Time & materials vs fixed price: which billing model for your dev shop

The three models, one paragraph each

Almost every services contract is a variant of one of three billing models. The difference that matters is not the invoice format — it is who is on the hook when the work takes longer than anyone expected.

• Time & materials (T&M). You bill the client for hours actually worked, at agreed rates, plus pass-through costs. The invoice moves up and down with real effort, so scope can change without renegotiating the contract. The client pays for whatever it takes — which means the client carries the risk of an overrun.
• Fixed price (fixed bid). You agree one total for one defined scope, and that total does not move even if the work runs long. Billing is usually tied to milestones or a delivery schedule. Because the price is locked, the vendor carries the risk of an overrun — every hour past the estimate comes out of margin.
• Retainer / dedicated team. The client buys ongoing capacity — a set number of people or hours every month — for a flat recurring fee. You are selling availability, not a finished deliverable. Both sides trade the overrun gamble for predictability: stable monthly revenue for you, a stable monthly bill and a reserved team for them.

The real question: who carries the risk

Strip away the labels and every model is really a way of splitting one risk — the gap between what you estimated and what the work actually costs.

• Fixed price pushes the risk onto the vendor. If the estimate is wrong, the vendor absorbs the extra hours. The rational response is to pad the estimate with a contingency buffer, so the client often pays a risk premium baked into the price whether or not anything goes wrong.
• T&M pushes the risk onto the client. The meter runs on real hours, so a slow month is the client's problem, not yours. The rational client response is to demand a cap, a not-to-exceed ceiling, and enough transparency to see the burn — otherwise they are signing a blank cheque.
• Retainer trades the risk for predictability. Neither side is betting on a single estimate; they are agreeing a steady rate of spend. The new risk is quieter: under-use, where the client pays for idle capacity, or over-service, where you quietly deliver more than the retainer covers.

There is no model that removes the risk — only models that decide who holds it. A fair contract prices that choice honestly instead of hiding it.

When each model fits

Match the model to how well you actually understand the work, not to which one feels safest to quote.

• Reach for T&M when the scope is unclear or evolving. Discovery work, R&D, integrations with systems you cannot see yet, and long engagements where priorities will shift are all a poor fit for a fixed number. T&M lets the work change direction without a contract amendment every time.
• Use fixed price only when the scope is genuinely well-defined and stable. A migration with a known source and target, a marketing site with signed-off designs, a clearly bounded feature — work you have done close cousins of before and can estimate with confidence. If you cannot write down exactly what 'done' means, it is not ready for a fixed bid.
• Choose a retainer or dedicated team for ongoing product work and staff augmentation. When the client needs a team to keep building month after month, or to extend their own team with skills they lack, there is no single deliverable to fix a price around. You are renting capacity, and a recurring fee fits that shape.

A useful tell: if the client can hand you a frozen specification, fixed price is on the table. If the specification will obviously change as you learn, it is not — and pretending otherwise just moves the argument to a change-request fight later.

Fixed-price discipline: estimate, then defend the scope

Fixed price is the highest-margin model when it works and the fastest way to lose money when it does not. The discipline has three parts.

• Estimate from the bottom up, then add contingency. Break the scope into tasks, estimate each, and add a buffer sized to how much you actually know — a small contingency for familiar work, a large one for anything novel. A single confident number on a complex project is a guess wearing a suit.
• Write the scope down precisely, including what is out. The statement of work should list assumptions and explicit exclusions, not just deliverables. Most fixed-price disputes are about something nobody wrote down.
• Run a real change-request process. When the client asks for something outside the agreed scope, treat it as a change request: re-estimate it, quote it, and get sign-off before building. This is not bureaucracy — it is the mechanism that keeps a fixed price fixed.

The number-one margin killer is fixed price on a vague brief. If the scope is fuzzy, every gap between what the client imagined and what you quoted becomes free work, because the price cannot move but the expectations can.

