9 ms·

Intrinsic Value of a SaaS Business: A 20 Year DCF

80 points by pstrazzulla 8y ago

vbhartia 8y ago
great write up. always been curious about this... how do you value growth... especially as Uber / Lyft IPO and face slowing growth. What basis do you use to get the discount rate of a SaaS company? This seems fairly arbitrary and is a key part to valuing a company
- pstrazzulla 8y ago
  Growth is important, especially if gross margins are high and >50% of new revenues goes to the bottom line to decrease losses or eventually increase profits. It's tough though, the real questions becomes when does growth stop, which is a function of market size and profitable customer acquisition channels. The discount rate is fairly arbitrary. You can use something like the CAPM to get to your cost of equity, but it's more designed for slower growth companies. I prefer to think of the gambler analogy where you're trying to figure out the percent chance of getting paid back. Of course, it's SO HARD to figure out the probability of a startup continuing on its current path given the many variables. This is where the "artists" of the world can capture a lot of value.
- clairity 8y ago
  > "What basis do you use to get the discount rate of a SaaS company? This seems fairly arbitrary and is a key part to valuing a company" this is one of the things you learn in an MBA program. you'd generally use multiple methods to triangulate a rational estimate. for example, you'd look at industry comparables and estimate a beta to plug into CAPM (as sibling commenter touches on). you can also do full financial projections (5-7 years out) based on expected performance, do DCF, and monte carlo that to back out a discount rate. you can also do a comparative ratio analysis (https://www.investopedia.com/terms/r/ratioanalysis.asp https://www.investopedia.com/terms/r/ratioanalysis.asp) on profitability, cash flow, asset efficiency, turnover, and the like. from my (limited) experience, startup valuations seem to be most sensitive to growth rate and the discount rate, so modelers spend a lot of time estimating these factors.
  - pstrazzulla 8y ago
    This guy is basically the god of the DCF: http://aswathdamodaran.blogspot.com/search?q=dcf http://aswathdamodaran.blogspot.com/search?q=dcf
    - clairity 8y ago
      ha, yes, we used damodaran's book in my corporate valuation class! i forgot to mention that you can also employ an option pricing model, like black-scholes or the binomial model (in simpler cases), in cases with multiple classes of equities, wherein you can back out a valuation (and subsequently, a discount rate) based what the various VCs involved are expecting.
      - pstrazzulla 8y ago
        Very true. Especially since the equity value of a startup basically looks like an out of the money call option where you're paying a small amount now for a potential large payoff of our money (but high chance of $0 outcome). Also, more volatility will increase the value of this option (whether it's founder or market volatility - aka Travis from Uber or Cypto or Cannabis).
ttul 8y ago
I wanted to know the long-term value of my recurring revenue company. Being a programmer at heart, I wrote a python script that tested out 100,000 random scenarios of customer churn over the next 50 years. This gave me a histogram of future cash flows from the business, allowing me to state with some level of certainty what the range of future cash flow was likely to be. I was really surprised at how valuable the company is when analyzed in this manner. And there’s no bullshit. You assume a churn rate and randomly try it out over thousands of scenarios.
- robodale 8y ago
  Hey...you...ummm <cough> <cough>...wouldn't mind sharing that script?
- harlanji 8y ago
  What approach did you use? I wrote a discrete event sim from scratch once and it was a good experience, simulating adding express lines to a bank for certain transactions. I can see it being overkill, or possibly a good method for tuning dials, depending what you did. I have a sim that I've been kicking around for a couple of months related to my current work environment that I think would help make informed arguments about rule changes over the long term.
  - ttul 8y ago
    I have begun using Monte Carlo simulations for various modeling purposes at the company over the last two years. For example, I created a script to model a price increase and the expected outcome from that increase given a certain rate of churn. That gave me the confidence to go ahead with a 30% increase knowing that in the median case we would end up with more revenue even if some huge number of customers said bye bye. The increase worked, and churn was lower than expected.
    - harlanji 8y ago
      Thanks for that and the business application, great result. This seems like a good method to make the case for serving more quality sensitive markets against the grain; that’s where I am coming from. Have heard the term, imagine it’ll be added to my tool shed soon. Thanks again.
justchilly 8y ago
The assumptions in your screen capture are dramatically more aggressive than any company in history. E.g. You are showing 60% yoy revenue growth 10 years in. For comparison: Google in year 10: 31%. Amazon in year 10: 23%. Dropbox in year 10: 27%. If these kinds of growth rates were possible, there would likely be much higher discount rates to go with them. Lots of investors do long term DCFs like this (and almost all update 5-year models on a rolling basis).They're just not as useful as a shorter term model given lack of visibility that far into the future and likelyhood of having to return funds well before then.
- pstrazzulla 8y ago
  TWLO, SHOP, etc are all growing at >50% 10 years in. It's very likely and common, especially in b2b SaaS where it's challenging to capture all of a fragmented market and/or you have a case where you can continually upsell with more product over time.
- aytekin 8y ago
  Yes, once you scale really big it is hard to keep a high growth rate. But if you have been always growing at 50% YoY, it is possible to keep that rate even at year 12. This is pretty common amount bootstrappers. I am speaking from my own experience at the company I founded: JotForm’s last 7 years were at 50%. Ahrefs recently announced a similar growth rate: https://medium.com/swlh/how-we-achieve-65-yoy-growth-by-ignoring-conventional-startup-advice-24a3eef619c1 https://medium.com/swlh/how-we-achieve-65-yoy-growth-by-igno...
aaavl2821 8y ago
The DCF is an alluring but dangerous way to value a company. It is alluring because it seems like a much more logical, first-principles way of valuing a company compared to slapping a market-based revenue multiple on a company's projected sales. It is alluring because it is precise. It is dangerous because it is not accurate. DCFs are very sensitive to assumptions, and confidence intervals for most assumptions are very wide. Two DCF models with credible, but different assumptions can yield hugely different valuations. DCFs are also dangerous because stocks are not valued solely based on fundamental cash flows. In startups especially, if you do a typical DCF with conservative assumptions, you will miss outlier returns when an acquirer's thesis hinges upon very aggressive assumptions or synergies that a DCF wouldn't capture. This happens all the time in biotech. It is not true that DCFs are independent of how the market values things. Many key inputs into the DCF, especially the discount rate and terminal value, are calculated in part based on how the market values things. In practice, investment bankers and investors use several different methodologies, all somewhat flawed, to triangulate around a valuation. Often the DCF yields the highest valuation of these methodologies.
- pstrazzulla 8y ago
  You're very spot on! DCFs are just one more way to triangulate. However, I think they can be very telling around where value drivers are, what what assumptions have to be in order to get to certain outcomes. It's definitely more useful than "SaaS companies trade at 7x forward revenues" as a DCF can help you capture the idiosyncrasies of your investment.
  - aaavl2821 8y ago
    I agree that DCFs are valuable in driving into assumptions. Taking a few "market" DCFs and seeing for which assumptions your view diverges from market is very helpful
    - amygdyl 8y ago
      "Some things I prefer to do in private." - Warren Buffet on DCF valuation.
FatDrunknStupid 8y ago
I feel like I'm on Mushrooms here whilst everyone is drinking cool-aid. Phil will tell you what the best software is if you pay him. Sorry mate.

