Five Forecasting Methods

There is no single best forecasting method. Each method has strengths and blind spots, and the best forecast comes from triangulating multiple methods against each other. The five methods are: rep call (bottom-up judgment), stage-weighted (pipeline math), historical run rate (trailing performance), velocity-based (throughput rate), and AI/ML (pattern recognition on full feature set). Using all five and investigating where they disagree produces a forecast that is more accurate than any individual method.

1. Rep Call (Bottom-Up) Each rep predicts which deals will close and for how much. The sum is the team forecast. Strength: reps have relationship context the data cannot capture. Weakness: systematic optimism bias — reps overestimate their probability of closing by 20-40% on average.
2. Stage-Weighted Each deal's value is multiplied by its stage probability (derived from historical conversion rates). The sum of weighted values is the forecast. Strength: objective, based on historical data. Weakness: treats all deals in a stage identically regardless of their individual characteristics.
3. Historical Run Rate The trailing 3-4 quarter average revenue is projected forward. Strength: simple and stable, accounts for seasonality if enough quarters are included. Weakness: does not account for pipeline changes, team size changes, or market shifts.
4. Velocity-Based Pipeline velocity (dollars per day) multiplied by remaining selling days. Strength: captures the current throughput rate regardless of individual deal predictions. Weakness: assumes the recent velocity rate will continue, which may not hold if the pipeline composition changes.