4 ms·

How Large Will the Paid Feed Reader Market Be?

37 points by julien 14y ago

ambirex 14y ago
This is pretty fun thought experiment. Let's actually say CNN has close to 100% market penetration, making the number of users around 25 million. if say 5% of those would be willing to upgrade to a premium for $25/year that would make the potential size of the market $31 million a year. Just using numbers I'm pulling out of my posterior.
davidjohnstone 14y ago
I really should make some time and do the final 5% (and get it online) on a feed reader I spent a couple of weeks building a few weeks ago…
hpaavola 14y ago
Sorry for stealing this topic a bit, but could someone suggest a RSS reader which works decently on mobile browsers? Google Reader on mobile is really good; a heading, site name and first line from the article. I've tried NewsBlur, but it wants you to use an app, and The Old Reader, but it lists all your feeds first and the aggregated list takes way too much space. EDIT: I'd rather avoid free ones, unless I can host it myself. Don't want to look for a new reader again next year.
- apathetic 14y ago
  Have you tried feedly?
  - hpaavola 14y ago
    I have the impression that it requires an app, even desktop browsers need special add-ons.
- callahad 14y ago
  I'm not sure how receptive the maintainer would be, but NewsBlur is open source. Everyone would win if someone stepped up and put together a pull request for a nice workflow designed for the mobile web. I'd gladly chip in to kickstart that.
- TylerE 14y ago
  Why are you so opposed to an app? This seems like the absolute IDEAL use case for an app - the only thing going over the wire is the content, not formatting, a bunch of JS, etc.
  - devindotcom 14y ago
    When the content must frequently be viewed in a browser, it makes sense for the reader to be in the browser as well. An app seems far from ideal when it's just a moving window to web-based content.
  - pbreit 14y ago
    IMO, an app makes no sense at all. Sending just the data over the wire is not limited to apps. And as has been pointed out, you frequently end up going to a web page so why not start out in the browser (app-encapsulated browsers usually stink).
- ricardobeat 13y ago
  We are building a (non-free) RSS reader called Creek, you should sign up for the beta: http://creek.io http://creek.io
- sebsauvage 13y ago
  Did you try Kriss Feed ? Opensource, single php file, fast, no database needed, works great on mobiles, keyboard shortcuts, OPML import/export... https://github.com/tontof/kriss_feed https://github.com/tontof/kriss_feed Screenshots: http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_5.png http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_... http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_6.jpg http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_... I love it.
asb 14y ago
I suspect a decent percentage of those 25 million CNN RSS subscribers actually don't use RSS. They opened google reader one day, added CNN, and later forgot about it.
gz5 14y ago
Not sure CNN is good proxy for RSS - many folks not using RSS for CNN because they get CNN via Twitter etc or visit it x times/day anyways? Seems like much (half?) of RSS market is folks monitoring streams of websites that they don't visit regularly or monitor w/ non-RSS solutions?
- voidfiles 14y ago
  Out of all the feeds that I checked CNN had the largest number of Google Reader subscribers. If you are right, I would argue that CNN market penetration is lower then 50% and thus the feed reader market would be bigger.
  - gz5 14y ago
    Good point, I would guess it is lower than 50% of total (irrespective of active vs. not active). Not sure where you sourced your GR data but maybe the ratios of GR/uniques across a few different types of sites to find out if GR ratios varies by site type could sharpen the straw man.
    - voidfiles 14y ago
      Agreed.
speek 14y ago
For those of you who need your RSS fix: I made a little thing yesterday that sends you an email (I have it running daily via cronjob) with whatever news/information is relevant to you -> https://github.com/mheld/daily-digest https://github.com/mheld/daily-digest It's not pretty, but it sure as hell works for me (https://dl.dropboxusercontent.com/u/63726/Screen%20Shot%202013-04-18%20at%207.23.54%20AM.png https://dl.dropboxusercontent.com/u/63726/Screen%20Shot%2020...) Please feel free to fork it!
Kylekramer 14y ago
Ludicrously optimistic. Not even questioning that CNN's Google Reader stats is all active users just throws the entire calculation out of whack. I'd be shocked if the RSS market is larger than 15 million.
jostmey 14y ago
Google did not seem to think they could make money from it.
- jarek 14y ago
  Google did not seem to think they could make enough money from it to make it worthwhile to them∗
bdarnell 14y ago
The stats cited here include both Reader and iGoogle, so you can't really extrapolate from these numbers. CNN was disproportionately popular with iGoogle users (where it is a part of the default configuration for new US users) while Engadget was disproportionately popular with Reader users (where it is/was one of the featured suggestions for new users)
PaulHoule 14y ago
It's hard to say. Note that such a system might read messages from other protocols (email, SMS, newsgroups, facebook, twitter, whatever) and it might come across as so cool and essential that people really feel they need it. What I can tell you is that Bing and Google have serious legal, business and technical reasons why they can't offer services that have an interesting degree of personalization based on your usage patterns. (Yes, they localize and they do some simple adjustments of your search results to fool you into thinking they're smarter than they really are) (And funny, Duck Duck Go gets a huge amount of press for offering an innovation-free zero-personalization search engine while the companies trying to beat the 70% P@1 barrier Google and Bing face are stealth mode companies you've never heard of.) Anyhow, text analysis has been getting radically better in the last few years and we're really on the threshold of building something that can automatically construct an interest model for you and customize things based on preferences. The algorithms are computationally expensive so if you've got a ARPU as bad as Facebook you can't afford them, but you might be able to make a subscription service profitable before somebody can make an advertising-based service. Ultimately the RSS reader market has to give up on the "view hundreds of feeds side by side" model and give up all the excuses that it can't be done.

