# Chapter 15: The company snapshot

*The Analyst's Handbook, Sifter's textbook for student investors · By Matt Lang and Bree Lang · Research · https://sifter.finance/textbook/ch15-snapshot*

The wall of stats on the ticker page, which numbers move without the company doing a thing, and how long a company's cash will last.

Where this is on Sifter: Research · Ticker · Overview

Open Research, type in any ticker, and take a look at what the Overview tab presents you with. There is a P/E, a forward P/E, an EV/EBITDA, a P/S, three varieties of margin, revenue growth, a dividend yield, a beta and a 52-week range, all lined up in a wall of statistics as if each number needed the same amount of your attention. Many of you may be looking at that wall and thinking, "am I supposed to have all of these memorized before I can call myself an analyst?" Fortunately, the answer is no. Nearly every number on the page can be computed from a few lines of the financial statements (the income statement, the debt and cash lines of the balance sheet, and a line or two of the cash flow statement) and a market quote, and in this chapter we will build the valuation half of the snapshot ourselves from exactly those inputs. (The remaining tiles, beta, alpha and the 52-week range, need a year or more of price history rather than a filing, and Chapter 6 already built the first two.)

Once the numbers are ours, we will do something more useful with them, which is to sort them into two bins. Some of the statistics are **price-driven**, which means that they wobble every trading day because the share price is somewhere in their formula. Others are **fundamentals-driven**, and these change only when the company reports something, which is four times a year for a quarterly filer and, on Sifter's tiles, once a year, because the tiles are read from the annual statements. In other words, part of the page is telling you about the mood of the market and part of the page is telling you about the business itself, and knowing which lines belong to which bin is most of the skill of reading the snapshot. The figure below is the card from the demo, which shows sixteen statistics in a grid (the live card now has five more, a P/B, a P/FCF, an FCF yield, a forward PEG and a free float, and we will get to them later in the chapter), and by the end of the chapter you will be able to sort every one of them.

![The Key Statistics card on a ticker Overview: sixteen tiles in a four-column grid, market cap, EV/EBITDA, P/E, forward P/E, P/S, EPS, revenue growth, three margins, dividend yield and annual dividend, one-year alpha, five-year beta, and the 52-week high and low.](https://sifter.finance/textbook/figures/ch14-keystats.png)

*Key Statistics, on the demo's Microsoft page. Research › Ticker › Overview. Most tiles open a drill-down with the statistic's history, and the two 52-week tiles jump to the chart. The demo shows a dated snapshot; on a live ticker the price-driven tiles (market cap, the three multiples, dividend yield, alpha and the 52-week range) move with the quote, and the fundamentals tiles wait for the next annual report.*

To do the sorting we need a company to practice on, and a made-up one is better than a real one here, because its numbers are round and its statements fit in a table. BRC Health is the practice company for the rest of the chapter. Its latest annual figures are below, and everything we build comes from those nine lines. One line on it, EBITDA, is a term you may not have met. It is earnings before interest, taxes, depreciation and amortization, and the note under the table says how it was arrived at.

**Practice company: BRC Health, latest annual figures**

| Line item | Value |
| --- | --- |
| Revenue | $200.0M |
| Cost of goods sold | $80.0M |
| Operating expenses | $70.0M |
| EBITDA | $80.0M |
| Net income | $40.0M |
| Shares outstanding | 20.0M |
| Share price | $50.00 |
| Total debt | $250.0M |
| Cash | $50.0M |

EBITDA is operating income before depreciation and amortization. Operating income here is revenue less cost of goods sold less operating expenses, $50.0M, and the operating expenses include $30.0M of depreciation and amortization, which is how EBITDA comes to $80.0M. To keep the arithmetic short, the table hands it to you directly rather than making you build it.

Let's start with the numbers that come from the statements alone, the margins, because they need no share price at all. Then we will bring the price in and watch what it does to everything it touches.

### Worked example: Worked example: building the snapshot from scratch

We will build eight numbers from the table (three margins, market cap, EPS, P/E, P/S and enterprise value) and finish with EV/EBITDA, and each one takes a single line of arithmetic.

