How to Quote Iterative Prototype Work Without Killing Your Margins

Prototype work rarely ends with the first part. In robotics, the first version is often only a starting point. The team machines Rev A, tests it, finds a clearance issue, changes a hole pattern, adjusts a mounting surface, and sends Rev B. A few days or weeks later, Rev C arrives.

This is normal. Robots change during testing because the real world exposes problems that the model does not show. A bracket may flex. A sensor may need to move. A cable may need more clearance. A gripper finger may need a different contact surface. The design improves through use.

For a CNC machine shop, this can be good work or painful work. The difference is how it is quoted and managed. If every revision is treated as a brand-new emergency, the shop spends too much time reviewing files, answering questions, and rethinking setups without being paid properly for that effort.

The first mistake is pricing prototype work as if it were stable production work. A first article carries uncertainty. The drawing may need review. The model may have hard-to-machine features. The tolerances may not match the real function. The customer may need feedback before the job is ready. That time belongs in the quote.

A first article is not just one part. It is the first proof that the design can be made and used. The shop is learning about the part, the customer, the material, and the risk. If the quote ignores that learning time, margin disappears before the second revision arrives.

Revision work should be handled differently. If Rev B is based on a part the shop already made, the quote should not start from zero. The shop should ask what changed. Did the customer move a hole? Change the thickness? Switch material? Add a finish? Increase quantity? A small change may be easy. A small-looking change may also affect the entire setup.

This is why revision history matters. A buyer may say the part is almost the same, but the shop needs to verify what changed. If the shop has notes from the previous job, it can quote faster and more accurately. If everything sits in old email threads, the team wastes time reconstructing the story.

There is also a difference between a revision batch and a pilot run. A revision batch may be five or ten parts for another round of testing. A pilot run may be fifty or one hundred parts for field use or early customers. Those are not the same conversation. A pilot run needs more attention to consistency, inspection, material availability, finishing, and delivery schedule.

Shops should explain this early. Robotics buyers usually understand iteration because they live with it every day. What they may not understand is how much each revision costs the shop in review, programming, setup, and communication. A clear explanation helps avoid frustration.

The wording can stay simple. The shop can say that the first part is priced as a first article, future revisions are quoted against the previous version, and larger batches are priced separately once the design is more stable. That feels fair because it matches how the product develops.

This also helps the buyer plan. A founder may be tempted to order one part at a time because the design is changing. That is reasonable in early testing. But once the team expects multiple versions, the shop can ask whether there is a likely next quantity. If the buyer may need ten parts after the first test, that should be discussed before the quote is finalized.

Shops should avoid giving away too much unpaid engineering time. Helpful feedback is valuable, but it should not become endless free design review. A short manufacturability comment is part of good service. Reworking the customer’s design over several revisions is different. The shop needs boundaries.

The goal is not to make prototype buyers feel difficult. It is to make the process sustainable. Robotics startups need iteration. Machine shops need margin. Both sides can get what they need when the work is separated into stages.

First article work proves the design. Revision work improves it. Pilot work tests whether the part can be repeated at a higher quantity. Quoting all three the same way creates confusion.

A shop that understands this can turn prototype chaos into repeat business. It can support the customer without absorbing every hidden cost. It can also become more valuable to the buyer because it remembers the part history and understands where the project is going.

Iterative prototype work can be profitable. It just cannot be treated like random one-off quoting forever.