One of the most important aspects of any analyst’s job is financial modeling. Different investment firms will use models differently—some might not even use them at all—and the aspects of models that are considered important will vary from firm to firm. But every firm will be aware of the importance of financial models to the market broadly, and so their awareness of models that exist in the market will inform their perception of a stock and its potential future value.

Let’s back up a bit. What exactly is a model? Simply put, a model is a table that tries to project future financial metrics based on past performance and analysts’ assumptions. That latter part is key; some models will be heavily influenced by analysts’ assumptions, while others will simply project future growth based on the past. The latter approach is less reliable, especially for new and fast-growing companies in quickly changing industries.

Models attempt to predict the future. They attempt to predict what numbers companies will report on their income statement, balance sheet, and statement of cash flows as they are reported to the SEC every quarter. Past reports are available broadly, including through SEC’s EDGAR and on sites like Google Finance:

financial-modeling-1

financial-modeling-21

financial-modeling-3

Here are screen shots of the financial data as reproduced on Google Finance for Google (GOOG). Note how the income statement begins with revenue, subtracts expenses from revenue, and shows us net income. The balance sheet shows how much cash the company has on hand, how many assets it owns, and how much debt it has. Finally, the cash flow statement shows how it uses its net income to pay dividends, buy shares, pay debts, and so on.

Because this data is reported every quarter and every year, we can create a time series of these data points and begin to see patterns as we see fit. Note that Google Finance does not try to calculate year-over-year changes in these numbers, although every model will. So, for instance, we might take net income in 2015 (16.3 billion) and compare it to 2014’s number (14.4 billion). Comparing the two, we see net income rose 13.2% on a year-over-year basis from 2014 to 2015, an impressive number.

If we produce a series of these comparisons, we can see how year-over-year changes trended from 2012 to 2015. We can go back even further, although few models use more than 5 years of data.

Now I am going to take a model from a sell-side analyst at BMO Capital Markets to show how these models are created by professionals.

financial-modeling-4

This is the income statement. Note how estimates for 2015 (the report is from 2014) extrapolate total revenue of $74.4 billion for 2015, a 12.7% year-over-year increase. How does this compare to actual results? Google’s 2015 gross revenues were $74.99 billion, a 3.9% difference. This shows that analyst estimates often fail to hit the mark, and in this case 4Q expectations in 2015 were far too conservative.

How can we make our models better, more accurate? That is a difficult question to answer, and no one really agrees. Most would admit that a better model often comes from a better understanding of the company’s fundamental business, so understanding how Google makes its money is key. This is why BMO doesn’t just show Google’s revenue, but breaks it out in the same way Google does. The analyst also tries to estimate future unique indicators of Google’s business, like CPC changes and paid click volume:

financial-modeling-5

From this example, we can come to three key conclusions about financial modeling:

  1. Financial models are an attempt to see future events based on analyst assumptions and past growth rates
  2. Financial models require access to data, an understanding of the fundamental business, and spreadsheet software
  3. Financial models can be made more accurate over time as analysts learn more about the company

The basis of financial modeling and the structure of income statements and balance sheets are taught in finance classes—sites like Investopedia also provide good tutorials on these rudimentary skills. What any budding analyst or investor needs to realize is the fundamental importance of financial modeling to capital markets and the need to utilize these models—and be aware of their existence—to improve their own investing performance.