Forecasting is the most important part of a financial model. It connects what is known — historical performance and current conditions — to what is expected to happen next. In investment banking, forecasts form the basis for valuation, deal structuring, and financial analysis. Whether preparing a discounted cash flow, an M&A model, or an IPO valuation, the forecast drives the numbers that matter.
This article explains how financial forecasts are built inside models — from revenue and margins to capital spending, working capital, and taxes — using assumptions that are reasonable, transparent, and defensible.
What Forecasting Means in Financial Modeling
In financial modeling, forecasting means projecting the company’s future financial performance based on a set of assumptions. The model takes historical data, applies logic and business drivers, and produces forward-looking income statements, balance sheets, and cash flows.
Forecasts are typically built on a monthly, quarterly, or annual basis. The forecast period varies depending on the transaction — for example, 5 to 10 years for a DCF, or just a few years for an M&A deal.
The goal is not to predict the future precisely, but to reflect a scenario that is consistent, internally linked, and credible. Each line item in the forecast should be supported by a clear assumption and should tie back logically to the other statements in the model.
Forecasting Revenue
Revenue is usually the first item forecasted in a financial model. It drives most of the other line items, including costs, working capital, and in some cases, capital expenditures. A strong revenue forecast begins with a clear understanding of how the company earns money.
The most common approach is to break revenue into drivers such as volume × price. For example, a company may sell units of a product at an average price, or earn subscription revenue based on number of users and average fee per user. In other cases, revenue may be forecasted by segment, region, or product line.
If detailed drivers are not available, revenue is often forecasted as a growth rate applied to historical revenue, based on past performance, industry trends, or management guidance.
The key is to use a method that is both reasonable and traceable. Assumptions should be explicitly shown — whether it’s a price per unit, a percentage growth rate, or a number sourced from public filings. Bankers often pressure-test revenue forecasts to assess how conservative or aggressive they are. This is especially true in M&A or IPO scenarios.
A good revenue forecast avoids excessive optimism. It balances past results, current conditions, and expectations for future performance — all in a structure that is easy to adjust as new information becomes available.
Forecasting Operating Costs and Margins
After revenue, the next step is to forecast operating costs. These include cost of goods sold (COGS), selling, general and administrative expenses (SG&A), and other direct and indirect costs. The goal is to estimate how much it will cost to deliver the expected revenue, and how that will affect profitability.
The most common method is to express costs as a percentage of revenue. For example, COGS might be forecasted at 60% of revenue, based on historical levels or industry benchmarks. SG&A is often projected in a similar way, or in some cases, modeled as a fixed amount with minor annual growth.
Margins — such as gross margin, EBITDA margin, or operating margin — help guide these forecasts. If historical gross margin is stable, future COGS can be derived by applying the same margin. If the company expects margin improvement due to scale or cost reduction, this should be reflected in the assumptions and clearly explained.
It’s important to distinguish between fixed and variable costs. Variable costs change with revenue, while fixed costs remain steady over a range of revenue levels. Models should reflect this difference where relevant, especially in downside or stress scenarios.
The best cost forecasts are simple, consistent, and grounded in the economics of the business. Overcomplicating this section adds noise without improving accuracy.
Forecasting Capital Expenditures and Depreciation
Capital expenditures (capex) represent investments in fixed assets — such as equipment, buildings, or software — that support future growth. Depreciation is the accounting expense that spreads those costs over the useful life of the assets.
In a financial model, capex is usually forecasted based on a percentage of revenue, a fixed dollar amount, or tied to specific expansion plans. Historical capex as a % of revenue can serve as a baseline, especially if the business has a steady investment pattern. In other cases, management guidance or planned projects can inform the forward estimates.
Depreciation should be linked to the projected capex and existing asset base. Some models use a straight-line method, applying a fixed depreciation rate based on asset life (e.g., 5 or 10 years). Others may roll forward the property, plant, and equipment (PP&E) schedule, adding new capex and subtracting depreciation each period.
The relationship between capex and depreciation affects free cash flow and valuation. If capex is consistently higher than depreciation, the business may be in a growth phase. If the reverse is true, the company could be underinvesting or in a low-capex industry.
Forecasts should reflect the nature of the business. Asset-heavy companies (e.g., telecom, industrials) typically require higher and more stable capex. Asset-light firms (e.g., software, services) may show lower and more flexible investment needs.
