Pricing an AI-built client app when the build is the fast part

Barbara·5 min read

Price an AI-built client app by what it changes for the client's business, not by the hours the build took. The build is now the cheapest part of the project. What the client pays for is the discovery that makes the app right, the data and content that make it useful, and the support that keeps it working.

Every freelancer who builds client apps with AI runs into the same question: if the app took an afternoon, how do I justify the invoice? Here is how we think about it.

Hourly pricing punishes you for getting faster

If you bill by the hour, every improvement in your tools lowers your income. The client gets the same app for less, and you get less for being better at your job.

Hourly billing also tells the client what to watch. They start counting hours instead of asking whether the app solved the problem. With an AI builder, that conversation goes badly fast: "the AI did it" becomes the reason to pay less.

The fix is to stop selling the build as the product.

What the client is actually buying

Break the project into the parts that carry value for the client. None of them is "generate the app".

Discovery. Finding the one task that matters. A repair shop owner says they want "an app for the business". After an hour of questions it turns out the problem is that customers keep calling to ask whether their phone is ready. That insight is worth more than any screen you will build, and it is the part the client could not do alone.

The data model. Deciding what the app records and what it leaves out. Which fields on the intake form, what counts as a status, what has to be kept and for how long. Get this wrong and the app is tidy but useless.

Content. Menus, price lists, care instructions, opening hours, the wording on every button. For customer-facing apps, content is most of what people see. It takes the client's time to gather and yours to shape.

Fitting it into the day. Which device it runs on, who opens it first in the morning, what happens when a new employee starts. An app that ignores how the business actually runs gets abandoned quickly.

Training and handoff. Showing staff how to use it, writing down where the app is published and where the exported code is kept, and who can change it.

Support and changes. The menu changes in spring. A new service is added. Someone deletes the app from the shared tablet. This is ongoing value, and it is the most natural place for a retainer.

Turn those parts into a quote

Once the value is visible, the quote follows a structure the client can read:

1Discovery and scope, as a fixed fee. It produces a one-page description of the app, the records it keeps and what it leaves out.
2Build and content, as a fixed fee tied to that scope, including a set number of revision rounds.
3Launch and training, as a fixed fee: publishing, setting up devices, a session with staff.
4Care plan, monthly: changes within an agreed size, fixes, and a response time.

Fixed fees per stage protect you from being paid less for working faster. They also protect the client from open-ended hourly bills. Describe each stage by the work it includes, not by a promised business result. You control the work; you don't control whether the café sells more on Tuesdays.

Charge the most where the client gains the most

Put the largest share of the price where the client's gain is largest. For most small businesses that is discovery and the care plan, not the build.

A cheaper build line is not a discount you are giving away. It is the honest part of the quote, and it makes the rest easier to defend. When a client sees a modest build line next to a substantial discovery line, the message is clear: the thinking is what costs money here.

Avoid justifying your price with what other agencies charge or what a traditional development project would cost. The client didn't ask for a traditional project, and the comparison invites a negotiation about someone else's work.

Be upfront about the tool

Some freelancers hide that they used an AI builder. We think that backfires. Clients tend to find out, and then every line in the invoice looks inflated.

Say it plainly: the app is built with an AI builder, which is why the build line is small and why changes are fast. What they are paying for is your understanding of their business and your responsibility for the result. If the client asks whether they could do it themselves, the honest answer is often "yes, the build". The questions, the decisions and the follow-through are still yours.

When the client wants it cheaper

Some clients will want only the build. That is fine, as long as the scope shrinks with the price. Offer a smaller package with less discovery and no care plan, and say what that leaves out. Never discount discovery and keep the same promises; that is how projects go wrong.

For clients building apps for their own use, our insight on how much it costs to make an app for your own use is a useful starting point for the conversation. For the bigger picture of agency work, read delivering client apps with an AI builder.

Leanfinit is free during the beta, so there is no tool cost to pass on to the client yet. Try a client prototype in the Leanfinit studio.