5 ms·

If an ad network wants to put restrictions on the volume of ad space in comparison with content on the page, they can do so, but requiring exclusivity isn't the

by ABCLAW 8y ago

If an ad network wants to put restrictions on the volume of ad space in comparison with content on the page, they can do so, but requiring exclusivity isn't the same thing as that.

AnthonyMouse 8y ago
Measuring the "content on the page" is hopelessly ambiguous and easy to scam. Even trying to price just based on how many other ad networks there are is problematic. For each additional ad network the value of each ad goes down. Even adding a second one may cut the value more than in half compared to exclusivity, because if the user sees the other network's ads first and they don't appear relevant they may not even look at yours, and a user who sees a larger number of ads may not look closely at any of them. If the price offered for any non-exclusive placement is more of a reduction over the price offered for exclusivity than the amount gained from the other ad network(s), how is that any different in practice than requiring exclusivity?
- ABCLAW 8y ago
  >Measuring the "content on the page" is hopelessly ambiguous and easy to scam. Is it? It seems pretty simple to audit your 10,000 largest ad-spend domains and check to see if they're actually just ad fraud. You'd need maybe 300 person-hours to do so, at an overhead-in cost of, let's high-ball it, 100$/hr, which puts you at a cost of $30,000. If you mean the process is impossible to automate, then don't try to automate it. You can catch 99% of fraud with very low-effort manual review in most industries. That Google doesn't do it doesn't speak to the inability to get it done. It speaks to their unwillingness. In short: exclusivity is there to choke out other ad networks. >If the price offered for any non-exclusive placement is more of a reduction over the price offered for exclusivity than the amount gained from the other ad network(s), how is that any different in practice than requiring exclusivity? The comparison you're making in the last paragraph is a common mistake in examining competition scenarios. The comparison isn't between pricing between the market leader and the incumbents following the anti-competitive behavior. It's the comparison between the prices we would have had no anti-competitive behavior existed in the first place.
  - AnthonyMouse 8y ago
    > It seems pretty simple to audit your 10,000 largest ad-spend domains and check to see if they're actually just ad fraud. It's not just a matter of fraud. The website could be CNN, but if they're filling their pages with a thousand different ads, none of the individual ads is going to be worth paying for, even if the page also contains legitimate content. And the thing that's impossible to measure is content volume. If you have one morsel on the page that drives traffic to it, you can easily add arbitrarily more to the bottom that is arguably "content" in order to make the ratio of ads to "content" be whatever you like. At that point you're trying to measure content quality, which is inherently subjective. > The comparison you're making in the last paragraph is a common mistake in examining competition scenarios. The comparison isn't between pricing between the market leader and the incumbents following the anti-competitive behavior. It's the comparison between the prices we would have had no anti-competitive behavior existed in the first place. That's not what I'm saying. Suppose we do a test. If a page has one ad on it, 0.5% of page visitors buy the product. If it has two ads on it, 0.2% of page visitors buy each advertiser's product, because the extra clutter causes people to pay less attention to both of them. So then your ad network prices that way. If the site puts your ad on their page with no others, you pay them 5 credits. If they put your ad on their page with someone else's, you pay 2 credits. If they want to use three ad networks then you pay 1 credit and it goes down from there. Just paying proportional to the value of the ad to the advertiser. Now nobody is going to put two ad networks on the page, because they'd get 2 credits from each ad network (total of 4) instead of getting 5 credits from one. Even if it was 4 vs. 2, you might as well save yourself the transaction costs of using the second network. The reason exclusivity is worth more isn't that there is less competition so you can charge more, it's that if you clutter up the page with more ads, their value falls off faster than their volume makes up for it. Moreover, suppose that isn't the case. Showing more ads would produce more value, at least until some threshold number of ads is shown. Then the first ad network would just offer to display more ads on your page and pay you appropriately more for them. You still end up with only one ad network -- whichever one will pay you the most per unit. The only way two ad networks end up on the same page is by reaching the point that the highest paying ad network knows showing more ads won't produce more advertiser value, and then scamming them by showing more anyway.
  - jefftk 8y ago

by ABCLAW 8y ago

If an ad network wants to put restrictions on the volume of ad space in comparison with content on the page, they can do so, but requiring exclusivity isn't the same thing as that.