Worked example: a fixed-price overrun

Say you quote a flat fee for an estimated 400 hours. The brief was thin, scope creep adds work, and delivery actually takes 560 hours. On T&M the client would have paid for 560; on a fixed price they pay for 400 and you absorb 160 hours — often the entire margin on the project and then some. The estimate did not just slip; it slipped onto your side of the table.

T&M discipline: caps, transparency, and trust

T&M is only as good as the trust behind it, because the client is paying for hours they mostly cannot see. Left uncontrolled, an open-ended meter feels to the client like a bill with no ceiling — and that erodes the relationship faster than any single overrun.

• Agree a cap or not-to-exceed (NTE). A cap sets a ceiling the invoice will not cross without a fresh conversation. The client gets budget certainty; you keep the flexibility of billing real hours underneath it. As you approach the cap you flag it early, rather than blowing through it.
• Report the burn every week. Send a short, regular view of hours spent versus budget remaining. Predictable reporting turns 'how much have we spent?' from an anxious question into a non-event, and it is one of the strongest reasons clients renew T&M work.
• Tie hours to something legible. Log time against tasks or tickets the client recognises, so the invoice reconciles to work they can see. Hours billed to a black box invite disputes; hours billed to named tasks get approved.

Worked example: capping T&M

A T&M engagement at a blended $80/hour with a not-to-exceed of $40,000 gives the client a firm worst case of 500 hours. You bill actual hours each month, but the moment cumulative billings near ~$36–38k you raise it with the client and agree either to close out inside the cap or to extend it deliberately. Nobody is surprised by the invoice, which is the whole point.

Hybrids that borrow the best of both

Most experienced shops rarely run a pure model. The useful hybrids all try to keep the vendor's margin safe while giving the client the predictability they want.

• Capped T&M. Bill real hours, but under an agreed ceiling. This is the workhorse hybrid: the client gets a worst-case number, you get paid for actual effort up to it. Overrun risk is shared — you carry it above the cap, they carry the variability below it.
• Fixed-price discovery, then T&M (or fixed) build. Sell a short, well-defined discovery or design phase at a fixed price, produce a real specification and estimate, then price the build once the scope is actually known. This converts an unknowable fixed bid into two smaller, honest commitments.
• Milestone-based fixed price. Split a fixed-price project into milestones, each with its own scope, price, and acceptance criteria. The client pays as value lands, you get paid through the project instead of only at the end, and a scope problem surfaces at the next milestone instead of at final delivery.

The common thread: none of these pretend the scope is more certain than it is. They stage the commitment so that price is set when knowledge is highest, not when it is lowest.

Cash flow and margin, model by model

The billing model does not just decide your margin — it decides when the money arrives, which for a small shop matters just as much.

• Fixed price: lumpy cash, estimation risk. Cash tends to arrive in milestone chunks — a deposit, a mid-point payment, a final payment on acceptance — so revenue is uneven and a delayed sign-off can strand a big receivable. Margin is high if the estimate holds and negative if it does not; the whole model rides on estimation quality. Mind the gap between an early deposit and the final payment, because you are funding the work in between.
• T&M: smoother cash, margin follows utilization. You invoice logged hours every month, so cash is steadier and tracks the work as it happens. Margin is the spread between your bill rate and your fully-loaded cost, protected as long as hours are captured and the people are actually billable. The risk moves from estimation to time capture: an unlogged hour is simply unbilled revenue.
• Retainer / dedicated team: the most predictable of the three. A flat recurring fee is the smoothest cash flow you can have and the easiest to forecast on both sides. The margin risk is utilization of the reserved team — bench time inside a retainer is pure cost — so you still track hours, not to build the invoice but to see whether the retainer is actually profitable.

Across all three, margin is only real once the invoice is paid. Late or disputed invoices turn paper margin into a cash-flow problem regardless of the model on the contract — see turning timesheets into client invoices and setting bill rates that protect margin.

A decision checklist

Run a new engagement through these questions before you quote. The answers usually point at one model.