Forecasting Working Capital
Working capital refers to the short-term assets and liabilities that affect a company’s cash flow. In financial modeling, the key components are accounts receivable, inventory, and accounts payable. Forecasting these items helps estimate how much cash is tied up in daily operations.
The most common method is to express each working capital item as a percentage of revenue or cost of goods sold. For example:
- Accounts receivable might average 60 days of sales outstanding.
- Inventory could be 45 days of cost of sales.
- Payables might reflect 30 days of payment terms with suppliers.
These assumptions are usually based on historical averages. The model then projects future balances and calculates the change in net working capital, which appears in the cash flow statement. An increase in working capital is a use of cash; a decrease is a source.
Working capital movements can have a major impact on free cash flow — especially in growing companies, where more revenue often means more cash tied up in receivables and inventory.
Assumptions should reflect business reality. A company with long customer billing cycles and limited supplier credit will need more working capital than one with faster turnover or better payment terms.
Forecasting working capital accurately helps make the cash flow forecast more realistic — and in turn, improves the reliability of the valuation or deal model.
Forecasting Interest and Taxes
Interest and taxes are often forecasted after the core operations and balance sheet items have been modeled. While they are not the primary drivers of value, they have a direct impact on net income and cash flow, and must be handled carefully.
Interest expense is typically tied to the company’s debt levels. In a simple model, it can be forecasted by applying an average interest rate to the projected debt balance. In more detailed models, interest is calculated separately for each debt instrument, considering fixed vs floating rates, amortization, and new financing needs.
Interest income may also be forecasted if the company holds excess cash. This is usually calculated by applying a short-term interest rate to the projected cash balance.
Interest forecasting can create circular references, since interest depends on cash and debt, and those in turn depend on net income, which includes interest. Most models handle this with a circularity switch, iterative calculations, or approximations that avoid looping logic.
Taxes can be forecasted using a statutory tax rate or an effective tax rate based on historical performance. If the company has net operating losses (NOLs), deferred tax assets, or other tax adjustments, these may require a separate schedule. Simpler models apply a flat tax rate to pre-tax income.
The goal is to make the tax and interest logic consistent with the rest of the model, without adding unnecessary complexity. In most investment banking models, clarity and speed matter more than full tax accuracy.
Forecasting Scenarios — Base, Upside, Downside
Most financial models used in investment banking include more than one forecast. Scenario analysis allows the team to test how changes in assumptions affect the company’s valuation, cash flows, or credit metrics.
The most common scenarios are:
- Base case – Reflects the most likely set of assumptions, usually based on management guidance or historical trends.
- Upside case – Reflects stronger performance, such as higher revenue growth, margin expansion, or delayed capital spending.
- Downside case – Tests more conservative outcomes, such as slower growth, higher costs, or weaker working capital efficiency.
Scenario logic is usually built into the model using assumption toggles, where a user can switch between cases. Each case feeds through to the full set of financial statements and outputs. This allows the banker to see how valuation or deal metrics change across outcomes.
In live transactions, scenario forecasts are used to negotiate deal terms, assess risk, and prepare client materials. They are especially important in M&A and LBO models, where return sensitivity to small changes can be significant.
A good model makes scenario analysis easy to run and easy to explain. Assumptions should be clearly defined, and the impact on outputs should be visible without manual adjustments.
What Makes a Forecast Credible
A forecast is only useful if others believe in it. In investment banking, credibility matters as much as technical accuracy. A model with clean logic will still be questioned if the assumptions don’t make sense.
Credible forecasts start with reasonable assumptions. These should reflect historical trends, current market conditions, and known business drivers. If growth rates or margins deviate sharply from the past, the model should clearly explain why.
Forecasts must also be internally consistent. Revenue growth should align with staffing needs, capex, and working capital. If one part of the forecast changes, related items must adjust accordingly. A model that shows rising revenue with flat expenses or fixed asset levels may signal poor logic or optimism.
Transparency is critical. Assumptions should be easy to locate, clearly labeled, and editable. Hidden inputs or hardcoded numbers reduce trust and slow down review.
Finally, the model must be easy to explain. In many cases, bankers must walk clients, investors, or internal committees through the forecast. The model should support a clear narrative and allow for quick adjustments when assumptions change.
A good forecast tells a realistic story with numbers — one that stands up to scrutiny and helps support real decisions.
Closing Thought
Forecasting is at the heart of financial modeling. The quality of a model depends on how clearly and consistently the future is projected. In investment banking, where deals depend on fast, high-stakes analysis, forecasts must be structured, defensible, and ready to adapt as new information emerges.