Denshin / Blog / Process
Fixed price or time and materials: how we actually decide
Fixed price moves the risk to the studio, which prices it in. Time and materials leaves it with you. How we decide between the two, and the split we recommend for most new products.
Denshin Team · Delivery · 21 October 2026 · 5 min read
Almost every client conversation reaches the same fork. They want a number they can take to whoever signs the cheque, and we want a way of working that does not punish either side for what nobody knows yet. Those two wants pull in different directions, and the contract model is where the tension gets settled.
There are two classic options. Fixed price: we agree a scope and a number, and we deliver that scope for that number. Time and materials: you pay for the hours actually spent, at an agreed rate, and the scope can move. Most agencies have a favourite and will tell you theirs is better for you. We use both, plus a third shape that we end up recommending more often than either.
Who carries the risk
The useful way to think about it is not cost, it is risk. Every project has some amount of unknown work in it. The contract decides who pays for that unknown work when it turns up.
Under fixed price, the studio carries it. That sounds good for the client, and sometimes it is. But a studio that carries risk prices it in. The fixed quote you receive is the estimate plus a buffer for everything that might go wrong, and if nothing goes wrong, you paid for the buffer anyway. The other cost is quieter: fixed price makes every change a negotiation. You see the first working screens in week three, realise the approval flow should have two steps instead of one, and now there is a change request, a revised quote, and an email thread about whether this was "in scope". Nobody enjoys that, and it teaches clients to stop asking for the changes that would make the product better.
Under time and materials, the client carries it. You pay for what was actually done, there is no padding for risks that never showed up, and changing your mind is cheap in process terms. The cost is that you do not know the final number, and you have to trust that the hours are being spent well. Some clients are comfortable with that. Many, reasonably, are not, especially the first time they work with us.
When each one fits
Fixed price works when the scope is small and genuinely known. A marketing site with a defined set of pages. A specific integration with an API we have used before. A second phase of a product where the first phase taught us how the codebase and the client both behave. In those cases our estimate is narrow, the buffer is small, and a fixed number is honest.
Time and materials works when the project is mostly discovery. A product that has not found its shape yet, an internal tool where the real requirements will only surface once people use it, work on a legacy system nobody has documented. Fixing a price on those means either padding heavily or eventually having an uncomfortable conversation, usually both.
The trap is a project that looks like the first kind and is actually the second. "A booking app, like the ones that exist already" sounds well defined until you learn about the cancellation policy that varies by season and the agent commissions that are calculated three different ways. We try to find those things in the first conversation, which is why we ask so many annoying questions before quoting. Our approach to that is written up in how we estimate when nobody really knows.
The shape we usually suggest
For most new products we propose a split. A short discovery phase at a fixed price, typically a couple of weeks, that ends with a written scope, wireframes or a clickable prototype, a data model, and an estimate graded by confidence. You own all of that whether or not you continue with us.
After that, the build runs as time and materials in fixed-length sprints, with a cap per sprint and a running total you can see. Each sprint has a planned set of work agreed at the start. If we finish early, the spare time goes to the next item on the list. If something blows up, we tell you the same day and you decide whether to spend more, cut something, or move it to a later phase.
This gives the client a number to plan around and a hard ceiling per period, while keeping changes cheap. It lets us price honestly without hiding a large buffer. And because the discovery phase is fixed and small, it doubles as a low-risk way for both sides to find out whether we work well together before committing to the expensive part.
Whichever model you pick
Write down what "done" means for each piece of work. Not a paragraph of intent, but the behaviour you will check before accepting it. Most fixed-price arguments are really arguments about this.
Agree how changes are handled before the first one happens. On fixed price, that means a simple change request with an estimate and your sign-off. On time and materials, it means the backlog is yours to reorder, and we tell you what moving an item costs.
Ask to see the hours. On time and materials you should get a breakdown you can read without a decoder ring. If a studio is reluctant to show you where the time went, that tells you something.
And whatever the model, budget for life after launch. Neither contract covers the second month, and the second month is when real users show up with real requests. We wrote about that in what a SaaS MVP actually costs.
Tags: Pricing, Contracts, Project Management, Process
All posts · Work with Denshin