[Scoundreller]

Ugh, one of the graphs is for YoY decline in CPC rates, but it it shows 2015 at the right and 2019 at the far left. Who does that?

And it shows it by quarter, so is this a rolling window?

Then:

> As long as Google can keep growing the blue line -- growth of paid clicks faster than the red line its ad click deflation -- then it is golden.

> Every three months Google has to find faster ways of expanding the total number of paid clicks by as much as 66%.

NO! The blue line, from the 68% value, was their growth in click quantity. And the direct comparison between CPC and clicks isn’t valid: GOOG could have the less clicks but increase its CPC revenue if it targets poorly. Total revenue is what we should look at.

Other than all this, this seems to be much ado about nothing. If GOOG can better target ads, they’ll get more clicks. And when an ad pays well, publishers create content that will show that ad, driving down rates but increasing clicks.

And GOOG’s ad growth is international. Where GDP is lower; as CPC rates will go lower. And international increasingly favours CPM in my opinion. A fast food chain or a brand of soap doesn’t need clickthroughs.

[dhimes]

And when an ad pays well, publishers create content that will show that ad

{rant: This- this right here- sums up my frustration with the evolution of the web. Getting ready to launch a new site, I looked around a bit at "SEO." The push was exactly backwards. It was, "given that you have some keywords you want to "optimize" for, here's the web content you need."

That's exactly backwards, and leads to people producing junk just to get clicks because certain keywords "perform well."

What we need is, "write something authentic, then find the keywords that will lead people to see it." Sure, make sure you put the keywords in the right places like headings and such, but the focus should be on writing something authentic, not noise. }

[Scoundreller]

Oh, I accept all that.

When I first started writing, I discovered what was popular versus not. Authentic content that nobody read didn’t seem like a good use of time.

Writing about how shitty my current credit card was and which I found to be better was very profitable. Reviewing a keyboard was not.

But having said that, some of the most valuable content was useful, but flew under the radar of other writers. E.g. how to navigate a particular government office’s procedures.

Google did a lot to help your case when they targeted ads based on the user and not the content. IE: showing you finance ads on non-finance websites. That really helped out publishers in otherwise not-too-profitable topics.

[floatingatoll]

Newest-left graphs would make sense in analysis domains that consider “the most recent time interval” exponentially more important than “the least recent time interval”, and would display the most valuable data (most recent) correctly on mobile screens of any width leaving truncation to cut off ancient histories.

[0xcde4c3db]

They're also used to make age the X axis without requiring the time period to be specified somewhere else.

[vgeek]

Growth in Asian/South American countries with less advertisers and lower CPCs.

There may also be disproportionate growth in branded CPCs, which have a significantly lower CPC (but higher CPM due to high CTRs) than commodity terms.

[Scoundreller]

Ya, dunno if they just calculated an effective CPC, or excluded non-CPC traffic in the numbers.

But again, from a shareholder point of view, nobody cares how any proportion changes as long as profit goes up, unless it’s at the expense of future revenue or at increased cost.