5 Revenue Forecasting Techniques Every Analyst Should Know

Revenue forecasting is arguably the single most important skill in equity research. Every other line item in the model -- cost of goods sold, operating expenses, working capital, capex -- either directly or indirectly depends on your revenue assumption. Getting it wrong cascades through the entire analysis.

The first approach is top-down market sizing. Start with the total addressable market (TAM), estimate the company's market share trajectory, and derive revenue. This works well for companies in large, well-defined markets where third-party data on market size is readily available. The risk is that TAM estimates can be overly optimistic.

The second is bottom-up unit economics. Decompose revenue into its building blocks: number of customers multiplied by average revenue per user (ARPU), or units sold multiplied by average selling price (ASP). This is more granular and often more defensible, but requires detailed operational data that may not always be publicly available.

Third is the growth rate extrapolation method. Analyze historical revenue growth rates, identify trends, and project them forward -- typically with some reversion toward industry averages as the company matures. This is quick and intuitive but can be dangerously simplistic if the business is going through a structural change.

Fourth, segment-level forecasting breaks the business into its reporting segments or product lines and forecasts each independently. This is essential for diversified companies where different parts of the business have very different growth profiles and margin structures.

Finally, regression and correlation analysis uses statistical relationships between revenue and external drivers (GDP, commodity prices, housing starts, etc.) to build a forecast. This adds rigor and can be especially powerful for cyclical businesses. The best analysts typically blend multiple approaches and triangulate toward a range rather than relying on a single method.