Here is my investment thesis. What’s yours?
“Businesses that disrupt/create huge markets with a non-replicable product offering a must-have service, which gets better the more people use it.”
Jean de la Rochebrochard (@jlr), a partner at TheFamily, recently published a very good piece on Medium titled “Here’s my investment thesis. What’s yours?” — below is my answer to his question.
When thinking about investment thesis, one has to ask oneself: what is the outcome I am aiming at for the companies I invest in? At Index, the answer is very clear: we want our investments to exit for > $1 billion in less than ten years post our investment — either through an IPO or a large trade sale. We are hunting “unicorns”.
Given that these exits are typically brokered by bulge bracket investment banks, one way to analyse them is through their lenses. What do they evaluate in a potential client? Four things: 1. Revenue, 2. Growth, 3. Margin, and 4. Risk.
And that’s it. Because these are the only four variables needed to build a Discounted Cash Flow (DCF) model — the closest thing there is to a “fair value” estimate for a company.
So here are the four key questions one needs to answer to estimate the unicorn potential of an investment:
1. Can it generate large enough revenues?
(for eg. at UBS we would not work on any transaction lower than $500m — at 10x current revenue multiple, this implies a $50m net revenue business as a minimum)
This obvious criteria has one essential corollary: the company must operate on a large enough market. Which means that they need to have enough people willing to pay a high enough price for whatever product/service they are offering.
There are two ways to size a market opportunity: top-down (adding up the revenues of all the companies operating on that market today) and bottom-up (coming up with an estimate for how many people would be willing to pay for the service, and at what price). Both approaches complement each other, especially for innovative tech companies which typically disrupt an existing market by undercutting the incumbents on the one hand (and hence shrinking the market), while creating a new use case attracting a larger number of new users on the other hand.
2. How quickly can it grow to reach this minimum revenue scale, and beyond?
It is (relatively) easy to grow fast for the first few years of the life of a startup, but maintaining that growth past $10m and then past $100m is a whole different game. The company will need to keep acquiring an ever greater number of users to overcome the natural churn of their growing user base, in order to maintain not only absolute growth, but also relative growth. The public markets only tolerate the risk of tech startups because they offer unparalleled growth rates.
The largest outcomes tend to have one common feature: something in their core product allows them to grow faster over time, while bringing acquisition costs lower — be it thanks to the network effects of peer-to-peer models or the natural virality of communication tools.
This is what enables this type of exponential charts:
3. How profitable can it be?
In the early days, “traction” is all that matters, whatever metric is being used to define it. But in the long run, Monthly Active Users (MAU) and other “activity” metrics, do not appear anywhere in a DCF model: net profit is the only number that counts.
Three main components determine the profit margin and its future evolution: variable costs (servers, costs of goods sold), fixed costs (salaries, rent) and marketing. Most software businesses have insignificant variable costs and high fixed costs. Marketing is where their economics differ widely and hence what VCs spend most of their time digging in. In other words: building a product always costs roughly the same, convincing people to use it is where costs differ widely and what sets apart exceptional companies from the rest.
The best businesses do not need to pay people anything to get them to use their product. Users either love it so much or need it so badly that they cannot imagine living without.
And, as importantly, they cannot find it anywhere else. The product has to offer a unique and proprietary selection (be it a feature, a technology, an inventory or a network of people).
If the company offers a non-replicable, must-have product, that sells itself with virtually no marketing, then its marginal cost will trend toward zero and margins will keep on expanding as the business grows.
4. How predictable are these profits in the long run?
This is the final piece of the investment puzzle: the defensibility of the business. The “moat” that investors and entrepreneurs are so obsessed about building around their activity. This concept has a financial translation too: the Weighted Average Cost of Capital (WACC), ie. the factor future profits get discounted by to value the business. The riskier the business is perceived, the higher its cost of debt and equity, and the lower the present value of its future profits.
Fred Wilson at USV recently posted their view about it and I think it sums it all up perfectly:
“So we asked ourselves, “what will provide defensibility” and the answer we came to was networks of users, transactions, or data inside the software.”
In other words, the best companies literally suck the air out of their market as they end up capturing an ever larger part of the users’ actions, which they then leverage to make their product better/smarter.
Blablacar’s offering keeps on getting better the more users join the network (as users become more likely to find a suitable date/time for their door-to-door journey), and the harder it becomes for anyone new to come in and disrupt them. Swiftkey’s product keeps on getting better the more users type through the app as the company gets access to more live data to improve their prediction engine, making it significantly harder for a new entrant to create a better product from scratch.
These characteristics make it very easy to plot the line of future earnings for these businesses, and lowering the WACC has as much impact on the value of a company than increasing the growth rate.
Conclusion
In the long run, these are the only four questions that matter to determine the potential of a business and whether it will be able to reach the “unicorn club”: how large can the revenues be? How fast can they grow? What costs will be needed to support them? How predictable are they going to be five/ten years from now? Obviously, the earlier the investment, the less precise these answers will be, but, whatever the stage, these questions remain valid.
And the outlier businesses all share the same characteristics: they disrupt/create huge markets with a non-replicable product offering a must-have service that keeps on getting better the more people use it.
BONUS
Here is a “quick and dirty” DCF model built using the Soulver app, which shows a back-of-the-envelope calculation of what is required to reach a $1bn valuation using this method.
