By Bryan Knipe, founder of TriPinnacle Advisors
Most company financial models are a forecast nobody believes. Revenue goes up and to the right because someone typed a growth rate, expenses are last year plus a percentage, and the cash line is whatever falls out. The board nods at it, the CEO ignores it, and decisions get made somewhere else.
A model a board can act on is a different thing. It explains how the business makes and spends money, shows what happens under different assumptions, says how long the cash lasts and ties the capital plan to the milestones that matter. I have built hundreds of forward-looking models: for new businesses and acquisitions at Optum, for the operations and finance I ran as COO/CFO, and since 2023 for life-science companies raising money, including the models behind more than $4M raised for a therapeutics company. This is how I build them.
1. Build from drivers, not plugs
Every important number should come from something you can observe and argue about. Revenue is customers or units times price, and customers come from a pipeline with conversion rates and sales cycles. Costs come from headcount by role and date, vendors and the cost to deliver each unit. When the board asks "what if sales take three months longer?", you change one driver and the whole model answers.
The biggest driver in most companies is people: who you hire and when. A hiring plan by role and month usually explains most of the spending and most of the risk.
2. Show scenarios, and what would trigger each
One forecast is a guess. Three scenarios are a plan: a base case you are managing to, a downside that assumes the things most likely to go wrong do go wrong, and an upside. The useful part is writing down what would tell you which scenario you are in, and what you would do in each. A downside case with a pre-agreed response, such as slowing hiring and extending runway, is worth more than a precise base case.
3. State the runway in months and dates
Every model should answer, on its first page: how many months of cash do we have, on what date do we hit the minimum cash we are willing to hold, and when do we need to start raising to avoid it? Raising takes longer than everyone expects, so the date to start is usually six to nine months before the cash runs short.
4. Tie the capital plan to milestones
The right amount to raise is the amount that gets the company to the next milestone that makes it more valuable, plus a margin for delay. In life sciences those milestones are concrete: a regulatory submission, a clinical readout, a first commercial customer. Raising enough to get halfway to one means raising again at the same value. The capital plan should show each milestone, what it costs to reach and the funding behind it, including equity, debt, grants and other non-dilutive money, with timing.
Non-dilutive funding deserves its own line. Grant programs like SBIR and STTR can fund real milestones, but they come with their own timelines, so the model should show when the money arrives, not just how much.
5. Get the unit economics right
For any business selling a product or service, the model should show what it costs to win a customer, what it costs to serve them and what they are worth over time. Unit economics that don't work at small scale rarely fix themselves at large scale, and investors know it.
6. Keep it readable
The best models have one summary page a board member can read in two minutes: revenue, spending, cash, runway, the key drivers and the scenarios. Everything else supports that page. Assumptions sit in one place, clearly labeled, with their source. If only the person who built the model can explain it, it won't be used.
7. Check plan against actuals every month
A model earns trust by being compared to what actually happened. Each month, the actuals from the books go next to the plan, the differences are explained and the forecast is updated. Your accountant or bookkeeper owns the actuals; the CFO owns the plan and the explanation. Over a few months, the comparison shows which assumptions were wrong, and the model gets better.
Strategic CFO, not accountant
This is the line I draw in my own fractional CFO work. Bookkeeping, the monthly close, tax and audit stay with your accountant. My job is the plan: the model, the scenarios, the capital plan and the board conversation that uses them. The two work best side by side, the books telling you where you are and the model telling you where you are going.
If your next raise, board meeting or major decision depends on a model nobody fully trusts, that is the place to start.