AnthonyMouse 8y ago

Measuring the "content on the page" is hopelessly ambiguous and easy to scam. Even trying to price just based on how many other ad networks there are is problematic. For each additional ad network the value of each ad goes down. Even adding a second one may cut the value more than in half compared to exclusivity, because if the user sees the other network's ads first and they don't appear relevant they may not even look at yours, and a user who sees a larger number of ads may not look closely at any of them. If the price offered for any non-exclusive placement is more of a reduction over the price offered for exclusivity than the amount gained from the other ad network(s), how is that any different in practice than requiring exclusivity?

ABCLAW 8y ago
>Measuring the "content on the page" is hopelessly ambiguous and easy to scam. Is it? It seems pretty simple to audit your 10,000 largest ad-spend domains and check to see if they're actually just ad fraud. You'd need maybe 300 person-hours to do so, at an overhead-in cost of, let's high-ball it, 100$/hr, which puts you at a cost of $30,000. If you mean the process is impossible to automate, then don't try to automate it. You can catch 99% of fraud with very low-effort manual review in most industries. That Google doesn't do it doesn't speak to the inability to get it done. It speaks to their unwillingness. In short: exclusivity is there to choke out other ad networks. >If the price offered for any non-exclusive placement is more of a reduction over the price offered for exclusivity than the amount gained from the other ad network(s), how is that any different in practice than requiring exclusivity? The comparison you're making in the last paragraph is a common mistake in examining competition scenarios. The comparison isn't between pricing between the market leader and the incumbents following the anti-competitive behavior. It's the comparison between the prices we would have had no anti-competitive behavior existed in the first place.
- AnthonyMouse 8y ago
  > It seems pretty simple to audit your 10,000 largest ad-spend domains and check to see if they're actually just ad fraud. It's not just a matter of fraud. The website could be CNN, but if they're filling their pages with a thousand different ads, none of the individual ads is going to be worth paying for, even if the page also contains legitimate content. And the thing that's impossible to measure is content volume. If you have one morsel on the page that drives traffic to it, you can easily add arbitrarily more to the bottom that is arguably "content" in order to make the ratio of ads to "content" be whatever you like. At that point you're trying to measure content quality, which is inherently subjective. > The comparison you're making in the last paragraph is a common mistake in examining competition scenarios. The comparison isn't between pricing between the market leader and the incumbents following the anti-competitive behavior. It's the comparison between the prices we would have had no anti-competitive behavior existed in the first place. That's not what I'm saying. Suppose we do a test. If a page has one ad on it, 0.5% of page visitors buy the product. If it has two ads on it, 0.2% of page visitors buy each advertiser's product, because the extra clutter causes people to pay less attention to both of them. So then your ad network prices that way. If the site puts your ad on their page with no others, you pay them 5 credits. If they put your ad on their page with someone else's, you pay 2 credits. If they want to use three ad networks then you pay 1 credit and it goes down from there. Just paying proportional to the value of the ad to the advertiser. Now nobody is going to put two ad networks on the page, because they'd get 2 credits from each ad network (total of 4) instead of getting 5 credits from one. Even if it was 4 vs. 2, you might as well save yourself the transaction costs of using the second network. The reason exclusivity is worth more isn't that there is less competition so you can charge more, it's that if you clutter up the page with more ads, their value falls off faster than their volume makes up for it. Moreover, suppose that isn't the case. Showing more ads would produce more value, at least until some threshold number of ads is shown. Then the first ad network would just offer to display more ads on your page and pay you appropriately more for them. You still end up with only one ad network -- whichever one will pay you the most per unit. The only way two ad networks end up on the same page is by reaching the point that the highest paying ad network knows showing more ads won't produce more advertiser value, and then scamming them by showing more anyway.
- jefftk 8y ago