1. **Margins (fundamentals-driven)** Gross = (200 − 80) / 200. Operating = (200 − 80 − 70) / 200. Net = 40 / 200. *Gross 60%, operating 25%, net 20%. These move only when the company reports.*
2. **Market cap and EPS** Market cap = 20M shares × $50 = $1,000M. EPS = $40M / 20M shares. *EPS = $2.00. Market cap is price-driven, and EPS is fundamentals-driven.*
3. **P/E and P/S** P/E = 50 / 2.00. P/S = 1,000 / 200. *P/E = 25, P/S = 5. Both have the price in the numerator, so they move every day the price does, with no news from the company at all.*
4. **Enterprise value** EV = market cap + total debt − cash = 1,000 + 250 − 50. *EV = $1,200M. EV is the price of the whole business, which is what buying every share and paying off the net debt would cost.*
5. **EV/EBITDA** 1,200 / 80. *EV/EBITDA = 15. This is the multiple that lets you compare companies with different debt loads, which P/E cannot do.*

So five of the nine numbers have the share price somewhere in them (market cap, P/E, P/S, enterprise value and EV/EBITDA) and four do not (the three margins and EPS). The quickest way to feel the difference is to move the price and leave everything else alone, which is what the check below asks you to do.

### Practice

BRC's stock jumps to $60.00 tomorrow with no other changes. What is the new EV/EBITDA?

**Solution.** The new market cap is the only thing that changed, so EV = 1,200 + 250 − 50 = $1,400M, and 1,400 / 80 = 17.5x. Take a moment to think about what moved and what did not. The price moved 20% and the multiple climbed from 15 to 17.5, but the company itself did not report, sell, hire or change in any way. This means that the move in EV/EBITDA told you about the mood of the market rather than about the business, and that is what we mean when we call a stat "price-driven".

The same logic runs in reverse. When a multiple moves, either the numerator or the denominator moved, or both moved by different amounts, and part of reading a snapshot is working out which. The question below asks you to rule one story out.

### Check yourself

A stock's P/E fell from 30 to 24 over a quarter. Which explanation is IMPOSSIBLE?

- The price fell while earnings held steady. This one is entirely possible, because a 20% drop in the price with earnings holding steady does this to the multiple all by itself.
- Earnings grew while the price held steady. Also possible. EPS growth of 25% at a flat price takes the multiple from 30 to 24.
- Both price and earnings fell by the same percentage. **(correct)** Right. When the numerator and the denominator fall by the same percentage, the ratio is UNCHANGED, which means that this story cannot explain a P/E that moved at all. A multiple that moved is telling you that price and earnings moved differently from one another, and figuring out which of the two did the moving is the actual analysis you are being asked to do.

One tile on the card needs its own explanation before the lab, because it looks like a duplicate and is not. The card shows two P/Es, and they divide the same price by two different earnings figures.

> **Forward versus trailing** The plain P/E is trailing, meaning the price divided by the last fiscal year's actual diluted EPS, while the forward P/E divides today's price by the consensus estimate of NEXT fiscal year's EPS, the first fiscal year ending after the one last reported (the same estimates the Analyst tab lays out year by year). On the platform the trailing figure is today's market cap over the last fiscal year's net income, which is the same thing as price over EPS when you use today's share count, and the P/S and EV/EBITDA tiles are built the same way, so all three move with the quote exactly as the worked example says they should. For a growing company the forward multiple is lower than the trailing one, because next year's earnings are expected to be larger than last year's. Be careful when you put two companies side by side, because comparing one company's trailing multiple to another company's forward multiple is the classic apples-to-oranges error in student pitches.

Now that we have both P/Es, let's look at the tiles that joined the card more recently, starting with the two that use cash instead of earnings. Accounting earnings include plenty of items that are not cash (depreciation is the usual example), so analysts like to check a company's price against its free cash flow, which is the operating cash flow on the cash flow statement minus capital spending (capex). P/FCF divides market cap by the latest fiscal year's free cash flow, and FCF yield turns the same ratio upside down and states it as a percent. Suppose BRC's cash flow statement shows $75M of operating cash flow and $25M of capex, which leaves $50M of free cash flow. At the $50 price, P/FCF is 1,000 / 50 = 20x and the FCF yield is 50 / 1,000 = 5%. In other words, every $100 of BRC's market value is backed by $5 a year of cash that the business could hand to its owners, and because the price is in both formulas, both tiles belong in the price-driven bin.

