How to price a freelance development project: 4 models ranked

I've used all four. Each one is correct for a specific shape of work. Here's the ranked breakdown, with the projects I'd use each for.

1. Fixed-price (default for most product work)

You quote a number. You deliver the scope. Time spent is your problem.

When it works: well-defined scopes (a landing page, a focused MVP, a website redesign). The client and you agree on what "done" means. You have done similar projects before so your estimate is realistic.

The catch: bad estimates eat your margin. Mitigate by adding 30% buffer on anything you've never built before, and being clear in writing about what is and isn't in scope.

This is my default. 70% of my work, almost always at the highest effective hourly rate.

2. Hourly (for unclear or exploratory work)

You charge by the hour. The client gets a weekly time report.

When it works: scope genuinely cannot be predicted ("help us figure out why our analytics is broken"). Long-term consulting where you advise more than build. Ongoing improvements to a system that the client owns.

The catch: caps your earnings to time. Also caps the client's appetite — they will say no to good ideas just to keep the hour count down. Best used as a complement to fixed-price work, not the main model.

3. Value-based (for high-leverage outcomes)

You charge a percentage of the value created, or a fixed price calibrated to that value.

When it works: when you can measure the outcome (a landing page tied to conversion rate, a backend optimization tied to infrastructure cost). When the client trusts you with their numbers.

The catch: requires trust both ways, and a measurement period that delays your invoice. Most realistic for ongoing work with an existing client, not first engagements.

Value-based pricing sounds great in books and is genuinely hard to start with. Try it on your second project with a returning client, not your first one with a stranger.

4. Retainer (for trusted ongoing work)

A monthly fee for a fixed allotment of hours or a fixed deliverable cadence.

When it works: a returning client with continuous needs. You want predictable income. They want predictable access. Maintenance contracts, monthly improvements, "fractional CTO" arrangements.

The catch: clients overestimate how much they'll use the retainer in month one and underestimate in month four. Build in roll-over rules and quarterly reviews to avoid resentment.

How I'd structure a freelance practice in 2026

• 60% fixed-price project work (the breadwinner)
• 20% retainers (the floor)
• 15% hourly consulting (the filler)
• 5% experimentation with value-based on a returning client

Don't pick one model and force it on every project. Pick the one that fits the shape of the work.