Fixed quote vs. time-and-materials: what each one really buys you.

Every studio that sells fixed-price work has an incentive to make time-and-materials sound reckless, and every studio that sells T&M has an incentive to make fixed price sound rigid. Both framings are marketing more than they're honest analysis. Here's the version without the thumb on the scale — including the real cases where T&M is the fairer deal for the client.

What each model actually prices

The framing that clarifies most of the confusion: neither model prices hours. Both models price risk. The question that actually matters is who's holding the risk when the initial estimate turns out to be wrong — and it will, to some degree, on almost every non-trivial project, because software estimation is inherently uncertain.

Under time-and-materials, the client holds the risk. If the work takes longer than expected — because the problem was harder than it looked, because the team was less experienced with the stack than assumed, because requirements shifted mid-build — the client's invoice grows to match. The vendor's incentive structure is, at best, neutral to that outcome and at worst mildly perverse: more hours is more revenue, even when it's a sign something went wrong.

Under a fixed quote, the vendor holds the risk. If the work takes longer than scoped, that's the vendor's cost to absorb, not the client's. The vendor's incentive is now aligned with finishing efficiently, because every extra hour comes out of their own margin rather than the client's budget.

Stated that plainly, fixed price sounds strictly better for the buyer, and for a large share of projects it is. But the model only works when the risk being transferred is actually knowable at quote time — which is where the failure mode comes in.

Where fixed quotes fail — and where T&M is genuinely fairer

A fixed quote is a bet the vendor is making on their own estimate. If the scope underlying that estimate is vague — "build us a customer portal," with no further definition — the vendor is pricing genuine uncertainty, and one of two things happens: they price high enough to cover the uncertainty, which means a well-defined project effectively subsidizes badly-defined ones across the vendor's whole client base, or they price optimistically to win the work and either eat the loss or, worse, cut corners to protect margin. Neither outcome serves the client well.

T&M is the more honest model in exactly this situation: genuinely exploratory work, where nobody — including a competent vendor — can accurately estimate the path before starting, because the path itself is part of what's being discovered. R&D-flavored engineering, a technical spike into an unfamiliar integration, or the very early, ambiguous phase of a novel product are all situations where forcing a fixed number onto real uncertainty just moves the risk around without actually reducing it. In those cases, a capped T&M arrangement — time and materials, with an agreed ceiling and regular check-ins — is often the fairer deal for everyone, because it prices what's actually known and stays honest about what isn't.

The line, in practice: if the scope can be genuinely defined — what pages, what integrations, what data model, what's explicitly out of scope — a fixed quote is the better instrument, because the uncertainty being priced is bounded. If the scope can't yet be defined because the project itself hasn't figured out what it's building, T&M with a cap is the more honest instrument, because pretending otherwise just hides the uncertainty inside an artificially confident number.

Real discovery is what makes a fixed quote honest

The failure mode that gives fixed pricing a bad name isn't the model — it's a fixed number generated without a real discovery phase behind it. A vendor who quotes a fixed price off a two-paragraph brief is making the same bet as one who quotes off a proper discovery process, but with far less information, which means the price either has hidden padding or hidden risk. Neither is visible to the client at signing.

A real discovery phase — scoping conversations, a written specification, explicit call-outs of what's in and out of scope, sign-off before the fixed number is issued — is what turns a fixed quote from a guess with a bow on it into an actual transfer of risk the vendor can responsibly absorb. If a vendor is willing to quote fixed price off a single call with no written scope document to follow, that's not confidence. That's a number built to be renegotiated later, once change orders start appearing.

Questions to ask any vendor about overruns

Whichever model is on the table, four questions expose how a vendor actually thinks about risk, regardless of what the pricing page says:

• "What happens if this takes longer than you've estimated?" Under a genuine fixed quote, the honest answer is "that's on us, unless the scope changes." Under T&M, the honest answer is "you'll see it in the invoice, and we'll flag it as soon as we see it coming." Any answer that dodges the question is the actual red flag.
• "What counts as a scope change, specifically?" Vague answers here are how fixed quotes quietly become T&M-in-disguise through change-order creep. A vendor should be able to give concrete examples from past projects.
• "How will I know if a T&M project is running over before the invoice tells me?" Regular, proactive check-ins against a budget — not a surprise at month's end — is what separates a well-run T&M engagement from an open tab.
• "What's your actual track record on estimates?" Ask directly. A vendor with a real, defensible answer has thought about this; one who's never tracked it probably hasn't been held to their own numbers before.

Why the delivery-or-discount guarantee exists

West Fork Digital quotes fixed price on essentially every engagement, backed by a specific, written term: delivered on time, or you pay less. The delivery date is set at kickoff and written into the contract. If we miss it for any reason that isn't a client-side delay or a signed scope change, the price drops 5% for every week we run late — automatically, no argument, no fine print.

That guarantee only works because it's paired with real discovery up front — a scoped plan, in writing, before the number is fixed — and weekly demos throughout, so slippage is visible in week two instead of discovered in week ten. A fixed price without those two things is just a number. With them, it's a genuine transfer of risk, backed by a term you can actually point to if it's ever tested.

The honest bottom line: fixed price protects the buyer when the scope is genuinely knowable, T&M is the fairer instrument when it genuinely isn't, and the vendor's willingness to be specific about overruns — in either model — tells you more than the pricing model itself does.