Why I bill by the hour, even on fixed-scope projects

Most clients I work with ask for a fixed price up front. It is the default, and it sounds responsible — you know what you will pay, the supplier carries the risk. In nearly every project where I have agreed to a fixed price, both sides ended up worse off than if we had used the model I actually prefer: hourly billing with an explicit budget ceiling and weekly invoicing.

Here is why.

Fixed price does not remove risk — it hides it

When a supplier quotes a fixed price, they are pricing in three things: the work itself, the cost of all known risks, and a fudge factor for unknown risks. The fudge factor on a typical SMB project is 25–50 %. You are paying for it whether or not those risks materialize.

If the risks materialize, the supplier has to deliver the work anyway — and that pressure usually shows up as quality cuts, scope-policing, or both. The client did not actually buy certainty; they bought somebody else’s anxiety.

The incentive trap

A fixed price aligns nobody’s interests with the actual goal:

• The supplier wants to ship as fast as possible to protect margin.
• The client wants to add scope before sign-off, while it is “still free”.
• Everything that is not in the original scope becomes a fight.

What you want instead is: the supplier wants to ship the right thing fast, and the client gets to redirect scope as new information comes in. That is the opposite of fixed price.

What I actually do

Every engagement starts with a written estimate as a range (“between 70 and 110 hours, most likely around 85”). The contract sets:

• An hourly rate.
• A budget ceiling I cannot cross without written approval.
• Weekly invoices with an itemized hours log.
• A standing one-hour weekly review where we re-rank the backlog.

The client always knows what they have spent, what is left, and what comes next. If a feature turns out to be twice as expensive as estimated, we have that conversation in week two, not in week six — and we decide together whether to do it differently, drop it, or extend the budget.

When fixed price is actually right

There is one case where fixed price is unambiguously the right model: the work is small, well-bounded, and the supplier has done it five times before. A logo. A landing page from a tight template. A one-off data import.

For anything where novelty plays a role — and any custom software project does — fixed price is the wrong tool.

What clients gain

The thing clients fear about hourly billing is “the bill never stops.” That fear is well-founded if you do not set a ceiling and if you do not get weekly visibility. With both in place, hourly is in practice more predictable than fixed price, because surprises surface early and cheap.

What they actually buy with hourly:

• The right to change their mind about scope without renegotiation.
• A supplier who has no reason to skimp on quality to protect margin.
• A line-by-line record of what was built and how long it took, which is the cheapest possible audit trail.

> estimate
{
  range:      "70–110 h",
  most_likely: "85 h",
  ceiling:     "120 h (hard cap, written approval to cross)",
  invoice:     "wöchentlich, Donnerstags",
  review:      "freitags 30 min, gemeinsam"
}

The clients who initially resist hourly are usually the ones who, after the first project, refuse to work any other way.