37 points by julien 14y ago

This is pretty fun thought experiment. Let's actually say CNN has close to 100% market penetration, making the number of users around 25 million. if say 5% of those would be willing to upgrade to a premium for $25/year that would make the potential size of the market $31 million a year. Just using numbers I'm pulling out of my posterior.

davidjohnstone 14y ago

I really should make some time and do the final 5% (and get it online) on a feed reader I spent a couple of weeks building a few weeks ago…

hpaavola 14y ago

Sorry for stealing this topic a bit, but could someone suggest a RSS reader which works decently on mobile browsers? Google Reader on mobile is really good; a heading, site name and first line from the article. I've tried NewsBlur, but it wants you to use an app, and The Old Reader, but it lists all your feeds first and the aggregated list takes way too much space. EDIT: I'd rather avoid free ones, unless I can host it myself. Don't want to look for a new reader again next year.

apathetic 14y ago
Have you tried feedly?
- hpaavola 14y ago
  I have the impression that it requires an app, even desktop browsers need special add-ons.
callahad 14y ago
I'm not sure how receptive the maintainer would be, but NewsBlur is open source. Everyone would win if someone stepped up and put together a pull request for a nice workflow designed for the mobile web. I'd gladly chip in to kickstart that.
TylerE 14y ago
Why are you so opposed to an app? This seems like the absolute IDEAL use case for an app - the only thing going over the wire is the content, not formatting, a bunch of JS, etc.
- devindotcom 14y ago
  When the content must frequently be viewed in a browser, it makes sense for the reader to be in the browser as well. An app seems far from ideal when it's just a moving window to web-based content.
- pbreit 14y ago
  IMO, an app makes no sense at all. Sending just the data over the wire is not limited to apps. And as has been pointed out, you frequently end up going to a web page so why not start out in the browser (app-encapsulated browsers usually stink).
ricardobeat 13y ago
We are building a (non-free) RSS reader called Creek, you should sign up for the beta: http://creek.io http://creek.io
sebsauvage 13y ago
Did you try Kriss Feed ? Opensource, single php file, fast, no database needed, works great on mobiles, keyboard shortcuts, OPML import/export... https://github.com/tontof/kriss_feed https://github.com/tontof/kriss_feed Screenshots: http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_5.png http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_... http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_6.jpg http://sebsauvage.net/rhaa/images/201303/rha_20130317_kriss_... I love it.

asb 14y ago

I suspect a decent percentage of those 25 million CNN RSS subscribers actually don't use RSS. They opened google reader one day, added CNN, and later forgot about it.

gz5 14y ago

Not sure CNN is good proxy for RSS - many folks not using RSS for CNN because they get CNN via Twitter etc or visit it x times/day anyways? Seems like much (half?) of RSS market is folks monitoring streams of websites that they don't visit regularly or monitor w/ non-RSS solutions?

voidfiles 14y ago
Out of all the feeds that I checked CNN had the largest number of Google Reader subscribers. If you are right, I would argue that CNN market penetration is lower then 50% and thus the feed reader market would be bigger.
- gz5 14y ago
  Good point, I would guess it is lower than 50% of total (irrespective of active vs. not active). Not sure where you sourced your GR data but maybe the ratios of GR/uniques across a few different types of sites to find out if GR ratios varies by site type could sharpen the straw man.
  - voidfiles 14y ago
    Agreed.

speek 14y ago

For those of you who need your RSS fix: I made a little thing yesterday that sends you an email (I have it running daily via cronjob) with whatever news/information is relevant to you -> https://github.com/mheld/daily-digest https://github.com/mheld/daily-digest It's not pretty, but it sure as hell works for me (https://dl.dropboxusercontent.com/u/63726/Screen%20Shot%202013-04-18%20at%207.23.54%20AM.png https://dl.dropboxusercontent.com/u/63726/Screen%20Shot%2020...) Please feel free to fork it!

Kylekramer 14y ago

Ludicrously optimistic. Not even questioning that CNN's Google Reader stats is all active users just throws the entire calculation out of whack. I'd be shocked if the RSS market is larger than 15 million.

jostmey 14y ago

Google did not seem to think they could make money from it.

jarek 14y ago
Google did not seem to think they could make enough money from it to make it worthwhile to them∗

bdarnell 14y ago

The stats cited here include both Reader and iGoogle, so you can't really extrapolate from these numbers. CNN was disproportionately popular with iGoogle users (where it is a part of the default configuration for new US users) while Engadget was disproportionately popular with Reader users (where it is/was one of the featured suggestions for new users)

PaulHoule 14y ago

It's hard to say. Note that such a system might read messages from other protocols (email, SMS, newsgroups, facebook, twitter, whatever) and it might come across as so cool and essential that people really feel they need it. What I can tell you is that Bing and Google have serious legal, business and technical reasons why they can't offer services that have an interesting degree of personalization based on your usage patterns. (Yes, they localize and they do some simple adjustments of your search results to fool you into thinking they're smarter than they really are) (And funny, Duck Duck Go gets a huge amount of press for offering an innovation-free zero-personalization search engine while the companies trying to beat the 70% P@1 barrier Google and Bing face are stealth mode companies you've never heard of.) Anyhow, text analysis has been getting radically better in the last few years and we're really on the threshold of building something that can automatically construct an interest model for you and customize things based on preferences. The algorithms are computationally expensive so if you've got a ARPU as bad as Facebook you can't afford them, but you might be able to make a subscription service profitable before somebody can make an advertising-based service. Ultimately the RSS reader market has to give up on the "view hundreds of feeds side by side" model and give up all the excuses that it can't be done.