If you’re trying to get into banking, equity research or any number of other competitive finance jobs, you’ve probably heard that you’ll need to learn “how to model.” But what does that really mean?

Do you know how to model?

Financial modeling encompasses a broad range of forecasting and valuation disciplines used across many fields of finance. Models can differ widely in complexity and application: Some are simple one-pagers built to get a “quick-and-dirty” estimate of next year’s Net Income. Some span 20+ worksheets and project every scenario of a multi-stage M&A deal.

Although models vary in scope and use, many share some common characteristics:

  1. Reported financials for past years are the basis for most projection models. Reported numbers are first adjusted to exclude non-cash & non-recurring items, as the focus is almost always on cash-recurring income. Reported accounting line-items are then combined into fewer, more succinct model line-items. Also, certain non-GAAP financial statistics and margin analyses are often included in the adjusted model financial statements.
  2. Projecting future years for the three main financial statements – the P&L, the Balance Sheet and the Cash-Flow Statement – is typically the first (if not the only) goal. P&L estimates for EBITDA and EPS as well as leverage statistics such as Debt/EBITDA and Interest Coverage are often the most important model outputs. In fact, “short-cut models” are often used to focus exclusively on P&L projections (“Just get me next year’s EBITDA and EPS!”). Time-consuming modeling of the Balance Sheet and Cash Flow Statement can thereby be avoided.
  3. Sensitivity & valuation analysis is frequently conducted after a projection model has been built. These analyses are often the real reason a model was built in the first place. For example, sensitivity analysis might be used to measure the impact on one model output – say next year’s EPS – from changing one or more model inputs, say revenue growth or the company’s debt level (“What happens to EPS if we increase sales growth to 5% next year?”). As the name suggests, valuation analyses involve estimating the value of a company with various techniques, e.g. comparable company or precedent transactions analysis, sum-of-parts analysis, or discounted cash-flow modeling.

In short, there is no single definition of “being able to build a model” – but usually what people want to know is: are you able to sensibly project the next few years of a company’s financial statements based on historical records. What other valuation techniques and industry specific knowledge are in question depends on the task at hand – whether you’re applying for a generalist M&A position, working on internal budgets for a tech company, or picking stocks at a buyside shop.