The forward PEG takes the forward P/E from the callout above and divides it by the growth in earnings that analysts expect. On the platform the growth figure is the consensus EPS growth rate per year, measured from next fiscal year's estimate out to the estimate as many as three years after it, and it goes into the formula as a number of percentage points. If BRC's consensus EPS for next year is $2.50, its forward P/E is 50 / 2.50 = 20, and if analysts expect earnings to grow 10% a year from there, the PEG is 20 / 10 = 2.0. A lower PEG means you are paying less for each point of expected growth, and investors use it to compare a fast grower on a high P/E with a slow grower on a low one. The tile is blank when forward earnings are negative or analysts expect them to fall, because there is no growth rate to divide by (and a negative one would make a shrinking company look cheap). The last of the new tiles, free float, is the share of the shares outstanding that trade freely, leaving out the stakes held by insiders and large strategic owners, so a BRC with a free float of 85% has 15% of its stock in the hands of people who are not trading it day to day. Free float comes from a filing rather than from the quote, which puts it in the fundamentals bin with the margins and EPS.

So far every company in this chapter has made money, and the multiples have been easy to interpret, but things get strange when the figure at the bottom of a multiple turns negative. Imagine a young drug developer (let's call it Kestrel Bio) trading at $40 a share after losing $2.00 a share last year. Its P/E is 40 / −2.00 = −20, and if its losses shrink to $0.50 a share while the price holds at $40, the P/E becomes 40 / −0.50 = −80. You may be thinking, "the company got better and the multiple got four times worse?", and you would be right to be suspicious of it. A negative P/E grows more negative as the losses shrink, and when Kestrel finally earns a profit, its P/E jumps from a very large negative number to a very large positive one without passing through any of the numbers in between. For this reason analysts call a multiple **NM**, short for "not meaningful", whenever the figure it divides by (earnings, EBITDA or book value) is negative.

The Research page follows the same convention. When a multiple in the header or on the Key Statistics card has a negative denominator, the page still shows the calculated number, followed by a small NM tag, and hovering over the tag explains the convention. The number is there so that you can see it, but every NM value is left out of the peer medians and comparisons, because a median that mixed Kestrel's −20 with a profitable competitor's 25 would describe neither company. An NM tile is also different from a blank one. A blank value shows a dotted dash, and hovering over the dash tells you why the figure is missing (the dividend yield is blank, for example, when the company paid no dividend in the last twelve months, and the PEG is blank for the reasons we just covered).

So what should we use for a company like Kestrel? The first option is a multiple whose denominator is not negative, and revenue is the usual choice, because a company can lose money for years while its sales keep growing. P/S on the Overview card (and EV/Revenue on the Comparables tab) will give you a number you can compare with other companies when earnings cannot. The second option is to turn the multiple upside down. **Earnings yield** is EPS divided by the price, the P/E flipped over, and unlike the P/E it moves smoothly through zero. Kestrel's earnings yield goes from −2.00 / 40 = −5% to −0.50 / 40 = −1.25% as its losses shrink, and if it then earns $0.50 a share the yield rises to +1.25%, which is the steady improvement the business actually made. In other words, the yield rises whenever the company improves, whether it is losing money or making it, while the P/E only behaves on one side of zero.

The statistic drill-down uses the same idea. Clicking a tile opens that statistic's history, and a Quarterly / Annual switch lets you choose between the last eight quarters (each on a trailing-twelve-month basis, which means that each point is the sum of four quarters) and up to six fiscal years from the annual reports. When the P/E or EV/EBITDA is negative today, or was negative in any of the periods on the chart, the drill-down charts earnings yield or EBITDA yield (EBITDA divided by enterprise value) instead, adds a line at zero, and prints a note above the chart saying why it switched. The question below asks you to work through Kestrel's two P/Es in the same way.