80 points by pstrazzulla 8y ago

great write up. always been curious about this... how do you value growth... especially as Uber / Lyft IPO and face slowing growth. What basis do you use to get the discount rate of a SaaS company? This seems fairly arbitrary and is a key part to valuing a company

pstrazzulla 8y ago
Growth is important, especially if gross margins are high and >50% of new revenues goes to the bottom line to decrease losses or eventually increase profits. It's tough though, the real questions becomes when does growth stop, which is a function of market size and profitable customer acquisition channels. The discount rate is fairly arbitrary. You can use something like the CAPM to get to your cost of equity, but it's more designed for slower growth companies. I prefer to think of the gambler analogy where you're trying to figure out the percent chance of getting paid back. Of course, it's SO HARD to figure out the probability of a startup continuing on its current path given the many variables. This is where the "artists" of the world can capture a lot of value.
clairity 8y ago
> "What basis do you use to get the discount rate of a SaaS company? This seems fairly arbitrary and is a key part to valuing a company" this is one of the things you learn in an MBA program. you'd generally use multiple methods to triangulate a rational estimate. for example, you'd look at industry comparables and estimate a beta to plug into CAPM (as sibling commenter touches on). you can also do full financial projections (5-7 years out) based on expected performance, do DCF, and monte carlo that to back out a discount rate. you can also do a comparative ratio analysis (https://www.investopedia.com/terms/r/ratioanalysis.asp https://www.investopedia.com/terms/r/ratioanalysis.asp) on profitability, cash flow, asset efficiency, turnover, and the like. from my (limited) experience, startup valuations seem to be most sensitive to growth rate and the discount rate, so modelers spend a lot of time estimating these factors.
- pstrazzulla 8y ago
  This guy is basically the god of the DCF: http://aswathdamodaran.blogspot.com/search?q=dcf http://aswathdamodaran.blogspot.com/search?q=dcf
  - clairity 8y ago
    ha, yes, we used damodaran's book in my corporate valuation class! i forgot to mention that you can also employ an option pricing model, like black-scholes or the binomial model (in simpler cases), in cases with multiple classes of equities, wherein you can back out a valuation (and subsequently, a discount rate) based what the various VCs involved are expecting.
    - pstrazzulla 8y ago
      Very true. Especially since the equity value of a startup basically looks like an out of the money call option where you're paying a small amount now for a potential large payoff of our money (but high chance of $0 outcome). Also, more volatility will increase the value of this option (whether it's founder or market volatility - aka Travis from Uber or Cypto or Cannabis).

