Most Companies Build a Budget. Almost None of Them Use It.

A founder walked me through his annual budget last month. Twelve months of monthly projections, headcount plan, revenue assumptions by channel, departmental spend broken out by category. It was detailed, realistic, and took his team about three weeks to build.

I asked what decisions changed after last month's board meeting.

Long pause.

What a budget actually is

A budget is a set of predictions. You're saying: given what we know today, here's what we expect revenue, costs, and cash to look like over the next 12 months. That's useful. It forces a conversation about priorities, surfaces resource constraints, and creates alignment on where growth is supposed to come from.

What it doesn't do — on its own — is tell you anything. Predictions only matter if someone is watching whether they came true and acting on the gap.

Datarails raised $70M last month and declared "FP&A software is dead." What they mean is the old model is dead: a static plan built once a year, reviewed quarterly, updated never. Most enterprise companies use FP&A software to automate a broken process. Growing companies don't use FP&A software at all. Both groups are running the same underlying problem.

The variance process most companies don't have

A working budget process has two parts. The first part is building the plan. The second is operating against it.

Operating against a budget means: actuals come in, you compare them to plan, you identify the gaps, someone owns the explanation, and a decision follows. Not every variance needs a decision. Some months you're over on travel because there was a conference. Fine. But revenue 12% below plan for two months in a row is a signal, not a conversation topic. It should trigger a reforecast, a pipeline review, or a change to the go-to-market motion.

Most companies stop at the report. The CFO or controller sends a P&L with actuals vs budget, usually six weeks after the period closed. The leadership team looks at it in a board meeting. Someone asks about the gap. A plausible explanation is offered. The slide moves on.

A budget that never drives a decision is a historical document. You spent three weeks building something to describe the past.

What makes variance useful is accountability with a short feedback loop. Each department head owns their numbers. They explain variances over a threshold — say, 10% on any line — and they propose a response. The finance team doesn't investigate the gap. The operator who made the spending or revenue decision does.

What this looks like in practice

Companies that actually use their budgets don't wait for the full close. They run a weekly flash: three to five numbers that reflect what's happening operationally. Cash balance, booked ARR, pipeline, burn. Not the full P&L — just the leading indicators that will show up in next month's actuals before next month's actuals are ready.

Monthly close produces the full picture. Actuals vs budget, by department, with owner accountability built in. If engineering is 15% over plan, the engineering lead explains why and what changes. If sales is behind on new logo revenue, the sales lead owns the update, not the CFO.

Quarterly, you reforecast. Not the full annual budget rebuild — a 90-day look forward that resets assumptions based on what you've actually seen. The original budget is still the plan. The reforecast tells you whether you're on track to hit it.

This isn't complicated. It is disciplined. The gap between companies that have this and companies that don't isn't software. It's whether leadership treats the budget as a living operating tool or a one-time planning exercise.

Software won't close that gap. A process will.