Two directions, two systematic errors
Top-down and bottom-up forecasting are complements, each carrying a systematic bias that compounds when either is used alone. Top-down is anchored to ambition. Bottom-up is anchored to individual rep conservatism. Neither bias is obvious in a single quarter. Both become visible over a full year.Companies that rely exclusively on top-down forecasting discover mid-year that the market share assumption required to hit plan was not grounded in capacity reality. Companies that rely exclusively on bottom-up forecasting report numbers the board treats as a floor, not an expectation.
Method comparison
| Dimension | Top-Down | Bottom-Up |
|---|---|---|
| Starting point | Market size, board target, prior year growth | Rep-level pipeline and deal probability |
| Primary use | Annual planning, goal setting, investor narrative | In-quarter forecast, pipeline review |
| Systematic bias | Optimism, decoupled from capacity | Sandbagging, undercommitment |
| Best owner | Finance, strategy | Sales ops, RevOps |
| Data required | Market data, historical growth rates | CRM data, deal-level probability |
How each method fails in isolation
A top-down-only plan distributes a target to managers who then distribute it to reps. The reps are assigned a number that was constructed from a market share argument, not from a pipeline model. They cannot trace the number to their own pipeline reality, which creates immediate credibility problems and often leads to inflated pipeline creation to show coverage.
A bottom-up-only forecast is a roll-up of rep-level estimates. Those estimates inherit every individual rep's tendency to shade conservatively, describe their best case as their commit, or inflate their pipeline to avoid negative attention. The company-level number is a statistical artifact of those individual distortions.
The combined approach
Start with top-down to set targets: what does the board need, what does the market support, what does headcount capacity allow? Then run bottom-up to forecast actuals: what does the current pipeline support at realistic conversion rates? The gap between the two is the planning gap. Closing it is a management decision, not a forecasting one.
Top-down forecasting and bottom-up forecasting each have their own mechanics. Both methods are only as reliable as the forecast accuracy discipline built around them.Frequently Asked Questions
What is the main bias in top-down forecasting?
Top-down forecasting is anchored to desired outcomes, not actual pipeline. The number starts with what leadership needs to be true and works backward. This creates optimism bias: the plan looks achievable on paper because the market share assumption is set to make the math work, not because rep-level capacity and pipeline quality support it.
What is the main bias in bottom-up forecasting?
Bottom-up forecasting is subject to sandbag bias. Reps undercommit to protect themselves from missing. Managers add a haircut to roll-ups as a habit rather than from analysis. The result is a conservative number that the company routinely beats, but that the board or investors cannot use for planning because it is not a true expectation of outcomes.
When should you use top-down versus bottom-up?
Use top-down for annual planning and board-level goal setting, where you need a market and capacity argument for your targets. Use bottom-up for in-quarter forecasting, where you need the most accurate read on what will actually close. The most reliable approach combines both: set targets top-down, forecast outcomes bottom-up, and close the gap explicitly.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like top-down vs bottom-up forecasting into prescriptive action for your team.
Schedule a Demo