3 ms·

The whole internet works on free-because-of-advertising, aside from Hacker News I guess which is just an advertisement for Ycombinator. Sure we want to pull th

by jbooth 16y ago

The whole internet works on free-because-of-advertising, aside from Hacker News I guess which is just an advertisement for Ycombinator. Sure we want to pull the rug out from under that one? What's so bad about content creators getting maximum value from the eyeballs they attract?

jacquesm 16y ago
You can advertise stuff without having to know everything about your visitors, standard demographics kept newspapers supplied with ink and papers for well over a century.
- jbooth 16y ago
  Well, turning "ip:25.43.251.5, user-agent:firefox" into "male, 18-19" actually requires a fair bit of that tracking stuff. Also, classifieds are what kept newspapers supplied with ink and papers. Branding-targeted advertising was a side revenue, and mass conversion-targeted advertising as we know it now didn't really exist yet (although you can call classifieds a small-scale conversion targetted ad, I suppose).
  - jacquesm 16y ago
    Indeed, classifieds are simply ads targeted at a local audience. As for the male 18-19 bit, that is something that a site would deduce from a typical sample or survey, not from invasive tracking. btw, how did you get my IP address? You got the age wrong ;)
    - jbooth 16y ago
      Well, RE: samples or surveys, users don't fill them out very often, which is why ad networks and information providers drop a cookie whenever they manage to find that out about someone for later recall. If you're just some blog, no way people are filling that out on your site.. so you get a deal with a tracking ad company that combines your traffic with data to make it more valuable than just run-of-the-network generic traffic.
      - jacquesm 16y ago
        The samples and surveys I'm referring to are paid for by the media company, typically conducted by third parties with some credibility in the field. Plenty of people are on a 'panel' of this kind, they periodically fill out surveys about the brands they consume, the tv programmes they watch and sites they visit. And yes, that's geared towards the larger content producers, it would take some doing to get similar metrics for smaller companies and blogs, but I can imagine that you could create a series of 'verticals' that would form the aggregate of a whole series of blogs and blogs could then be categorized as being part of a vertical to be included in the surveys in a statistically relevant manner. So it's not a survey that suddenly pops up on the site (though that's been done, the quality of those is typically not as good).
- wonderzombie 16y ago
  Newspaper ads were way, way more profitable, though. Newspapers were the biggest game in town (everybody read newspapers), space was limited (you need to print news, too), paper was limited (there are N pages in a newspaper), and so on. In other words, online advertising is sufficiently different in a bunch of fundamental ways such that I'm skeptical of this particular analogy. That's just my gut, though— I have no data. It would be worth seeing some analysis of how well demographic vs. targeted (?) ads worked. Then again, I'll bet you many folks who buy and display ads have already done the analysis, and that's precisely why this advertising exists.

by jbooth 16y ago

jacquesm 16y ago

You can advertise stuff without having to know everything about your visitors, standard demographics kept newspapers supplied with ink and papers for well over a century.

jbooth 16y ago
Well, turning "ip:25.43.251.5, user-agent:firefox" into "male, 18-19" actually requires a fair bit of that tracking stuff. Also, classifieds are what kept newspapers supplied with ink and papers. Branding-targeted advertising was a side revenue, and mass conversion-targeted advertising as we know it now didn't really exist yet (although you can call classifieds a small-scale conversion targetted ad, I suppose).
- jacquesm 16y ago
  Indeed, classifieds are simply ads targeted at a local audience. As for the male 18-19 bit, that is something that a site would deduce from a typical sample or survey, not from invasive tracking. btw, how did you get my IP address? You got the age wrong ;)
  - jbooth 16y ago
    Well, RE: samples or surveys, users don't fill them out very often, which is why ad networks and information providers drop a cookie whenever they manage to find that out about someone for later recall. If you're just some blog, no way people are filling that out on your site.. so you get a deal with a tracking ad company that combines your traffic with data to make it more valuable than just run-of-the-network generic traffic.
    - jacquesm 16y ago
      The samples and surveys I'm referring to are paid for by the media company, typically conducted by third parties with some credibility in the field. Plenty of people are on a 'panel' of this kind, they periodically fill out surveys about the brands they consume, the tv programmes they watch and sites they visit. And yes, that's geared towards the larger content producers, it would take some doing to get similar metrics for smaller companies and blogs, but I can imagine that you could create a series of 'verticals' that would form the aggregate of a whole series of blogs and blogs could then be categorized as being part of a vertical to be included in the surveys in a statistically relevant manner. So it's not a survey that suddenly pops up on the site (though that's been done, the quality of those is typically not as good).
wonderzombie 16y ago
Newspaper ads were way, way more profitable, though. Newspapers were the biggest game in town (everybody read newspapers), space was limited (you need to print news, too), paper was limited (there are N pages in a newspaper), and so on. In other words, online advertising is sufficiently different in a bunch of fundamental ways such that I'm skeptical of this particular analogy. That's just my gut, though— I have no data. It would be worth seeing some analysis of how well demographic vs. targeted (?) ads worked. Then again, I'll bet you many folks who buy and display ads have already done the analysis, and that's precisely why this advertising exists.