I want to think out loud about Reddit $RDDT after its latest quarter, mainly because of management’s plan to move to a logged-in-only experience and, over the medium term, roll it out across the whole platform. The critics have a point. Logged-out traffic, much of it coming from search, has historically been Reddit’s most important user acquisition channel, and maybe more importantly, it has been free. Putting that traffic – which will be much less in the first place as Google moves to AI overviews – behind a login wall will almost certainly slow the funnel (sort of a double whammy), both for casual visitors who never sign up and for people who might eventually have become loyal logged-in users. Where I am unsure is how much that matters for long-term intrinsic value. Reddit earns gross margins of over 90%, and that gives it a lot of room to buy the users it used to get for free. Paying for acquisition will never look as good as free acquisition, because the return on free acquisition is basically infinite. Still, as Mauboussin keeps pointing out, value gets created whenever a company reinvests above its cost of capital, even if the headline return on invested capital comes down along the way. The question then becomes what Reddit can afford to pay per user. With US ARPU at $13-14 a quarter, a US user brings in roughly $50 of gross profit a year. If that user sticks around for three years and I insist on lifetime value of three times the acquisition cost, Reddit could spend around $50 per user and still earn an attractive return. That’s before ARPU keeps growing towards what Meta earns from its apps, which is a ceiling Reddit won’t reach anytime soon but shows how much runway is left. Internationally, at roughly $9 of gross profit per user a year, the math becomes harder. So if Reddit gives up some margin and puts it into acquisition spend at attractive returns, I don’t see why intrinsic value can’t keep compounding. Would probably be a fun project to do more work on the user acquisition math.
View original →
