3 ms·

There’s a sense in which free ad-supported apps are “redistributive”. Whether you’re rich or poor, Facebook costs the same and the experience and features you h

There’s a sense in which free ad-supported apps are “redistributive”. Whether you’re rich or poor, Facebook costs the same and the experience and features you have access to are identical. Effectively, the rich users are subsidizing the app for the poor users (the rich users’ eyeballs being worth much more). I personally view that as a strong upside compared to paid models. I’m not sure where I’m going with this. What does paid social media look like? Perhaps more like traditional media: only deep pockets will have access to “virality” and wide distribution. I could certainly see a market for that with the Instagram crowd.

supernovae 5y ago
It's been years since there was organic viral videos... that ship sailed - we're at the point of pay for play, sex or extremely high risk/danger/endangerment content for eyeballs..
hedora 5y ago
In the past, we had paid social media (kind of) with services like CompuServe, AOL, etc. People tended to self-segment by demographics. Also, there was more than one network to choose from, since things were segmented by which internet connection you had. (This was also the heyday of BBS’s and Usenet, which probably wasn’t a coincidence.) Maybe having N vertically integrated communities with different norms is better than what facebook, twitter, etc built via their horizontally-integrated services.
- specialist 5y ago
  Most (all?) prior mediums were initially pay to play. Print, radio, telegraph, TV, cable, fax, ATMs. Then there's some kind of phase change. Like maybe when costs reduce enough for Metcalfe's Law to kick-in. [I don't quite recall what some other theorists have said. Shirky's observations about cost of coordination. McLuhan's distinctions between passive and active engagement (aka push vs pull).] Apple's privacy power play may be the first time someone manages to put the toothpaste back into the tube, for an entire medium and by extension ecosystem. Some mediums, like rail and phones, manage to remain pay to play. Maybe there's something to be learned by compare and contrast.
  - lanstin 5y ago
    Newspaper advertising never had an incentive to attack newspaper readers.
    - majewsky 5y ago
      > Newspaper advertising never had the means to attack newspaper readers. FTFY - They totally would have done it if it had been possible and profitable.
banach 5y ago
These are not the only alternatives. Messaging apps allow people to stay in touch with each other without any of the timeline corporate surveillance society nonsense.
arkadiyt 5y ago
DuckDuckGo has advertisements and is profitable. Nothing about the free-ad model for apps requires tracking users.
- adrr 5y ago
  Search ads you don’t need user targeting. You target on the search term and most likely it will be relevant for the user. Only standard display ads need targeting to be able to show relevant ads to the user.
  - lanstin 5y ago
    Content based ads still seem valuable. If I am on Don’s how to fix something pages, then add for something fixing tools seem possible to sell without involving my data
  - criddell 5y ago
    There's always the context of the page on which the ad will appear. If I search for "teach your kid how to read", the search engine can display ads based on the query and the destination page seems like it should show the exact same types of ads.
    - warkdarrior 5y ago
      How would the destination page know the exact type of ads to show, other than getting some info from the search engine? And wouldn't that be tracking?
      - criddell 5y ago
        There are non-tracking ad networks that will analyze the content of the page and insert relevant ads.
have_faith 5y ago
> the rich users are subsidizing the app for the poor users That would only hold if the majority of paid ad placements on a social media site targeted rich people. As is evident by the ads that appear surrounding elections, the ads targeting average people to dump their savings into crypto and all of the gambling (might as well be gambling...) gaming ads that doesn't seem like a safe assumption to make although I have no data to support either assumption.
- howinteresting 5y ago
  It is well established that the ARPU from richer demographics (income levels, countries etc) is much higher than that from poorer ones. The redistributive effect is real, though perhaps there may be other ways (e.g. reparations) that have less harmful side effects.
brundolf 5y ago
Maybe some room in the market will be made for federated and/or nonprofit social-media?
msoad 5y ago
This is something I thought about too. The dilemma for Facebook and Twitter to offer paid service is interesting Wealthy eyeballs worth more (a lot). Wealthy people most likely to pay to not see ads. A super high price to justify not showing ads to wealthy people will make the whole offering moot. I don't have the numbers but if it's anything like the rest of the economy, top 10% is probably spending something like 60% of total spent. So removing the top 10% from the targetable population for ads can be a huge hit to ad revenues.
phh 5y ago
I totally agree with you. I wouldn't necessarily say that rich people subsidize poor ones, other one we wouldn't see Facebook racing for Brazil and Google racing for India. But yes, one major property of ads is how it adapts to the user's revenue. I believe that anyone who wants to take on ads, need to understand this first. And to accomplish an ad-free realk, one needs to somehow have people pay differently based on their incomes. And I can't see any way without tracking on how to do that. One company who managed to do that, to some extent, is Netflix. Their price range from 3$/month (lowest tier in India) to 20$/month (highest tier in France), for pretty much the same service. It's still only one order of magnitude of adaptation. While ads scale from pretty much 0 (poor people who can "mis"spend only 5$/month), to I think maybe 100$ for some clicks. Basically, from my point of view, ads are a way to raise money based on what people can give, while removing the free-rider problem that doing it directly would raise. Can we make a useful crypto money that fixes this? (you're allowed to a big computation budget, since ads have a very expensive global budget)
crvdgc 5y ago
> Whether you’re rich or poor, Facebook costs the same and the experience and features you have access to are identical. Or you can say, the amount of work that's extracted from the user is the same. Because the major "service" of social media isn't the webapp that has a feed timeline, or a fancy UI, or availability (which can be easily copied by competitors). It's the network effect that comes from user's participation. It's very possible there's a net profit for the majority of users (either from clicking the ads, or from simply using the social media), so the "subsidized" users' loss is not a primary concern here.

