No investor believes your year-three revenue number, and they are not asking for it because they do. Financial projections for a fundraise are read as an X-ray of your thinking: do you understand where revenue actually comes from, what it truly costs to build and sell the product, and when the cash runs out? A well-built model signals a founder who knows their own business. A model that says "we will capture one percent of a large market" signals the opposite, and does it in one slide. This guide builds the model piece by piece, with the emphasis on physical products — where the cash mechanics are harsher than software founders expect.

The Three Parts Every Model Needs

  • Revenue forecast. How many units, at what price, through which channels. Monthly detail for the first year or two, quarterly after that. Anything beyond a few years is a shape, not a number.
  • Cost structure. Direct cost of goods kept strictly separate from operating expense — payroll, engineering, marketing, certification and testing, facilities.
  • Cash flow. The statement investors read first. When money leaves, when it arrives, and the month the bank balance hits zero. Your funding ask is derived from this, not chosen before it.

A profit-and-loss statement that looks healthy while the cash flow statement goes negative in month seven is the most common failure mode in hardware modeling. Cash timing, not margin, is what kills physical-product companies.

Build Bottom-Up, Never Top-Down

The classic error is the market-share calculation: a large addressable market multiplied by a small, arbitrary percentage. Investors recognize it instantly, and it tells them you have not done the work. A credible forecast is assembled from the bottom: how many qualified leads each channel produces per month, what fraction convert, what it costs to acquire a customer, what growth rate each channel realistically sustains. Multiply small, reasoned assumptions together and you get a number you can defend under questioning — which is the actual test.

Every assumption needs a source: results from a small paid campaign, a pilot deployment, a preorder page, pricing observed on comparable products. Whatever traction you have collected is the factual foundation of the revenue line, so build the model on top of it rather than beside it. The pricing logic deserves its own scrutiny too — see how to price a product.

Hardware Reality: The Costs Founders Forget

A physical product carries a cost structure that a software-shaped spreadsheet cannot represent. Diligence exposes the omissions quickly. Make sure yours includes:

  • Fully landed unit cost. Not just factory price — packaging, freight, duties and tariffs, incoming inspection, plus a realistic scrap and warranty-failure allowance. Build it from a real bill of materials, not a placeholder.
  • Non-recurring engineering and tooling. Injection molds, fixtures, test jigs, and regulatory certification are large sums that land entirely before the first sale.
  • Inventory and working capital. Your contract manufacturer wants payment months before your customer pays you. That gap lives only in the cash flow statement, and it is the line most often left out.
  • Minimum order quantities. Your first purchase order is sized by the factory's floor, not by your forecast — see minimum order quantity. Model the inventory you will actually be holding.
  • Second-run engineering. Almost no product ships its first build unchanged. Budget a revision.

For the full catalog of line items that go missing, see hidden costs of hardware development.

From Model to Funding Ask

The cash flow statement gives you runway, and runway gives you the ask. Convention is to raise enough to fund something like a year and a half to two years of operation — and, more importantly, enough to reach the next milestone that raises valuation: a validated prototype, first revenue, a regulatory clearance, a signed distribution agreement. An amount derived from milestones is far more persuasive than a round number that sounds impressive, and it feeds directly into the valuation conversation covered in how to value a pre-revenue startup.

Include a clear use-of-funds breakdown: what fraction goes to engineering, tooling, inventory, certification, and go-to-market. Investors read use of funds as a proxy for operational judgment.

What Goes in the Deck, What Goes in the Data Room

The pitch deck gets the summary only — a revenue trajectory, the milestones, the use of funds, and the ask. Nobody wants a spreadsheet on a projector; see investor pitch deck for a physical product. The full workbook, with assumptions on a visible tab and formulas that a stranger can trace, belongs in the investor data room for diligence. Keep one canonical file, version it, and make sure the number on the slide matches the number in the model — mismatches between the two are noticed and remembered.

Three Things That Destroy Credibility

The unexplained hockey stick

Growth that explodes in year three with no mechanism behind it. If a curve bends sharply upward, something specific must cause it — a channel coming online, a distribution deal, a cost step-down at volume. Name the cause or flatten the curve.

Profitability from month one

Investors know hardware burns cash early. A model that shows immediate profit does not look impressive; it looks naive, and it undermines every other number on the sheet.

Assumptions with no source

A modest model where you can defend every line beats an impressive model that collapses on the second question. Practical tip: build a pessimistic and an optimistic scenario next to your base case. When an investor asks "what if sales slip two quarters," having a prepared answer is worth more than the answer itself.

Grounding the Cost Side in Real Engineering

The revenue side of a model is judgment. The cost side does not have to be — unit cost, tooling, and schedule can be estimated from actual design and manufacturing work. That is where Projects House fits: translating a product concept into an engineering budget and timeline you can put in front of an investor and defend. Note that this is general educational guidance on model structure, not financial or accounting advice; have a qualified adviser review anything you present to investors. More tools for this stage are in our startup fundraising guide.

Want the cost and schedule assumptions in your model based on real development and manufacturing numbers? Send us your project through the contact form and we will come back with a grounded engineering estimate you can build the model around.