Your forecast is not your revenue

Build a spreadsheet that projects where a product could go, and something strange happens to how you read it. The number at the bottom — the annual total, the "if this converts at X, we make Y" — starts to feel like money you have rather than money you've imagined. It's on the screen. It's specific. It came out of a model you built carefully. So it reads as a fact.

It isn't. It's a chain of assumptions multiplied together, and every link in that chain is a guess wearing a decimal point. Traffic times conversion times price times repeat rate, each estimated, each optimistic in the small ways estimates always are. The output looks like a measurement. It's a hypothesis with good production values.

A model is a question, not an answer

There's nothing wrong with the spreadsheet. A projection is a useful thing — it tells you what would have to be true for a plan to work, and it points at which assumption matters most. That's its whole job: to turn a vague hope into a set of testable claims. "We'll do well" becomes "we need a two-percent conversion on five hundred visits," and now you have something you can actually go check.

The failure isn't building the model. The failure is reading it as a report instead of a question. A report tells you what happened. A model tells you what you're assuming. Confuse the two and you start making real decisions — spending, hiring, promising — on the strength of numbers that no event has ever confirmed.

Label the assumptions, or they'll disguise themselves as facts

The cheapest safeguard we've found is almost embarrassingly simple: mark every number as either measured or assumed, and never let the two share a cell without a flag. Measured means an actual event produced it — a sale cleared, a visitor arrived, a real transaction settled. Assumed means you reasoned your way to it. Both belong in the plan. Only one belongs in the sentence "here's how we're doing."

When you keep them visually separate, two useful things happen. Assumed numbers stop quietly inflating your sense of progress, because you can see at a glance how much of the picture is real versus projected. And the measured column — usually small, often uncomfortable — becomes the thing you're trying to grow, which is exactly the thing you should be trying to grow. The forecast is the map. The measured column is where you actually are.

Steer by the measured number

This changes what you do on a slow week. If the projection is your reality, a flat week feels like a betrayal of the plan, and the temptation is to spend your way back onto the curve. If the measured column is your reality, a flat week is just information: the assumptions haven't been confirmed yet, so the correct move is to go get evidence — not to pour budget onto a conversion rate no real buyer has ever demonstrated.

Keep the forecast. Let it inspire the target and expose the assumption that matters. But make decisions on the number an event actually produced. The map is worth having. Just don't mistake it for the ground under your feet — and never fund a route on the map before you've measured a single step of it.