3 ms·

So, growing 100% from 10,000 to 20,000 users allowed App.net to drop annual pricing from $50 to $36, which implies their expenses grew by only 44% (from around

by cs702 14y ago

So, growing 100% from 10,000 to 20,000 users allowed App.net to drop annual pricing from $50 to $36, which implies their expenses grew by only 44% (from around $500,000 to around $720,000). If expenses continue to grow at similar fractions of the rate of user growth, App.net should be able to price the service for around $5 per year with just over a million users. At that pricing point, App.net would become a compelling economic proposition for many consumers.

jmduke 14y ago
How is $5/year with one million users more compelling than free/year with five hundred million users?
- ceejayoz 14y ago
  Not having to deal with 300 million spam users?
  - hobonumber1 14y ago
    How is it spam if you're responsible for choosing who you listen to vs who you ignore?
    - dannyr 14y ago
      If you want to interact with the community, it's an issue. Post something on Twitter mentioning "IPad", you'll get a lot of spam accounts replying. It's hard to engage with other users if you have to comb through the spam.
    - ceejayoz 14y ago
      You choose who you follow, but you don't choose who @replies to you.
  - ChrisLTD 14y ago
    The reason parts of the tech celebrity circle (like Gruber, Siracusa, and Arment) supported App.net was because of Twitter's new hostility to third-party developers, not because of spam.
- rexreed 14y ago
  It may not be, but the odds of building another free-to-use 500 Million+ social network without raising a $1B in the process are pretty slim. I think they are trying to prove that an alternative exists to ad-driven business models for growth (the people are the product, not the customers). In which case, they aren't going for size, but for value. That's why there must be a cost.
rexreed 14y ago
I don't think this has anything to do with cost-based pricing. In a non-commodity market, you don't charge based on how much it costs... you charge based on how much the market will bear. I don't think the choice of $36 vs. $50 has anything to do with a 44% expense rate. It's simple. It's hard to charge $50 a year ($4.17/mo) when you also have a $5/mo plan ($60/year). It's all driven by the $5/mo plan. Simply put, compared to $5/mo, $50 a year is not compelling enough to motivate annual subscriptions. This makes their $36/year ($3/mo) a compelling purchase to motivate annual subscriptions. Likewise, they actually make more money with people on a monthly plan, assuming they have a customer LTV > 12 months. So, they're making $60/yr on monthly subscribers (20% more +$10), and their annual plan, while a 23% reduction from previous pricing (-$14), which means only a $4 discount on an average customer basis assuming they have an even mix of monthlies who stick around at least 12 months and their annuals. So, they would just need to grow their market such that they only need to have more monthlies than annuals, and they're making MORE money in the aggregate. (Do the math -- you'll see a mix of 58.33% monthlies and 41.67% annuals on the new plan is roughly break even with 100% annuals on the old plan). Or even more starkly to show how this works -- if they didn't grow their annual base AT ALL, and dropped the price, they would only need 23.4% as many monthlies as annuals to break even to cover the difference. So, if they had 10,000 annuals, even if they didn't grow that at all and gave them all a price break, they would only need 2,334 monthlies to make up the difference (just under 19% of the total combined audience). And odds are high that monthlies will be a lot more than 19% of their total base. Oh, and they'll probably grow their annual subscriptions anyways. They can see that their rate of growth for monthlies, combined with a more compelling discount for an annual program will actually increase the bottom line. Solid pricing strategy.
- teawithcarl 14y ago
  Well said. Its a wide-open place and welcomes engineering opinions ... the price (+/-) has been hotly debated inside App.net. I admire Dalton for moving down ONLY slightly on price. He actually raised it on the monthly, yet allowed people to test cheaply. It's worth the premium, Dalton is wise not to give it away. PS - members are a dream demographic, and spend $5 on apps easily. Some developers have made $ thousands via app sales already. The developer revenue share model (just announced) accrues additionally on top of app sales each month.
teawithcarl 14y ago
Two points. 1) App.net expenses are quite low. Nowhere near your $500K-$720K range. Since it's not "free", it's going to be quite profitable. 2) App.net is NOT shooting for 1 million users. Not yet. Paul Graham challenged people in his PyCon keynote address to "build a Google Search for the top 10,000 geeks on earth". Dalton is responding to that in "social". He's not trying to scale quickly to 1 million users ... rather, he's wisely getting 10,000 top geeks involved, by sharing revenue and building a blackbox with integrity long term. Rather than falling for a "must scale like Twitter" mindset ... ponder why Paul Graham himself suggested a smaller, ultra-geek laser focus. Dalton is executing beautifully. Paying developers well is smart, and it's profitable. Economically when it does scale, this model has the potential to be off the charts. Patience.
- waterlesscloud 14y ago
  A search engine is judged by its utility, which is why having 10,000 geeks use it is a good sign. A social network is not judged by the same sort of criteria, and so it's not at all clear to me why someone would apply the same strategy.
  - nasmorn 14y ago
    I actually think of search as a heavily branded product. Like sugary drinks. Sure there are people that try somerhing new sometimes but that doesnt hurt the behemoths. I wonder if it is even possible to make a searchengine so much better than the "googles" that people referring to it as such could notice.
lkrubner 14y ago
That would be true if App.net pays out $20,000 a month to developers, for forever. App.net promises to pay $20,000 to developers, but is that a fixed cost? If it is a fixed cost, then you are right, the more people who sign up, the lower the subscription price can be. A million users could fund $20,000 a month very easily. But is $20,000 a month fair to the developers? By the time App.net has a million users, we can assume there will be a lot of apps competing for a slice of that $20,000. What happens when there are 1,000 apps in the eco-system. Do they all get $20 a month? If yes, then hasn't App.net failed at its primary stated mission, of helping fund app development? If App.net is to remain true to its mission, then that $20,000 should go up over time. And if it goes up over time, it is no longer so clear that subscription levels can remain as low as they are now. We should ask how much money App.net is making. Similar services are upfront about what percentage they take from the subscriber's money. App.net needs to offer at least that level of transparency.