supernovae 5y ago

It's been years since there was organic viral videos... that ship sailed - we're at the point of pay for play, sex or extremely high risk/danger/endangerment content for eyeballs..

hedora 5y ago

In the past, we had paid social media (kind of) with services like CompuServe, AOL, etc. People tended to self-segment by demographics. Also, there was more than one network to choose from, since things were segmented by which internet connection you had. (This was also the heyday of BBS’s and Usenet, which probably wasn’t a coincidence.) Maybe having N vertically integrated communities with different norms is better than what facebook, twitter, etc built via their horizontally-integrated services.

specialist 5y ago
Most (all?) prior mediums were initially pay to play. Print, radio, telegraph, TV, cable, fax, ATMs. Then there's some kind of phase change. Like maybe when costs reduce enough for Metcalfe's Law to kick-in. [I don't quite recall what some other theorists have said. Shirky's observations about cost of coordination. McLuhan's distinctions between passive and active engagement (aka push vs pull).] Apple's privacy power play may be the first time someone manages to put the toothpaste back into the tube, for an entire medium and by extension ecosystem. Some mediums, like rail and phones, manage to remain pay to play. Maybe there's something to be learned by compare and contrast.
- lanstin 5y ago
  Newspaper advertising never had an incentive to attack newspaper readers.
  - majewsky 5y ago
    > Newspaper advertising never had the means to attack newspaper readers. FTFY - They totally would have done it if it had been possible and profitable.

banach 5y ago

These are not the only alternatives. Messaging apps allow people to stay in touch with each other without any of the timeline corporate surveillance society nonsense.

arkadiyt 5y ago

DuckDuckGo has advertisements and is profitable. Nothing about the free-ad model for apps requires tracking users.

adrr 5y ago
Search ads you don’t need user targeting. You target on the search term and most likely it will be relevant for the user. Only standard display ads need targeting to be able to show relevant ads to the user.
- lanstin 5y ago
  Content based ads still seem valuable. If I am on Don’s how to fix something pages, then add for something fixing tools seem possible to sell without involving my data
- criddell 5y ago
  There's always the context of the page on which the ad will appear. If I search for "teach your kid how to read", the search engine can display ads based on the query and the destination page seems like it should show the exact same types of ads.
  - warkdarrior 5y ago
    How would the destination page know the exact type of ads to show, other than getting some info from the search engine? And wouldn't that be tracking?
    - criddell 5y ago
      There are non-tracking ad networks that will analyze the content of the page and insert relevant ads.

have_faith 5y ago

> the rich users are subsidizing the app for the poor users That would only hold if the majority of paid ad placements on a social media site targeted rich people. As is evident by the ads that appear surrounding elections, the ads targeting average people to dump their savings into crypto and all of the gambling (might as well be gambling...) gaming ads that doesn't seem like a safe assumption to make although I have no data to support either assumption.

howinteresting 5y ago
It is well established that the ARPU from richer demographics (income levels, countries etc) is much higher than that from poorer ones. The redistributive effect is real, though perhaps there may be other ways (e.g. reparations) that have less harmful side effects.

brundolf 5y ago

Maybe some room in the market will be made for federated and/or nonprofit social-media?

msoad 5y ago

This is something I thought about too. The dilemma for Facebook and Twitter to offer paid service is interesting Wealthy eyeballs worth more (a lot). Wealthy people most likely to pay to not see ads. A super high price to justify not showing ads to wealthy people will make the whole offering moot. I don't have the numbers but if it's anything like the rest of the economy, top 10% is probably spending something like 60% of total spent. So removing the top 10% from the targetable population for ads can be a huge hit to ad revenues.

phh 5y ago

I totally agree with you. I wouldn't necessarily say that rich people subsidize poor ones, other one we wouldn't see Facebook racing for Brazil and Google racing for India. But yes, one major property of ads is how it adapts to the user's revenue. I believe that anyone who wants to take on ads, need to understand this first. And to accomplish an ad-free realk, one needs to somehow have people pay differently based on their incomes. And I can't see any way without tracking on how to do that. One company who managed to do that, to some extent, is Netflix. Their price range from 3$/month (lowest tier in India) to 20$/month (highest tier in France), for pretty much the same service. It's still only one order of magnitude of adaptation. While ads scale from pretty much 0 (poor people who can "mis"spend only 5$/month), to I think maybe 100$ for some clicks. Basically, from my point of view, ads are a way to raise money based on what people can give, while removing the free-rider problem that doing it directly would raise. Can we make a useful crypto money that fixes this? (you're allowed to a big computation budget, since ads have a very expensive global budget)

crvdgc 5y ago

> Whether you’re rich or poor, Facebook costs the same and the experience and features you have access to are identical. Or you can say, the amount of work that's extracted from the user is the same. Because the major "service" of social media isn't the webapp that has a feed timeline, or a fancy UI, or availability (which can be easily copied by competitors). It's the network effect that comes from user's participation. It's very possible there's a net profit for the majority of users (either from clicking the ads, or from simply using the social media), so the "subsidized" users' loss is not a primary concern here.