1. How clear is the scope? Frozen and signed-off → fixed price is possible. Fuzzy or exploratory → T&M, or a fixed-price discovery first.
2. How often will requirements change? Rarely → fixed price. Frequently → T&M or a retainer, so change does not mean a contract amendment.
3. What is the client's risk appetite? Wants a guaranteed number → fixed price or capped T&M. Comfortable paying for real effort with transparency → T&M.
4. How long is the engagement? One bounded deliverable → fixed price or capped T&M. Ongoing, multi-month product work → retainer or dedicated team.
5. How confident is your estimate? You have delivered close cousins before → fixed price is defensible. First of its kind → do not fix a price on it yet.
6. Can you see and cap the risk? If you must take fixed price on imperfect scope, cap your exposure with milestones and a tight change-request process.

When the answers conflict — clear scope but a nervous client, or an ongoing need with a fixed budget — reach for a hybrid rather than forcing a pure model. The model is a tool for allocating risk fairly, not a loyalty badge.

How Helia HR does this

Helia HR is an HR system with delivery ops built in, so whichever model a project uses, the same timesheets and rates feed the invoice:

• A billing model per project. Set each project to time & materials or fixed price — the choice drives how its invoices are generated, so the contract you signed and the invoice you send stay in sync.
• T&M billed from actual hours. Approved timesheet hours flow into a client invoice at your per-project (or per-assignment) rates — the model where "bill what you logged" has to be effortless, or it quietly leaks.
• Fixed price & milestones. Draft invoice lines are fully editable, so a fixed fee or a milestone payment is a line you set, while the project still tracks actual hours against the estimate underneath.
• Margin per project, whatever the model. The profitability view compares revenue against fully-loaded cost, per project and per customer, FX-consolidated — so you see which contracts and models actually earn, not just which ones bill.

Timesheets, capacity, invoicing and profitability come in the Delivery pack; the HR basics come in the base.

FAQ

Is T&M or fixed price better?

Neither is better in general — they allocate risk differently. Fixed price suits well-defined, stable scope and puts overrun risk on the vendor; T&M suits evolving or unclear scope and puts that risk on the client, usually under a cap. Match the model to how well the work is understood, and use a hybrid when the answer is in between.

How do you cap a T&M contract?

Agree a not-to-exceed (NTE) ceiling in the contract: you bill actual hours, but the total will not pass the cap without a fresh, signed agreement. Pair it with weekly burn reporting so the client sees hours-spent versus budget-remaining, and flag it early when billings approach the cap rather than after crossing it. Capped T&M gives the client a worst-case number while you still get paid for real effort.

What is the dedicated team model?

A dedicated team is a form of retainer where the client pays a flat monthly fee for a reserved group of people who work only on their account. It suits ongoing product development and staff augmentation, where there is no single deliverable to price. The client gets predictable capacity and priority; the vendor gets predictable revenue but carries the risk of keeping that team fully utilized.

Which billing model is the most profitable?

Fixed price has the highest ceiling — if your estimate holds, every efficiency gain is your margin — but also the highest downside, because an overrun comes straight out of profit. T&M offers a steadier, more predictable margin tied to the spread between your bill rate and cost. Retainers are the most predictable of all but depend on keeping the reserved team busy. The most profitable model is the one whose risk you can actually control.

Why is fixed price on a vague brief so risky?

Because the price is locked but the expectations are not. Every gap between what the client pictured and what you quoted becomes unpaid work, since you cannot raise the price but the client can keep asking. Fixed price only protects your margin when the scope is precise and a change-request process handles anything outside it.

Can you switch billing models mid-engagement?

Yes, and it is common — for example a fixed-price discovery phase that converts into T&M or a retainer for the build once the scope is known. Agree the trigger and the new terms up front so the switch is a planned step, not a renegotiation under pressure. Changing models is easiest at a natural boundary such as a milestone or a contract renewal.