When a Startup Should Bring In a Development Firm

Two founders, a prototype held together with a 3D printed bracket, and a spreadsheet that says the first production run costs $340,000. The question on the table is whether to spend the seed money hiring engineers or to spend it with an outside development firm. Get this wrong in either direction and you burn a year: hire too early and payroll eats the runway before there is a product to sell; outsource too much and you end up owning a product nobody inside the company understands.

There is no universal answer, but there are reliable signals. Most of them are about the shape of the work ahead, not about the size of the company.

The five signals that say "now"

1. The next stage needs a discipline nobody on the team has

A founder with a strong mechanical background hits the point where the product needs a custom PCB, a wireless stack, an enclosure that can be injection molded, and a regulatory strategy. Each of those is a career, not a weekend. Hiring four specialists is impossible at seed stage; a firm that already has all four on staff is the only realistic way to move. The tell is that you are reading tutorials on a topic that will appear in one design review and never again — any capability you need intensively for eight weeks and then never is a candidate for outside help.

2. There is a hard external deadline

A trade show, a customer pilot, a grant milestone, a crowdfunding date, a retail buyer's window. Hiring a good hardware engineer in the US takes three to six months from posting to productive contribution. If your deadline is inside that window, hiring is not a plan — it is a wish. A firm starts in one to three weeks and staffs up and down as the phase requires.

3. Design has to become manufacturable

The most common inflection point. A prototype that works on a bench is not a design a factory can build. Someone has to do the DFM pass, write drawings and tolerances, specify materials, select a contract manufacturer, and build the manufacturing data package. Founders who have not done this before underestimate it by a factor of two or three, and the mistakes surface after tooling is cut.

4. You are burning months on rework

Three revisions of the same board, an enclosure that has failed drop testing twice, a mechanism that keeps binding. Rework loops that a team cannot break usually mean a missing analytical capability — FEA, thermal, tolerance stack-up, EMC pre-scan — rather than a missing effort.

5. Investors are asking who is going to build it

Due diligence on a hardware startup always reaches the question of execution capability. A credible engineering partner with a named scope, a schedule, and a fixed budget is an acceptable answer. "We will hire a team after the round" is a much weaker one, because it converts a technical risk into a hiring risk.

What you should never outsource

The failure mode of outsourcing is not cost. It is hollowing out the company. Keep these inside, always:

• The product decision. What it does, who it is for, what it costs, what gets cut. A firm can advise; the founders must decide.
• Customer contact. Every user interview, every pilot visit, every complaint. This is where the product actually gets designed.
• The core novel technology. If your algorithm, chemistry, or mechanism is the reason the company exists, keep the deepest work in-house even if the surrounding engineering goes out.
• IP ownership. Non-negotiable and contractual, not a matter of trust — the terms that decide it are covered in who owns the IP when a company develops your product.
• Enough technical literacy to judge the work. Someone on your side must be able to read the drawings and challenge the tradeoffs, which is a real argument for having a technical co-founder even when the engineering is outsourced.

The cost math, honestly

Founders compare a firm's hourly rate to a salary and conclude the firm is expensive. That comparison is wrong in both directions, so do it properly.

The rough rule: below roughly 3,000 to 4,000 hours of a given discipline per year, a firm is cheaper on a fully loaded basis. Above that, and if the work is continuous rather than phase-shaped, hiring wins. Most seed-stage hardware startups are far below that threshold in every discipline except the one their founders already cover, and the general framing is laid out in outsourcing product development versus building an in-house team.

Do not buy the whole thing at once

The riskiest move a startup can make is signing a single large end-to-end contract with a firm it has never worked with. The safer sequence is to buy a small, bounded piece first and see how the relationship behaves.

1. Definition phase. Requirements, architecture options, cost model, risk list, and a schedule. Usually a few weeks and a modest fraction of the total, and it is the highest-leverage money in the whole project — see what a product definition phase costs.
2. One technical de-risking spike. The single hardest unknown, proven or disproven on a bench.
3. Full development. Only after the first two have shown you how the firm communicates, estimates, and handles bad news.

This staged approach is also what makes hiring a firm for a single phase a legitimate strategy rather than a compromise. And the selection criteria that matter — portfolio depth in your category, who actually does the work, and how change orders are priced — are covered in choosing a product design firm.

The equity question

Cash-poor founders regularly ask whether a firm will work for equity. Some will, partially, and the arrangements usually look like a discount of 20% to 40% off standard rates in exchange for a small equity stake or warrants, with the remaining cash still payable. Very few firms take pure equity, for a simple reason: they have payroll every two weeks and your exit is years away and improbable.

Understand what you are buying. Discounted rates can come with lower priority when a cash-paying client has an emergency, and an outside firm on the cap table complicates later rounds. The tradeoffs are worth reading in full in whether an engineering firm will work for equity before you propose it.

When to wait

Bringing in a firm too early is its own waste. Wait if:

• You cannot describe the product in a page — you will pay engineering rates for people to help you decide what you want.
• You have not talked to twenty potential customers. Requirements written without them get rewritten.
• The core technical question is unproven and cheap to test yourself on a breadboard.
• You have under three to four months of runway. A firm cannot save a company that is out of money; it will simply spend the last of it faster.

In those situations the better use of a small budget is the scrappier path described in building a hardware MVP, followed by a definition phase once you know what you are actually building.

The hybrid that usually works

Most successful early-stage hardware companies end up in the same shape: one or two in-house engineers who own the core technology and hold the institutional knowledge, plus a development firm covering the disciplines that appear in bursts — industrial design, electronics, firmware, regulatory, and manufacturing transfer. The in-house people direct and review; the firm executes. Knowledge stays inside because the internal engineers sit in every review, and the firm's documentation deliverables are contractual, not optional.

Projects House works with startups in exactly this configuration, through a global engineering and manufacturing network, taking a scope from definition through production readiness while the founders keep control of the product and the IP. If you are weighing whether the next phase belongs in-house or outside, describe where you are through our contact form.