Design for Manufacturing for Founders

Design for manufacturing — DFM — is one of those phrases that gets used constantly in product development circles and almost never explained in a way that's useful to someone who isn't already an engineer.

If you've worked with a product designer or done any research on manufacturing, you've probably encountered it. Maybe your designer mentioned a "DFM review." Maybe you read somewhere that DFM can dramatically reduce production costs. Maybe a factory kicked back your files and said they weren't "DFM compliant" without explaining what that actually meant.

This post is the plain-English version. What DFM is, why it matters more than almost any other single decision in your product development process, what the most common mistakes look like, and — importantly — what you need to understand as a founder versus what you simply need to trust your designer on.

You don't need to become an engineer. You do need to know enough to tell whether the designer you've hired is actually thinking about this.

What DFM is — in plain English

Design for manufacturing is exactly what it sounds like: designing a product so that it can actually be manufactured — consistently, at the quality you need, at a cost that makes sense.

That sounds obvious. Of course a product should be designed so it can be built. But in practice, many products are designed for how they look, not how they're made. The designer creates something beautiful in CAD. The render looks great. The client approves it. Then the factory gets the files and the problems start.

The gate location creates a visible weld line right on the front face. The wall thickness varies in a way that causes sink marks in the final part. The tolerances on a critical dimension are tighter than the manufacturing process can hold reliably. A snap-fit feature that looked clean in CAD requires a two-part mold action that doubles the tooling cost. None of these are catastrophic failures — but each one is a cost, a delay, or a quality compromise that a DFM-aware design could have avoided.

DFM is the discipline of catching those problems during design, before tooling is cut and money is committed. It's the conversation between what you want the product to be and what's practical to build.

Why most of your manufacturing cost is locked in before production starts

This figure surprises almost every first-time founder. By the time a design is finalized and approved for production, most of the product's manufacturing cost is already determined — locked in by choices made at the design stage. Commonly cited estimates run as high as 70%, but the exact share varies by product.

Material selection drives cost. Wall thickness drives cost. The number of components and how they assemble drives cost. The surface finish drives cost. The tolerances on every critical dimension drive cost. All of these decisions happen during design. After they're made, the factory executes against them. The factory can be efficient or inefficient, but they can't change the cost structure that the design established.

This is why DFM review early in the process is so much more valuable than DFM review after the design is locked. When you're still in sketches and early CAD, a single comment — "if you move that rib, you can eliminate a side action in the mold" — might save thousands of dollars in tooling and take minutes to implement. After the mold is cut, the same change can mean paying for a tool modification and waiting weeks for it, plus whatever the factory charges for the change order.

The earlier in the process you're thinking about how the thing gets made, the less expensive the whole project is.

The most common DFM mistakes first-time founders make — and their designers let happen

These are the patterns that show up repeatedly, across product categories.

Choosing materials last

The order most founders follow intuitively: design the product, then figure out what to make it from. It feels logical — you're designing the form, and the material is a finishing detail.

In reality, the manufacturing process and the material need to be selected alongside the design, not after it. Different materials behave differently in different processes. A form that works beautifully in compression-molded silicone requires a completely different approach in injection-molded TPU. The wall thicknesses that work in aluminum die casting will cause sink and warp in polypropylene. The surface finish that's achievable on a CNC-machined part may be impossible on a rotomolded one.

When material selection is deferred, you often end up with a design that was conceived for one material and forced into another — and the product suffers for it in ways that are visible to the customer.

Designing for aesthetics without designing for assembly

A product that looks perfect in a render can be an assembly nightmare in production. Parts that require three hands to align. Fasteners that are hidden until you need to service the product. Snap fits that require precise orientation to engage but provide no alignment feature to help the assembler find that orientation on the 500th unit of the day.

Assembly time is real money. A worker who takes four minutes to assemble your product instead of two minutes doesn't sound like much until you're at 5,000 units. Design choices that make assembly consistent and repeatable — alignment features, snap-fit geometry, clear part orientation, minimized fastener count — pay for themselves at production volumes.

Ignoring wall thickness and draft angles until the factory flags them

These two issues are responsible for an enormous share of first-sample problems.

