SteadySharesSteadyShares
All guides
TutorialsIntermediate· 8 min read

Tutorial: analyse a software company

The one thing to remember

In software, the question is not whether customers buy. It is whether they stay, and whether they spend more each year.

What you will be able to do

Judge a software business on the things that actually predict its future.

Figure

How analyse a software company works, in one picture

1Start with retention, because it dominates everything2Then ask what a customer costs to acquire3Check gross margin, and be suspicious if it is low4Look for the switching cost5Then find the stock-based compensation

The same argument as the text, as a chain. Each step is what makes the next one possible.

Figure

The only five moats there are

1
Brand
People pay more for the same thing
2
Switching costs
Leaving is painful or expensive
3
Network effects
It gets better as it gets bigger
4
Cost advantage
It can undercut and still profit
5
Scale in a small market
Not worth invading

If you cannot name which of these a company has, it probably does not have one. It is merely doing well, which is a different and far more temporary condition.

  1. 1

    Start with retention, because it dominates everything

    Net revenue retention asks: of the money existing customers spent last year, how much are they spending this year, after cancellations and upgrades? Above 100% means the existing base grows on its own, without a single new customer. That is the closest thing to magic in business.

    Below 100% means the company is filling a leaking bucket, and every new sale is partly replacing one it lost.

    A company with 90% retention can grow fast and still be worthless, because the growth stops the moment the marketing budget does.

  2. 2

    Then ask what a customer costs to acquire

    Sales and marketing spend, divided by the new customers it produced. Compare that to what a customer is worth over their life. If it costs more to win a customer than they will ever pay you, scale makes the problem bigger, not smaller.

  3. 3

    Check gross margin, and be suspicious if it is low

    Financials

    Real software runs at 75 to 85%. If a company calls itself software and posts a 45% gross margin, it is probably a services business with a login screen, and it should not be valued like software.

  4. 4

    Look for the switching cost

    Overview

    Software's moat is rarely the software. It is that the customer has built their workflow, their data and their integrations around it, and ripping it out would take a year. That is why boring, deeply embedded, unglamorous products are often far better businesses than exciting ones.

  5. 5

    Then find the stock-based compensation

    Software companies pay staff in shares. It is a real cost, it dilutes you every year, and it is the first thing excluded from 'adjusted' profit. Check the share count over five years. If it keeps rising, the company's growth is partly being paid for out of your slice.

Try it
What a P/E is actually sayingInteractive
P/E ratio
20.0
Years of earnings you pay
20 yrs
Payback allowing for growth
12 yrs

Ordinary. The market expects steady, unremarkable growth.

The P/E is years of today's earnings you are paying. Growth shortens the payback, which is why fast growers deserve higher multiples.
You have got it when

You can name the company's net revenue retention and its five-year change in share count.

Go and do it in SteadyShares

Read next

The bottom line

In software, the question is not whether customers buy. It is whether they stay, and whether they spend more each year.

See the 30 live screens

Every one shows its exact method, and the circumstances in which it is wrong. Free, and no account to look.