or the fixed-price premium looks high relative to the risk.

When fixed price still wins: you need absolute budget certainty for internal approval, or the scope is thoroughly understood after discovery and the premium is consequently small.

Frequently asked questions

Written by Jas Singh. Last updated 8 October 2026.

Overrun cause figures from the Panorama Consulting 2026 ERP report (170 organisations, January 2025 – January 2026) as reported by secondary sources; cited as directional.