ttul 8y ago

I wanted to know the long-term value of my recurring revenue company. Being a programmer at heart, I wrote a python script that tested out 100,000 random scenarios of customer churn over the next 50 years. This gave me a histogram of future cash flows from the business, allowing me to state with some level of certainty what the range of future cash flow was likely to be. I was really surprised at how valuable the company is when analyzed in this manner. And there’s no bullshit. You assume a churn rate and randomly try it out over thousands of scenarios.

robodale 8y ago
Hey...you...ummm <cough> <cough>...wouldn't mind sharing that script?
harlanji 8y ago
What approach did you use? I wrote a discrete event sim from scratch once and it was a good experience, simulating adding express lines to a bank for certain transactions. I can see it being overkill, or possibly a good method for tuning dials, depending what you did. I have a sim that I've been kicking around for a couple of months related to my current work environment that I think would help make informed arguments about rule changes over the long term.
- ttul 8y ago
  I have begun using Monte Carlo simulations for various modeling purposes at the company over the last two years. For example, I created a script to model a price increase and the expected outcome from that increase given a certain rate of churn. That gave me the confidence to go ahead with a 30% increase knowing that in the median case we would end up with more revenue even if some huge number of customers said bye bye. The increase worked, and churn was lower than expected.
  - harlanji 8y ago
    Thanks for that and the business application, great result. This seems like a good method to make the case for serving more quality sensitive markets against the grain; that’s where I am coming from. Have heard the term, imagine it’ll be added to my tool shed soon. Thanks again.

justchilly 8y ago

The assumptions in your screen capture are dramatically more aggressive than any company in history. E.g. You are showing 60% yoy revenue growth 10 years in. For comparison: Google in year 10: 31%. Amazon in year 10: 23%. Dropbox in year 10: 27%. If these kinds of growth rates were possible, there would likely be much higher discount rates to go with them. Lots of investors do long term DCFs like this (and almost all update 5-year models on a rolling basis).They're just not as useful as a shorter term model given lack of visibility that far into the future and likelyhood of having to return funds well before then.

pstrazzulla 8y ago
TWLO, SHOP, etc are all growing at >50% 10 years in. It's very likely and common, especially in b2b SaaS where it's challenging to capture all of a fragmented market and/or you have a case where you can continually upsell with more product over time.
aytekin 8y ago
Yes, once you scale really big it is hard to keep a high growth rate. But if you have been always growing at 50% YoY, it is possible to keep that rate even at year 12. This is pretty common amount bootstrappers. I am speaking from my own experience at the company I founded: JotForm’s last 7 years were at 50%. Ahrefs recently announced a similar growth rate: https://medium.com/swlh/how-we-achieve-65-yoy-growth-by-ignoring-conventional-startup-advice-24a3eef619c1 https://medium.com/swlh/how-we-achieve-65-yoy-growth-by-igno...

aaavl2821 8y ago

The DCF is an alluring but dangerous way to value a company. It is alluring because it seems like a much more logical, first-principles way of valuing a company compared to slapping a market-based revenue multiple on a company's projected sales. It is alluring because it is precise. It is dangerous because it is not accurate. DCFs are very sensitive to assumptions, and confidence intervals for most assumptions are very wide. Two DCF models with credible, but different assumptions can yield hugely different valuations. DCFs are also dangerous because stocks are not valued solely based on fundamental cash flows. In startups especially, if you do a typical DCF with conservative assumptions, you will miss outlier returns when an acquirer's thesis hinges upon very aggressive assumptions or synergies that a DCF wouldn't capture. This happens all the time in biotech. It is not true that DCFs are independent of how the market values things. Many key inputs into the DCF, especially the discount rate and terminal value, are calculated in part based on how the market values things. In practice, investment bankers and investors use several different methodologies, all somewhat flawed, to triangulate around a valuation. Often the DCF yields the highest valuation of these methodologies.

pstrazzulla 8y ago
You're very spot on! DCFs are just one more way to triangulate. However, I think they can be very telling around where value drivers are, what what assumptions have to be in order to get to certain outcomes. It's definitely more useful than "SaaS companies trade at 7x forward revenues" as a DCF can help you capture the idiosyncrasies of your investment.
- aaavl2821 8y ago
  I agree that DCFs are valuable in driving into assumptions. Taking a few "market" DCFs and seeing for which assumptions your view diverges from market is very helpful
  - amygdyl 8y ago
    "Some things I prefer to do in private." - Warren Buffet on DCF valuation.

FatDrunknStupid 8y ago

I feel like I'm on Mushrooms here whilst everyone is drinking cool-aid. Phil will tell you what the best software is if you pay him. Sorry mate.