A fixed-price contract built on guessed scope. A T&M engagement without reporting. A dedicated team with no backlog. These aren’t project failures — they’re pricing-model failures that were predictable from day one. The model you choose decides whether estimation errors destroy margin or get absorbed without drama. Most agencies pick one by accident.
Choosing a model starts with knowing what you know
In CTO groups and Reddit threads the same story surfaces every week. An agency signs a fixed-price deal, the scope shifts, and months later they’re explaining to the client why the original number no longer works. Or the reverse — the client wants T&M but has no handle on what the final number might look like, so they walk.
The problem isn’t the model itself. It’s the data underneath it.
Fixed price rewards certainty. T&M rewards transparency. A dedicated team rewards discipline. Each model asks a different question of your estimation process — and estimation, not what’s easiest to sell, should decide which model fits.
Fixed price — works when you actually know the scope
A single number for a defined product. The client knows their budget, you know your margin. On paper it’s the cleanest deal in the business.
In practice it works when you’ve done this exact kind of work before. Same type of module, same stack, same team profile. You’re not estimating from instinct — you’re drawing on data from a hundred similar projects.
Most agencies skip that condition. Without precise scope, documented assumptions, and a real change request mechanism, fixed price is a bet. A bet that you won’t be wrong. Projects have a way of proving you wrong.
Time and materials — when uncertainty is honest
Bill for actual time spent. No fixed number, no pretense of knowing everything upfront. Flexible, fair, uncomfortable for clients who need predictability.
T&M fits where change is part of the deal: discovery phases, MVPs with evolving requirements, fast iterations shaped by real feedback. The client doesn’t commit to a final number — they commit to a process.
Without reporting though, that process looks like a black box. Break the work into visible stages. Show progress in real time. The client needs to see their spending turning into something concrete, or T&M starts to feel like a blank check they’re afraid to cash.
Dedicated team — renting capability, not buying a project
A team at a fixed monthly rate. Developers, PM, QA. Scalable, cost predictable, good for longer engagements. The client gets skilled people without the hiring drag.
The trap is that a loose team burns budget fast. Without constant scoping and prioritization, the monthly rate buys a lot of activity but not much progress. The discipline here isn’t about estimating the whole project — it’s about estimating velocity. How much can three developers actually deliver per month? And does the backlog point them at the right things?
Hybrid — most agencies should start here
Discovery on T&M, fixed price for the clear parts, T&M again for the uncertain bits. It’s not elegant — it’s pragmatic. And pragmatism distributes risk better than any pure model.
The hard part is separating scope into known and unknown buckets. Each needs its own estimation approach and billing rules. Most teams don’t practice that split, which is exactly why hybrid feels uncomfortable until you’ve done it a few times.
Outcome-based — demands the most trust
Paid for results. A conversion lift, a feature launch, a performance target met. The strongest alignment of interests and the most exposure for the agency.
This model needs measurable business outcomes and real influence over achieving them. It also needs precise estimation of the cost to reach the result — if that estimate is wrong, you’re working at a loss. And without hard metrics, “did the outcome happen?” becomes a negotiation.
Most agencies aren’t ready for this model. The ones that are built estimation discipline first.
The thread that runs through all of them
Every model comes back to the same question: how well do you know the project before you commit?
Certain scope lets you choose fixed price and protect margin. Uncertain scope forces T&M or hybrid so you’re not betting your own money. Dedicated team and outcome-based models live or die on your ability to estimate velocity and risk.
Agencies that pick models deliberately — instead of stumbling into them — invest in estimation infrastructure. A component library. A knowledge base of past projects. Benchmarks from similar engagements. Data replaces gut feel. The pricing model becomes a decision based on evidence, not hope.
That’s the shift Apropo is built for. Instead of estimating on paper and hoping the model holds, you get estimation grounded in your own project data. The sanity check compares your proposal against benchmarks from hundreds of engagements. The interactive web proposal lets the client toggle scope and watch the price change in real time — whether you’re selling fixed price, T&M, or hybrid.
Next time you’re choosing a pricing model, start with the data. What do you actually know about this project? The answer determines whether your model protects margin or bleeds it.