Hacker News new | ask | show | jobs
by jyxent 1100 days ago
Reddit reported that they had 430 million monthly active users in 2019.

$500 million per year / 430 million users = $1.16 per user per year

$1.16 / 12 months = $0.10 per user per month.

2 comments

Pretty big difference between users who open the site once a month to look at a meme someone sent them vs those who are online for multiple hours a day. The ones who use third party apps are more likely to be in the latter bucket, and the loss of revenue from them will be significant more than the bottom tier of users.
Revenue is generated by ads. The power users are also the savvy users with ad block. Their value added is making the content for average users to engage and consume. Those are the users that will make up the base of the as revenue.

Alienating power user seems like the exact opposite.

Let's do some back-of-the-napkin math. Selig estimates that the average user would cost him $2.50/mo and his annual costs would be $20m/yr, so it seems like Apollo has 600–700k monthly active users — something like 0.16% of all 430m active Reddit users. That is simply not going to represent a significant opportunity cost to Reddit's overall ad revenue.

Plus, I'd bet that third-party app users are far more likely to be mods and/or heavy content creators, which means that alienating them will have outsized negative effects on the engagement of the rest of the userbase.

Where do you account for opportunity cost and future growth/loss in these (also old and also imprecise) estimates?
Ok, multiply that by 5, it’s still nowhere near 2.5.
Well, guess what? Your starting point is just an estimate and its accuracy cannot be sourced, so this is all meaningless
Exactly. Regardless of estimates, 20 mil to run custom Reddit client is ridiculous.