4 ms·

That's an interesting argument but I suspect that advertisers care only about the number and quality of referrals. The specifics of the process Google uses to g

by ed 19y ago

That's an interesting argument but I suspect that advertisers care only about the number and quality of referrals. The specifics of the process Google uses to generate those referrals are abstracted away from the advertiser so they probably don't care how many users pass over their ad so long as they get n quality leads. But my comment may have been a little confusing without explanation. The total number of aggregate clicks is dependent on (a) the number of impressions and (b) the ability to convert those impressions into clicks. The number of aggregate clicks will increase so long as a, b, or a and b together, continue to increase. If either a or b are increasing, there will be growth. If both a and b are increasing, there will be accelerating growth. In this sense, we're just seeing the end of accelerating growth, not the end of growth all together. What's more, (b) will increase discretely as Google modifies its ad-selection algorithm, so it should look more like a step-wise function than anything else. It doesn't seem like a cause for alarm when the graph of (b) plateaus for a bit!

wanorris 19y ago
All excellent points. Thanks for posting this. I agree that 2 months should never be a cause for alarm for something like this. It's only that for so long they had been able to keep building significant continuous improvements in clickthrough rate that makes it a potential warning sign. At some point, clickthrough rate could actually end up plateauing on an ongoing basis as they run out of ways to keep improving the rate. Then their ad business would be solely dependent on impression growth (which I assume would still be fine) for clickthrough growth. Of course, as you point out, referral quality is also an important factor, so even if they get the same number of clicks, that might be great for advertisers if it's better traffic. Is this the end of growth in clickthrough rate? Unlikely. But it's always good to keep an eye on these things in case it turns out to develop into a trend.

by ed 19y ago

wanorris 19y ago

All excellent points. Thanks for posting this. I agree that 2 months should never be a cause for alarm for something like this. It's only that for so long they had been able to keep building significant continuous improvements in clickthrough rate that makes it a potential warning sign. At some point, clickthrough rate could actually end up plateauing on an ongoing basis as they run out of ways to keep improving the rate. Then their ad business would be solely dependent on impression growth (which I assume would still be fine) for clickthrough growth. Of course, as you point out, referral quality is also an important factor, so even if they get the same number of clicks, that might be great for advertisers if it's better traffic. Is this the end of growth in clickthrough rate? Unlikely. But it's always good to keep an eye on these things in case it turns out to develop into a trend.