Wall thickness: In injection-molded plastics, uniform wall thickness is a fundamental principle. Sections that are significantly thicker than the rest of the part cool more slowly, which causes sink marks on the visible surface and internal stress in the part. Extremely thin sections may not fill completely. A good DFM review calls out every wall thickness issue before the mold is built. A design that doesn't account for this will produce bad parts.

Draft angles: For any part that's molded, cast, or formed in a tool, surfaces that are perfectly vertical create friction that resists ejection. Draft angles — a slight taper on vertical surfaces — allow the part to release cleanly. Typically 1–2 degrees per side, sometimes more depending on texture and depth. Missing draft is one of the most common reasons a first sample comes back with surface defects or pull marks. It's completely avoidable if the designer is thinking about it during modeling.

What you need to understand vs. what you just need to trust your designer on

Here's the practical part. You're a founder, not an engineer. You don't need to become fluent in GD&T or memorize the shrinkage rates for every thermoplastic. But you do need to know enough to evaluate whether the designer you've hired is doing their job.

What you need to understand:

That DFM exists and that your designer should be doing it — not as a separate paid step, but as part of how they work
That material selection and manufacturing process should be decided during design, not after
That the decisions made during design lock in most of your manufacturing cost
That a DFM review from a factory is different from DFM built into the design — and that the former is much less useful than the latter

What you need to be able to ask:

"What manufacturing process are you designing this for, and what constraints does that create?"
"Have you worked with this material and process before?"
"What DFM considerations have come up in this design, and how did you address them?"
"Can you show me where there are tooling implications in the current design?"

These questions don't require technical expertise to ask. They require judgment to evaluate the answers. A designer who's thinking about manufacturability will have real, specific answers. A designer who isn't will give you vague reassurances.

What you can trust your designer on: The specifics. Exact draft angles, precise wall thickness specifications, tolerance callouts, mold flow considerations, gate location tradeoffs — these are technical decisions your designer should be making and should be able to explain in plain English if you ask. The way you verify that trust is through the questions above and through their portfolio. Designers who build products that actually get manufactured and come back from production without surprises are designers who think about manufacturability.

How to tell if a designer is thinking about manufacturability — or just making renders

In the work itself: A DFM-aware designer's CAD models will have draft on molded surfaces, consistent wall thickness, minimal undercuts, and snap-fit features that are designed to assemble, not just look right. If you can share files with a factory for a preliminary review before committing to a studio, their feedback will tell you whether the designer understands manufacturing.

In the conversation: A designer who asks about your target manufacturing process, your cost targets, and your production volume early in the engagement is thinking about DFM. A designer who starts with moodboards and aesthetic direction and never asks about manufacturing is probably not. Both might produce beautiful renders. Only one will produce a design you can build.

In the portfolio: Look for products that actually exist in the world. Shipping product is the evidence. If a portfolio is all renders and concept models, you don't know whether those designs were ever validated by a factory. If the portfolio includes physical products — photos of real things held in real hands — that's a meaningful signal.

In the process: Good DFM isn't a separate meeting or a line item on a proposal. It's how the design work gets done. If a designer describes DFM as an add-on phase that happens after the design is approved, that's a process structure that will cost you money.

What a DFM review actually looks like in practice

When DFM is built into the design process — as it should be — it's not a dramatic event. It's an ongoing conversation.

In early concept work, it's the designer choosing a manufacturing process that fits the product requirements and budget, and making form decisions with that process in mind. It's asking "is this shape achievable in this material and process, and at what cost?"

In detailed design, it's the CAD modeling that applies draft, maintains wall thickness, designs assembly features, and calls out tolerances that the manufacturing process can actually hold. It's the note in the design review that says "this undercut creates a side action in the mold — here's the tradeoff between keeping it and removing it."

Before the factory handoff, it's often a formal review: the designer walks through the design with the factory, the factory flags any remaining concerns, and those concerns are resolved before tooling begins. This review is much less expensive than discovering the same issues after the mold is cut.

What it isn't: a post-design checklist that runs through after the creative work is done. That's better than nothing, but it's not DFM built into the design. Changes at that stage are more expensive and more disruptive than changes made during design.

DFM is built into every engagement at Sunfish Labs — it's not a separate service or an optional add-on. If you want to talk through a specific design challenge or get a read on whether your current files are manufacturing-ready, get in touch.