4 ms·

Totally agree that Reddit are being glorious ass-munching hentais here. I fucking hate all that Reddit has become as a company and as a product. But also, dude

by dcow 3y ago

Totally agree that Reddit are being glorious ass-munching hentais here. I fucking hate all that Reddit has become as a company and as a product. But also, dude just raise your prices. I read the whole announcement and truly don't understand why Apollo can't be $20/year. I don't know anybody who attributes a meaningful difference to $10/year and $20/year. I'm not a user but if I was faced with that type of price change and some language around needing to adjust pricing because Reddit is now charging for API access, I'd not give it a second thought. It really really seems weird to want to die on this hill when you don't need to. Maybe it is the harbinger of the end for Reddit and we're just overdue. But I see no reason the founder of a popular Reddit reader couldn't secure some temporary funding to weather the transition, or simply negotiate a longer lead time rather than spending all the time in talks and ugly back and forth.

flutas 3y ago
> I don't know anybody who attributes a meaningful difference to $10/year and $20/year. You forgot to factor in other things. Base API cost: $2.52/mo (0.00024/call @ 345req/day) App Store Taxes / Fees: $1.08/mo (I doubt he qualifies for the reduced fees at this point) Just adding the appstore tax makes this $43.20 a year. That doesn't factor in any servers he has to pay for (push notifications). The app dev fee $99 per year, not a huge amount but small parts add up. Add in the cost to pay his server engineer, or any profit for him to live off (likely less users, so has to be more $ per user) of while making the app and it probably goes to something like $60/yr.
- dcow 3y ago
  He said he could make the pricing work at half what reddit is asking, for $1.26/mo base API cost. I assume he has the best insight into his own business. I was just conservatively spitballing that doubling his revenue would mean he could manage the price Reddit is asking (since he said he could make due with half). Just looking at it from the other angle.
  - flutas 3y ago
    > He said he could make the pricing work at half what reddit is asking, for $1.26/mo base API cost. He said a much more reasonable thing would be to cut the price in half and give a 3 month transition period to make it "feasible for more developers, myself included." > However in a perfect world I think lowering the price by half and providing a three month transition period to the paid API would make the transition feasible for more developers, myself included. These concessions seem minor and reasonable in the face of the changes. What that would likely mean is removing as many API calls as possible and removing features as a result. Which means fewer users would want to pay for it. > I was just spitballing that doubling his revenue would mean he could manage the price Reddit is asking (since he said he could make due with half). Also, as a tidbit. His current subscription pricing is $5/mo for ultra. If we want to take that as his revenue for an ongoing subscription to double (since API access is going to be monthly), then the app would be $10/mo or $120/yr.
    - dcow 3y ago
      Fair. That definitely changes the situation a lot.
    - zamadatix 3y ago
      What, then, are you arguing his proposal was making feasible?
zamadatix 3y ago
The "Why not just increase the price of Apollo" section says he's already sold 50,000 yearly subscriptions at normal price. He's going to offer a pro-rated refund and, at that point, he'll be out $250,000 of revenue he already recognized while still having all of the lifetime user loss to contend with. Even if he could optimize the API usage to halve it in the next 30 days that's still $1.25/m ($15/y) in API fees per user. After the App Store cut, the $20/y would only barely cover API cost of users who sign back up. Something like $30/y might make more sense to cover overall cost and bring in profit but then it comes to the question of "is it really worth trying to put all of this effort in to save a fraction of the userbase and MRR with a company that won't even give him more than a months notice to make these kinds of changes". While I think there is a way to keep Apollo running in some form or another, I by no means blame them for going to this hill to let it die on. Waaaay too much work for far too little reward to have the burden of massive risk from dependency on Reddit's future whims lingering in the background.

by dcow 3y ago

flutas 3y ago

> I don't know anybody who attributes a meaningful difference to $10/year and $20/year. You forgot to factor in other things. Base API cost: $2.52/mo (0.00024/call @ 345req/day) App Store Taxes / Fees: $1.08/mo (I doubt he qualifies for the reduced fees at this point) Just adding the appstore tax makes this $43.20 a year. That doesn't factor in any servers he has to pay for (push notifications). The app dev fee $99 per year, not a huge amount but small parts add up. Add in the cost to pay his server engineer, or any profit for him to live off (likely less users, so has to be more $ per user) of while making the app and it probably goes to something like $60/yr.

dcow 3y ago
He said he could make the pricing work at half what reddit is asking, for $1.26/mo base API cost. I assume he has the best insight into his own business. I was just conservatively spitballing that doubling his revenue would mean he could manage the price Reddit is asking (since he said he could make due with half). Just looking at it from the other angle.
- flutas 3y ago
  > He said he could make the pricing work at half what reddit is asking, for $1.26/mo base API cost. He said a much more reasonable thing would be to cut the price in half and give a 3 month transition period to make it "feasible for more developers, myself included." > However in a perfect world I think lowering the price by half and providing a three month transition period to the paid API would make the transition feasible for more developers, myself included. These concessions seem minor and reasonable in the face of the changes. What that would likely mean is removing as many API calls as possible and removing features as a result. Which means fewer users would want to pay for it. > I was just spitballing that doubling his revenue would mean he could manage the price Reddit is asking (since he said he could make due with half). Also, as a tidbit. His current subscription pricing is $5/mo for ultra. If we want to take that as his revenue for an ongoing subscription to double (since API access is going to be monthly), then the app would be $10/mo or $120/yr.
  - dcow 3y ago
    Fair. That definitely changes the situation a lot.
  - zamadatix 3y ago
    What, then, are you arguing his proposal was making feasible?

zamadatix 3y ago

The "Why not just increase the price of Apollo" section says he's already sold 50,000 yearly subscriptions at normal price. He's going to offer a pro-rated refund and, at that point, he'll be out $250,000 of revenue he already recognized while still having all of the lifetime user loss to contend with. Even if he could optimize the API usage to halve it in the next 30 days that's still $1.25/m ($15/y) in API fees per user. After the App Store cut, the $20/y would only barely cover API cost of users who sign back up. Something like $30/y might make more sense to cover overall cost and bring in profit but then it comes to the question of "is it really worth trying to put all of this effort in to save a fraction of the userbase and MRR with a company that won't even give him more than a months notice to make these kinds of changes". While I think there is a way to keep Apollo running in some form or another, I by no means blame them for going to this hill to let it die on. Waaaay too much work for far too little reward to have the burden of massive risk from dependency on Reddit's future whims lingering in the background.