### Check yourself

Kestrel Bio's share price is the same as it was a year ago, but its P/E went from −20 to −80. What happened to the business?

- Its losses grew four times larger. This is the tempting answer, because the multiple looks four times worse, but a bigger loss at the same price makes the P/E smaller in size, not larger. A loss of $8.00 a share at a $40 price gives a P/E of −5, not −80.
- Its losses shrank to a quarter of their old size. **(correct)** Right. At a $40 price, a P/E of −20 means a loss of $2.00 a share and a P/E of −80 means a loss of $0.50 a share, so the business improved a lot while the multiple looked as if it fell apart. The earnings yield tells the same story the right way around, rising from −5% to −1.25%, and this is why the page tags the P/E as NM and the drill-down switches to the yield.
- It turned profitable. A profitable company has a positive P/E, so a P/E that is still negative tells you the company is still losing money. When Kestrel does turn profitable, its P/E will jump from negative to positive in one step.

Every valuation tile on the card is one line of arithmetic on the statements and the quote. The lab below has you rebuild one of them from a real filing, so that the next time a tile looks wrong you know how to check it.

### Lab: Lab: rebuild a real margin from the filings

1. Open Research, load any of your holdings, and read its net margin off the Overview tab.
2. Go to the Filings tab and download the latest 10-K's financials as XLSX.
3. From the income statement, divide net income by total revenue.
4. Compare your number with the Overview stat. The Overview's margins are computed from the data vendor's latest annual income statement, so first make sure you are looking at the same fiscal year. The sheet's header names its years, and the Overview's Revenue tile tells you which year the vendor is on (the two can be a filing apart in either direction). Same year, and the margins should agree to the rounding. A difference then means you grabbed the wrong line, and finding the right one will teach you more than a match would have.

Everything on the Overview card has been about price, either what the market is paying for the company or how much the company earns for that price. The Fundamentals tab opens with a different question, which is whether the company can pay what it owes, and the section at the top of the tab, Balance Sheet & Liquidity, collects the figures an analyst checks before trusting any multiple. Its first row covers cash and debt, where Cash & short-term investments adds Treasury bills and other investments the company can sell within a year to the cash itself, total debt includes lease obligations (a lease is a promise to pay, much like a loan), and **net debt** is total debt minus cash and short-term investments. A negative net debt means the company holds more cash than it owes, and the page shows that as net cash.

BRC holds no short-term investments, so its net debt is 250 − 50 = $200M, and dividing by its $80M of EBITDA gives a net debt / EBITDA of 2.5x. This means that BRC would need about two and a half years of operating profit to pay off its debt, which is below the 3x level where lenders usually start to treat leverage as high. (On the page, the EBITDA in this ratio is the sum of the last four quarters, and the ratio is NM when EBITDA is negative.) Net debt comes from the newest balance sheet the company has filed, which is often a quarterly one rather than the annual report, and the DCF tab that Chapter 16 builds starts from the same figure.

The second row asks whether the company can cover the bills that come due soon. The current ratio divides current assets by current liabilities, so a figure above 1x means the company holds more short-term assets than it owes over the next year, and the quick ratio does the same thing without inventory, which can be slow to turn into cash. Debt / equity compares borrowing with the shareholders' book equity. Interest coverage divides operating income by interest expense over the last four quarters, and if BRC pays $10M a year in interest, its $50M of operating income covers that bill 50 / 10 = 5 times. A company with coverage close to 1x is earning just enough to pay its lenders, and one bad year could leave it unable to make the payments.

For a company that is losing money, the most important figure in the section is a different one. Kestrel Bio has no operating profit to cover anything with, so the first question an analyst asks about it is how long its cash will last. When a company's free cash flow over the last four quarters is negative, the page adds a Cash burn row with that free cash flow, the average burn per quarter (the negative free cash flow divided by four) and the **cash runway**, which is cash and short-term investments divided by the annual burn. In other words, the runway is the number of years the company could keep going at its recent pace before the money runs out, if it raised nothing new along the way.

**Cash runway**

```
Runway (years) = cash and short-term investments ÷ (−free cash flow over the last four quarters)
```

STI is short-term investments. The denominator is the annual burn, the free cash flow over the last four quarters with its sign flipped so that a burn is a positive number, and the page shows a runway only when that free cash flow is negative.

