Follow-on reserves

We’ve invested slightly north of $60M into Turkish startups so far. About $30M of this number went towards our initial investments into companies, and the remaining $30M went towards follow-on investments into companies where we were already shareholders. This means that, so far, we have a 1:1 initial:follow-on investment ratio.

In startup markets with abundant capital, a 1:1 initial:follow-on investment ratio is reasonable. Some investors prefer a 1:2 ratio, but a greater allocation towards follow-on investments is rarely seen. Because there’s a lot of capital in the market, it’s easier for your best performing companies to get the follow-on capital they deserve from other investors. Funds therefore don’t need to keep this capital in reserve themselves.

In Turkey, I think that a 1:1 or even 1:2 initial:follow-on investment ratio isn’t enough. Although a company may be performing very well, since there’s limited capital available, they may not be able to access the outside funding they deserve. As a result, investors like us who are often the first money into a company need to keep greater reserves to support these strong performing companies. My gut feeling is that a 1:3 or 1:4 initial:follow-on investment ratio is ideal for investors operating in a market like Turkey.

Another reason why startup investors need to keep more reserves in Turkey is because of the number of great startups. There are a smaller number of great startups than in markets with abundant capital (hence the reason for the abundant capital), so you need to have the ability to increase or at least retain your ownership in your winners to achieve great returns. If you choose to make more new investments rather than to keep this money to back your winners as you get more information about each company’s performance, your ownership stake in the winners that determine the success of your portfolio will be gradually reduced. You’ll end up with a much smaller ownership stake in these rare winners at exit, and that will hurt your returns.

Also published on Medium.