Designers and Engineers on One Team

There is a version of product development where the industrial designer produces a beautiful concept, hands it over the wall, and the mechanical engineer spends the next two months explaining why none of it can be manufactured. The designer feels overruled, the engineer feels handed an impossible brief, and the founder pays twice. It is the most common structural failure in a first product program, and it is a process problem rather than a personality one.

The two disciplines are not adversaries with different taste. They are optimizing different, genuinely conflicting variables, and the job of the program is to make those conflicts visible early enough to be cheap.

What Each Discipline Actually Owns

The overlap column is where every argument happens: the exterior surface. It is simultaneously the brand and the molded part. If you have not yet decided whether you need both disciplines, the question of when to bring in an industrial designer is worth answering before you staff the project, not after the engineering is underway.

Where the Friction Really Starts

Draft angle

A designer draws a crisp vertical face because it looks precise. A molded part needs typically 1–3 degrees of draft per side, more on textured surfaces, or it will not eject. Add draft late and every adjoining surface shifts, which visibly changes proportions the designer chose deliberately. Agreed draft rules stated on day one cost nothing; discovered in month four they force a restyle.

Wall thickness and the sink mark

Designers ask for a thick, solid-feeling boss or a chunky corner. Injection molding punishes thickness variation with sinks, warp, and long cycle times. The constraint is not negotiable and is explained in the wall thickness rule that drives every plastic part. Designers who internalize it stop drawing shapes that will fail, and engineers who understand why the designer wanted mass can usually deliver the same perceived heft with a rib pattern and a weighted insert.

Parting lines and gaps

The designer wants the split line where it disappears; the toolmaker wants it where the tool opens cleanly. Both are right. Resolve it jointly on a printed section view, not in email.

Gaps, flush, and stack-up

A designer specifies a 0.5 mm uniform gap between two housing halves. Real parts vary. Whether that gap can hold depends on the tolerance stack-up across the assembly, and the honest answer is often that the gap must grow to 0.8 mm or the parts must be located differently. Better to know at concept stage than at first article inspection.

CAD: Two Toolchains, One Truth

Industrial designers typically work in surface modelers built for continuous, high-quality curvature. Engineers work in parametric solid modelers built for features, constraints, and drawings. These are different mathematics and different intents, and the handoff between them is where geometry gets lost.

What works in practice:

• The designer delivers surfaces, not solids. A clean, trimmed, watertight surface set with defined continuity is what an engineer can thicken, offset, and split. Why that matters is covered in surface modeling that manufactures well.
• Deliver control curves too. If the engineer has the driving curves rather than only the finished surface, they can regenerate the shape after a change instead of asking for a new file.
• Name the master. One model is authoritative at any moment and one named person owns it. Two authoritative models is how a product ends up with a housing that no longer matches its bezel.
• Use STEP for exchange and expect fidelity loss. Neutral formats break parametric history and can degrade tangency. Plan review time to check the imported geometry rather than assuming it arrived intact.
• Version everything with dates and a change note. The single most common cause of a wasted week is an engineer working from a superseded surface set.

Run Them Concurrently, Not in Sequence

The sequential model, where design finishes and engineering begins, is what produces the over-the-wall failure. The alternative costs slightly more in coordination and saves entire cycles.

1. Shared brief. Both disciplines are in the room when the requirements are written, and both sign off. A good design brief states not only the aesthetic direction but the manufacturing process, target cost, and known constraints.
2. Concept phase with an engineering reality check. The designer generates four to six directions. The engineer reviews each for feasibility and rough cost before anyone falls in love with one. Killing an unbuildable concept in week two is free.
3. Package study in parallel. While form develops, the engineer builds a rough internal package: board outline, battery, motor, fasteners, service access. The designer works around a real volume rather than an imagined one. For anything containing a PCB this is decisive, and the constraints are laid out in designing an enclosure around a board.
4. Joint model reviews on a fixed cadence. Weekly, screen-shared, both disciplines present. Every issue gets a decision and an owner in the meeting.
5. Physical models early and often. A printed part settles arguments that renders cannot. Both disciplines should be holding the same object.
6. DFM review before the surface freezes. Bring the manufacturer into the conversation while change is still cheap; the logic is set out in design for manufacturing.

Practices That Prevent the Big Fights

• Define who decides on aesthetics. Usually the designer, unless the change is required for function, safety, or a stated cost target. Write the rule down at kickoff.
• Make constraints explicit and dated. A constraint discovered in month four is a schedule event. Publish the known list on day one and update it visibly.
• Give the engineer the why. "The radius has to stay 8 mm because it is the brand cue across the whole family" gets respected. "Because I said so" does not.
• Budget for iteration. Programs that plan for a fixed number of design rounds and then discover they need more are the ones that blow up. Understanding the stages in the industrial design process helps set that expectation honestly.

Projects House runs industrial design and mechanical engineering as one team on the same program, with a shared model and a single point of accountability, so the handoff problem does not exist. Tell us about your product through our contact form.