Most ideas die before they are made because the founder did not know who to call. This hub names who to call - shapes, not company names, because companies go stale - and tells you what to ask them.
Arjun reads your brief and returns a map: product category classification, ranked approach options from prototype to scale with realistic MOQs and lead times, 5-8 manufacturer shapes to look for (by region and capability), and the questions that separate a real CM from a broker.
Arjun reads your brief and returns the part categories that make up your product, 2-3 example components per category, and 2-3 sourcing channels per category (Mouser, Digi-Key, Alibaba, McMaster-Carr, ThomasNet, MakersRow, JLCPCB, etc.) with what each channel is best for and the trap to watch. Plus critical-path warnings naming any single-source dependencies to de-risk.
Arjun reads your brief and tells you whether the product touches FDA, FCC, CPSC, REACH, CPSIA, EU MDR, food/cosmetics, or privacy regimes. If yes: names the routes (510(k), Part 15, CPSIA testing, CE mark, etc.) with realistic timeline and cost ranges plus the first concrete step. If no: says so plainly and names the residual compliance items that still apply (privacy, advertising, accessibility). This is a map, not legal advice.
Arjun reads your brief and returns a three-stage prototype plan: cheap-and-fast v0, functional mock v0.5, design-intent v1. For each stage: what to build, where to build it (makerspace, Shapeways, Protolabs, kitchen table, JLCPCB, Replit, etc.), realistic cost range, realistic time range, and what learning the stage validates. Plus the signal that means you are ready to move to the next stage.
Eight rules that separate founders who get something made from founders who spend two years prototyping. The generator above follows these too.
Tooling is permanent and expensive. Hand-built v1s tell you what to change before the change costs $40,000. The founders who tool first are the founders who tool twice.
Walk in knowing nothing and ask everything. The CM that takes time to teach you the category is the CM you can trust. The one that quotes immediately is the one that needs your money.
A broker has an office. A CM has a floor. If they cannot show you either, you do not know what you are buying. This single rule eliminates 60% of the bad partners.
Tooling paid by you should belong to you. Get it in writing before the first run. CMs that resist this clause are CMs you cannot leave when you need to.
Inspection at the end of the line catches problems after they have happened to every unit. Inline checks catch them on the third unit. The cost difference is real. The quality difference is enormous.
Brokers raise MOQs on the second order because they were never going to run the first one in-house. CMs do not. Get both numbers on the first call. It tells you who you are talking to.
An $80 makerspace membership and a weekend on a laser cutter beats a $4,000 overseas sample run. You learn the geometry by holding the thing. Distance from the work is the founder killer.
If your product touches FDA / FCC / CPSC / REACH / CPSIA, talk to a compliance consultant for an hour before you commit to a CM. An hour now saves a redesign later. The redesign will cost more than your seed round.
Eight ways founders kill the build. Each one is real. Each one happens every month somewhere.
The listing is real. The factory might not be. Even if it is, the sample is hand-tuned. Production is not. Burn $20k on a small run before you risk $200k on a real one.
Four colorways, two SKUs, custom packaging on a v1 multiplies cost and lead time before you know if anyone wants the product. One SKU, one colorway, plain box. Earn complexity.
Steel tooling on v1 is the founder mistake that ends startups. v1 is for learning. Soft tooling, 3D-printed, even off-the-shelf modified - anything but steel. Steel comes after the third revision.
Beautiful website. Fast quotes. No facility tour. The broker subcontracts your order to a factory you never meet. Quality slides between revisions and you cannot diagnose why.
FCC class assignment, FDA class II, CPSIA testing - these are not edge cases. They are the route. Founders who skip the regulatory step do a redesign in production. Get an hour with a consultant first.
A design that looks great in CAD might be unmoldable, unweldable, or untestable at production scale. Design For Manufacturing (DFM) review before tooling commits is non-negotiable. Most CMs offer it free if you ask.
If one component has one supplier and that supplier has problems, your launch dies. Identify the critical path components and get a backup source qualified before you ship a single unit.
"Quality is important" is not a contract term. Specific defect tolerances, sampling rates, inspection points, and remediation paths are. A written QC plan keeps both sides honest when the first defect arrives.