by cs702 14y ago

jmduke 14y ago

How is $5/year with one million users more compelling than free/year with five hundred million users?

ceejayoz 14y ago
Not having to deal with 300 million spam users?
- hobonumber1 14y ago
  How is it spam if you're responsible for choosing who you listen to vs who you ignore?
  - dannyr 14y ago
    If you want to interact with the community, it's an issue. Post something on Twitter mentioning "IPad", you'll get a lot of spam accounts replying. It's hard to engage with other users if you have to comb through the spam.
  - ceejayoz 14y ago
    You choose who you follow, but you don't choose who @replies to you.
- ChrisLTD 14y ago
  The reason parts of the tech celebrity circle (like Gruber, Siracusa, and Arment) supported App.net was because of Twitter's new hostility to third-party developers, not because of spam.
rexreed 14y ago
It may not be, but the odds of building another free-to-use 500 Million+ social network without raising a $1B in the process are pretty slim. I think they are trying to prove that an alternative exists to ad-driven business models for growth (the people are the product, not the customers). In which case, they aren't going for size, but for value. That's why there must be a cost.

rexreed 14y ago

I don't think this has anything to do with cost-based pricing. In a non-commodity market, you don't charge based on how much it costs... you charge based on how much the market will bear. I don't think the choice of $36 vs. $50 has anything to do with a 44% expense rate. It's simple. It's hard to charge $50 a year ($4.17/mo) when you also have a $5/mo plan ($60/year). It's all driven by the $5/mo plan. Simply put, compared to $5/mo, $50 a year is not compelling enough to motivate annual subscriptions. This makes their $36/year ($3/mo) a compelling purchase to motivate annual subscriptions. Likewise, they actually make more money with people on a monthly plan, assuming they have a customer LTV > 12 months. So, they're making $60/yr on monthly subscribers (20% more +$10), and their annual plan, while a 23% reduction from previous pricing (-$14), which means only a $4 discount on an average customer basis assuming they have an even mix of monthlies who stick around at least 12 months and their annuals. So, they would just need to grow their market such that they only need to have more monthlies than annuals, and they're making MORE money in the aggregate. (Do the math -- you'll see a mix of 58.33% monthlies and 41.67% annuals on the new plan is roughly break even with 100% annuals on the old plan). Or even more starkly to show how this works -- if they didn't grow their annual base AT ALL, and dropped the price, they would only need 23.4% as many monthlies as annuals to break even to cover the difference. So, if they had 10,000 annuals, even if they didn't grow that at all and gave them all a price break, they would only need 2,334 monthlies to make up the difference (just under 19% of the total combined audience). And odds are high that monthlies will be a lot more than 19% of their total base. Oh, and they'll probably grow their annual subscriptions anyways. They can see that their rate of growth for monthlies, combined with a more compelling discount for an annual program will actually increase the bottom line. Solid pricing strategy.

teawithcarl 14y ago
Well said. Its a wide-open place and welcomes engineering opinions ... the price (+/-) has been hotly debated inside App.net. I admire Dalton for moving down ONLY slightly on price. He actually raised it on the monthly, yet allowed people to test cheaply. It's worth the premium, Dalton is wise not to give it away. PS - members are a dream demographic, and spend $5 on apps easily. Some developers have made $ thousands via app sales already. The developer revenue share model (just announced) accrues additionally on top of app sales each month.

teawithcarl 14y ago

Two points. 1) App.net expenses are quite low. Nowhere near your $500K-$720K range. Since it's not "free", it's going to be quite profitable. 2) App.net is NOT shooting for 1 million users. Not yet. Paul Graham challenged people in his PyCon keynote address to "build a Google Search for the top 10,000 geeks on earth". Dalton is responding to that in "social". He's not trying to scale quickly to 1 million users ... rather, he's wisely getting 10,000 top geeks involved, by sharing revenue and building a blackbox with integrity long term. Rather than falling for a "must scale like Twitter" mindset ... ponder why Paul Graham himself suggested a smaller, ultra-geek laser focus. Dalton is executing beautifully. Paying developers well is smart, and it's profitable. Economically when it does scale, this model has the potential to be off the charts. Patience.

waterlesscloud 14y ago
A search engine is judged by its utility, which is why having 10,000 geeks use it is a good sign. A social network is not judged by the same sort of criteria, and so it's not at all clear to me why someone would apply the same strategy.
- nasmorn 14y ago
  I actually think of search as a heavily branded product. Like sugary drinks. Sure there are people that try somerhing new sometimes but that doesnt hurt the behemoths. I wonder if it is even possible to make a searchengine so much better than the "googles" that people referring to it as such could notice.

lkrubner 14y ago

That would be true if App.net pays out $20,000 a month to developers, for forever. App.net promises to pay $20,000 to developers, but is that a fixed cost? If it is a fixed cost, then you are right, the more people who sign up, the lower the subscription price can be. A million users could fund $20,000 a month very easily. But is $20,000 a month fair to the developers? By the time App.net has a million users, we can assume there will be a lot of apps competing for a slice of that $20,000. What happens when there are 1,000 apps in the eco-system. Do they all get $20 a month? If yes, then hasn't App.net failed at its primary stated mission, of helping fund app development? If App.net is to remain true to its mission, then that $20,000 should go up over time. And if it goes up over time, it is no longer so clear that subscription levels can remain as low as they are now. We should ask how much money App.net is making. Similar services are upfront about what percentage they take from the subscriber's money. App.net needs to offer at least that level of transparency.