### Practice

Kestrel Bio holds $240M of cash and short-term investments, and its free cash flow over the last four quarters was −$160M. What is its cash runway, in years?

**Solution.** The annual burn is $160M, which works out to $40M a quarter, so the runway is 240 / 160 = 1.5 years. Take a moment to think about what 1.5 years means for someone who owns the stock. It is under the two-year line where the page shows a warning, because a company that expects to run low within two years usually has to raise money well before the cash is gone, and the usual ways of doing that are new debt, money from a partner or an equity raise that dilutes the people who already own the shares.

A runway is a rough guide rather than a forecast, and the page says so under the figure. It assumes the burn of the last four quarters continues unchanged, with no new debt, equity or partner funding, and it does not count credit lines the company has arranged but not yet drawn, so check the liquidity discussion in the 10-K before you quote a runway in a pitch. The burn can also speed up or slow down (a drug developer that starts a large trial will spend faster than it did last year). The Dilution row comes next, because the most common way for a cash-burning company to stretch its runway is to sell new shares. It shows the change in the weighted-average share count over three fiscal years, where a rising count means each existing share owns a smaller slice of the company, and stock-based pay as a share of revenue, which is a real cost to shareholders even though no cash leaves the business.

The last row, Working capital, measures something that matters for every company, profitable or not, which is how long its cash is tied up in running the business. Imagine that BRC buys a shipment of medical supplies on day 0. Its supplier gives it 30 days to pay, the supplies wait in the warehouse for 60 days before they are sold, and BRC's customers take 45 days to pay their bills after the sale. BRC pays out cash on day 30 and does not get it back until day 105 (60 days to sell plus 45 days to collect), which means that its cash is tied up for 75 days on every shipment.

Each of those three pieces has its own tile on the page. **Days sales outstanding** (DSO) is the average number of days the company takes to collect after a sale, **days inventory** (DIO) is the average number of days inventory waits before it is sold, and **days payables** (DPO) is the average number of days the company takes to pay its suppliers. Put together, they make the **cash conversion cycle**.

**Cash conversion cycle**

```
Cash conversion cycle = DSO + DIO − DPO
```

Each piece is measured in days. DSO compares receivables with revenue, DIO compares inventory with cost of goods sold, and DPO compares payables with cost of goods sold, each scaled to a 365-day year. The page takes all four figures on a trailing-twelve-month basis from the data provider.

For BRC the cycle is 45 + 60 − 30 = 75 days, the same 75 days we counted on the calendar. A shorter cycle means less cash is stuck in receivables and inventory, and a cycle that grows longer from one year to the next can be an early sign that customers are paying more slowly or that products are not selling. Some companies even run a negative cycle, collecting from their customers before they pay their suppliers, which means that the suppliers are financing part of the business. Notice that DPO is the only piece with a minus sign, so a company can shorten its cycle without selling or collecting any faster, and the check below has you work through one example of it.

### Practice

BRC negotiates with its supplier and now pays in 45 days instead of 30. Its DSO (45 days) and DIO (60 days) do not change. What is its new cash conversion cycle, in days?

**Solution.** The new cycle is 45 + 60 − 45 = 60 days, down from 75. BRC did not sell or collect any faster, but by holding on to its supplier's money for an extra 15 days it cut the time its own cash is tied up by the same 15 days. This means that a shrinking cash conversion cycle deserves a second look, because it can come from the business running better or from the company paying its suppliers later, and the three day counts on the page tell you which of the two it was.

Make sure you feel good about both halves of this chapter. The Overview card tells you what the market is paying for the company, which of its tiles move with the price and which multiples stop making sense when a denominator turns negative, while the Balance Sheet & Liquidity section tells you whether the company can pay its bills and, for a company that is losing money, how long it can keep going. In practice it helps to look at the balance sheet before trusting a multiple, because a company with eight months of runway will probably raise money soon, and the new shares it sells will change every per-share figure on the card. Chapter 16 takes the free cash flow we met in this chapter and builds a fair value out of it.