Why forecasting feels harder with commission sales
When you have salaried reps, forecasting often starts with activity metrics. You know how many calls they make, how many demos they book, and you can roughly predict output from input. With commission only agents, you have less control over daily activity, which makes traditional forecasting models feel unreliable.
But here is the thing: commission sales forecasting is not harder, it is just different. And in many ways, it can be more accurate because you are forecasting from real deal flow rather than assumed activity levels.
Start with pipeline weighting
The most reliable method is to weight your pipeline by stage. Assign a probability to each deal stage based on historical conversion rates.
For example, if deals at the proposal stage close 40% of the time, a $10,000 proposal contributes $4,000 to your weighted forecast. Sum these across all agents and you have a bottom up revenue projection.
Track these conversion rates monthly. As your data improves, so does your forecast accuracy.
Factor in agent ramp curves
New agents rarely produce at full capacity in their first month. Most commission agents take two to four weeks to learn the product, build initial pipeline, and close their first deal.
Build a ramp curve into your model. If an experienced agent closes $20,000 per month at full capacity, assume $5,000 in month one, $12,000 in month two, and full run rate by month three. This prevents you from overestimating revenue when you are onboarding new agents.
Use cohort based projections
Group your agents by start date and track their revenue over time. This gives you a cohort curve showing average revenue per agent at each month of tenure. When you add new agents, apply the cohort curve to project their likely contribution.
Platforms like Zepys make this easier because all deal data is tracked in one place, giving you clean cohort data without manual spreadsheet work.
Adjust for seasonality
Every business has seasonal patterns. If your product sells more in Q4, your commission agents will reflect that. Build seasonal multipliers into your forecast using at least 12 months of historical data.
The bottom line
Forecasting with commission agents is about patterns, not control. Track your pipeline, understand your ramp curves, and let the data guide your projections. The longer you run the model, the more accurate it becomes.