Revenue Model & Build

The top-down approach starts with total addressable market (TAM) and estimates the company's market share penetration over time. It's useful for early-stage companies or when detailed operational data isn't available. The bottom-up approach builds revenue from granular operational drivers: units sold × average selling price, number of subscribers × average revenue per user (ARPU), or number of stores × revenue per store. Bottom-up is more defensible and preferred in investment banking because it connects revenue directly to operational assumptions that can be verified and stress-tested. Most sophisticated models use a combination: a bottom-up build for the near term and a top-down sanity check for the outer years.

Breaking revenue into segments — by product line, geography, customer type, or channel — dramatically improves model accuracy and analytical utility. A retail company might be modeled as: same-store sales growth + new store openings × average store revenue. A SaaS company might use: beginning ARR + new ARR + expansion ARR − churned ARR. Each segment should have clearly identified drivers with distinct growth rates and margin profiles. The revenue build connects directly to operating margin and gross margin assumptions because different segments often carry different profitability, which affects the entire P&L forecast.

Start by analyzing 3-5 years of historical revenue data to identify trends, seasonality, and growth drivers. Break historical revenue into the same segments you'll project forward. For each segment, identify 2-3 key drivers and build your projections from those drivers. Cross-check your bottom-up build against management guidance, consensus estimates, and industry growth rates. Build in sensitivity analysis on the most impactful drivers — typically volume/units and pricing — to show the range of outcomes. The revenue model feeds directly into COGS (to calculate gross margin), operating expenses, and ultimately free cash flow.

The right revenue model structure varies significantly by industry. For subscription/SaaS businesses: cohort-based models tracking new customers, retention rates, and ARPU expansion. For retail: same-store sales + new store rollout schedule. For manufacturing: volume × price with separate assumptions for each product category. For financial services: assets under management × fee rate, or loan book × net interest margin. For healthcare: patient volume × reimbursement rates by procedure. Knowing which revenue model structure to use for a given industry demonstrates the kind of sector awareness that impresses interviewers and is essential for producing credible projections in live deal models.