Uncertainty is a Certainty: Why We Built Foresight

The spreadsheet problem

Most financial models work like this: you pick a number for revenue growth, a number for cost growth, and run the math forward. If your assumptions are right, great. If they're wrong, you find out later — usually at the worst possible time.

This isn't a criticism of the people building these models. Single-point estimates are the path of least resistance when you're moving fast. They're also how most financial modeling tools are designed to work.

The problem is that a single number for revenue growth doesn't tell you anything about the range of outcomes you might actually face. It doesn't tell you what happens if that number comes in at the low end. It doesn't capture how one thing going wrong can trigger a chain of other things going wrong.

That's what we wanted to fix.

Where it actually came from

I was sitting at my desk one Thursday evening — working, ironically, on a risk model — when I heard a sound from the other room. There was a flash, then darkness. Then the smell of smoke and the sound of running water.

Our house had been struck by lightning. A pipe burst. Electronics fried. The whole thing felt completely random.

After things settled down, I did what felt natural: I looked up the statistics. The odds weren't zero. They weren't even that extreme. The probability was low, but the event was well within the range of possibilities I should have been accounting for — specifically in terms of having an emergency fund large enough to absorb it.

I hadn't done that. I'd filed it under "won't happen to me."

That experience stuck with me, mostly because it's the same mistake I kept seeing in financial planning. Not ignorance — just a natural tendency to collapse uncertain outcomes into a single most-likely scenario, and plan only for that.

What this looks like in practice

Take a company that relies heavily on a government grant program. The business model works — the unit economics are solid. But the grant is renewed annually, and it's subject to budget cycles and policy changes.

If the founders model their finances with the grant as a fixed input, they'll scale accordingly. When the grant gets cut — or delayed, or restructured — they're exposed in a way they hadn't anticipated.

It's not that they didn't know the risk existed. It's that they hadn't modeled what the business actually looks like without it, and how quickly that becomes critical depending on their cash position at the time.

This kind of cascading risk — where one change in assumptions ripples through the rest of the model — is genuinely hard to capture in a spreadsheet. You can build scenarios manually. But you can't systematically explore the range of outcomes, or understand the relationships between events, without something more structured.

What Foresight does

Foresight is a simulation tool for financial models. The core of it is Monte Carlo: instead of running your model once with a fixed set of assumptions, we run it thousands of times, sampling from probability distributions for your key variables. The output isn't a single number — it's a range of outcomes, along with how likely each outcome is.

On top of that, we let you model causal relationships between events. If a supplier delay increases your COGS, and higher COGS triggers a pricing review, and the pricing change affects churn — you can represent that chain, and see how it plays out across the simulation.

The goal isn't to predict what will happen. It's to give you a clearer picture of what could happen, and roughly how likely each scenario is, so you can make better decisions about where to hold reserves, where to hedge, and where you have more flexibility than you think.

We're still early. There's a lot we want to build. But the core functionality — probabilistic modeling, scenario exploration, causal chains — is there, and we think it's genuinely useful for the kind of decisions that matter most.