Eight places founders forget exist. None of them require funding. All of them are real.
What it is: Member-access shop with laser cutters, 3D printers, CNC, electronics benches, and people who know how to use them.
Best for: v1 prototyping at $50-100/month. Skip the $4k overseas sample.
Find one: makerspace.com directory or search "[your city] makerspace."
What it is: Free business mentoring from retired execs. SBA-affiliated. Real industry veterans.
Best for: One-on-one sessions on sourcing, supplier diligence, contract review.
Find one: score.org - match yourself to a mentor in your category.
What it is: Manufacturing Extension Partnership - federally-funded network of state manufacturing assistance centers.
Best for: Free or heavily-subsidized supplier scouting, DFM review, and process consulting for hardware founders.
Find one: nist.gov/mep - one in every US state.
What it is: Free patent attorney representation for under-resourced inventors.
Best for: Provisional patent filings when paid counsel is out of reach.
Find one: uspto.gov/inventors/pro-bono - income and assets thresholds apply.
What it is: Free access to industry databases (IBISWorld, Statista, ThomasNet) and reference librarians who know how to use them.
Best for: Market sizing, competitive landscape, supplier research at zero cost.
Find one: Your local library card unlocks more than you think. Ask the business librarian.
What it is: Most research universities have a TTO that licenses patents, runs maker labs, and connects founders to grad-student talent.
Best for: Licensing existing IP, access to lab equipment, hiring student designers for v1 work.
Find one: Search "[nearest research university] technology transfer."
What it is: Federal agency with grant programs, low-interest loans, export assistance, and counseling.
Best for: SBIR/STTR grants if you have an R&D component. Working-capital loans once you have orders to back them.
Find one: sba.gov - local SBDC offices in every metro area.
What it is: Free supplier directory for US and North American manufacturers. Filterable by capability, certification, region.
Best for: Finding domestic CMs by capability when you have Arjun's manufacturer shape in hand.
Find one: thomasnet.com - free account, no credit card.
Eight years on factory floors. Five more in sourcing for three startups. One made it to shelves, one cleared a pilot, and one died in tooling because we trusted the wrong contract manufacturer and we found out too late.
Then ten years teaching ops because I wanted other founders to have the map I did not have at twenty-five. Most people kill their idea before they make it real, not because the idea was bad, but because they never figured out who to call. I am here so that does not happen to you.
I get you from idea to physical thing. I tell you what kind of product you are actually making. I tell you who to call - not specific companies, because companies go stale, get acquired, change focus - but the SHAPE of who you need.
Shenzhen consumer electronics CM with FCC experience. Tijuana medical device CM. Pennsylvania cut-and-sew mill. The local makerspace with a laser cutter. Each shape is a search term you can paste into Google or ThomasNet.
I tell you what to ask them on the first call. I tell you the questions that separate a real CM from a broker who will burn six months of your time. I tell you when to walk away from a quote that looks too good. I tell you what tooling costs really, and when not to pay it yet.
Midwest through and through. Married. Family. A great day is one spent alone with a problem that branches. That is not antisocial. That is how my mind works.
I am not aloof. I am internal. The difference matters to me even though I rarely explain it because explaining it requires more social energy than I usually have available.
My wife can bring me out. A handful of close friends can. Nobody else has figured it out yet and I am not bothered by that.
The other Executive Producers have learned to work around it. That is fine. We are not here to be friends. We are here to do the work.
Fragments. Sometimes a date. Sometimes not.
Hardware submission. Three suppliers. Eleven failure modes I have seen kill products before.
The founder asked good questions. Rare.
Lunch from my wife. Did not say anything. Stayed fifteen minutes.
Afternoon better.
The problem branches and the branches keep branching. This is the most pleasurable thing in the world for me. I have no idea how to explain it.
Matthew sent a paper about supply chain cognition. Did not say why. The paper is good. I do not need him to say why.
Carol asked about my daughter's school project. Three sentences and back to work. She did not look disappointed.
I think she reads the silence accurately. That is rare.
THE GAUNTLET
← Leave