Anyone else notice the same?

Precisely as described.

I assume that the lower values early in the day reflect Far East/Middle East/Russian Fed smart pricing, during the day European/USA values and then overnight (for me) lower value ad clicks as the day heads towards the Pacific/Japan etc.

Then again, I'm most probably wrong!

Just thought yesterday to start exactly this discussion, the development of eCPM during the day.

Was in the past much more equal distributed during the day.

I agree completely. The downside is, you think you're going to have a good day but as the evening progresses the eCPM keeps shrinking. Just have to get used to it i guess :o/

Since you cannot know when your clicks are updated, you cannot know when in the day pays more or less. Are the clicks and money you see as of 9am a result of people clicking 3hrs. before? 10 hours? a day?

I don't understand how the same people can talk about how the same people who complain about the unpredictable updating of stats can turn around and take the messed up numbers as valid or indicative of anything.

You don't know when the numbers are from. Could someone explain why people who have been in the program for years, don't seem to understand the fundamentals?

Since you cannot know when your clicks are updated, you cannot know when in the day pays more or less. Are the clicks and money you see as of 9am a result of people clicking 3hrs. before? 10 hours? a day?

True. Clicks on the stats are not in real time, still about the OP, yes, the same trend here.

perhaps off topic, but what I don't understand is the VERY LOW amount of clicks I get on Adsense compared to any banner I might put on my sites. Perhaps a click is not a click unless you stay on the landing page for a while.

It depends where are you located ;)
For me (europe, the eCPM is low at start (daytime in Europe, Asia, Africa) and is rising when the light comes to US ;)

Coachm - we are discussing a trend that is the same every day. The time of day does not really matter, nor does click recording time. It is the beautiful bell curve that is interesting... ;-)

Coachm - we are discussing a trend that is the same every day. The time of day does not really matter, nor does click recording time. It is the beautiful bell curve that is interesting... ;-)

If the time of day doesn't really matter, than why are you guys talking about the time of day?

Actually it's not interesting. The pattern has been noted almost since adsense began. It's old. And certainly the trend is NOT the same everyday, for everyone.

You can conclude nothing from the pattern. For example, perhaps the higher priced clicks are held to be verified until the next day. So, they are added in to your stats only after midnight, or 3 am. or whatever.

That's a plausible explanation. But what's the point?

My EPC starts to fall as the weak US Dollar/economy wakes up (before the year 2008 it was completely different story, as US actually could compete with Europe back then).

If the time of day doesn't really matter, than why are you guys talking about the time of day?

Time of day is actually just numbers of the graph. It is the actual time that is not important, but the trend of the span of time. Sorry if I was confusing.

That's a plausible explanation. But what's the point?

I have been starting to think about times of day that I drive traffic to my site (ie - sending out a newsletter). In the past I would send it out when I thought it had the best chance for a good open rate. But this trend got me thinking maybe I should be driving this newsletter traffic to my site during my peak Adsense eCPM times of the day? An interesting experiment perhaps. Particularly when you are talking about 100,000 newsletter readers hitting your site for a few hour span. That could net me a few hundred bucks more based on the time of day that more advertisers are competing for my ad slots.

Same could be said other traffic driving techniques used by folks on a daily basis I suppose.

But I can completely understand some folks not finding interest in such trends. It is borderline nerdy... ;-)

OK. A very intelligent sensible response.

Time of day is actually just numbers of the graph. It is the actual time that is not important, but the trend of the span of time. Sorry if I was confusing.

No, time of day is time of day. Anyway, moving on. I still suspect you may not understand that you have NO idea what clicks are being counted when, so you can't identify what visitors are doing with respect to time, what advertisers are doing with respect to time, and therefore you can't draw any conclusions about what to do.

I have been starting to think about times of day that I drive traffic to my site (ie - sending out a newsletter). In the past I would send it out when I thought it had the best chance for a good open rate. But this trend got me thinking maybe I should be driving this newsletter traffic to my site during my peak Adsense eCPM times of the day?

I understand. Check out the experts re: when they recommend.

An interesting experiment perhaps. Particularly when you are talking about 100,000 newsletter readers hitting your site for a few hour span. That could net me a few hundred bucks more based on the time of day that more advertisers ae competing for my ad slots.

It's an empirical question. Test it, then. But be aware that as a result of your being totally in the dark (being kept that way by google), you could just as easily lose money as gain. Or one time you could gain, and another lose.

But I can completely understand some folks not finding interest in such trends. It is borderline nerdy... ;-)

It's not nerdy. It's full of logical faults, wrong assumptions about the numbers, and wrong conclusions about what the reports you are given mean.

You cannot draw ANY conclusions about time of day based on the google reports. PERIOD.

If you want to do blind trial and error to test, it's fine with me. I'm betting with 100,000 subs, you can find better ways to make serious money. I know I could. I made 1,700 bucks extra in March from my 8,000 person list, and a single newsletter issue.

(edit. I can't find the error in the formatting.

I wonder if using the integrated Analytics would offer more insight into the black box known as G? We don't use Analytics due to our paranoia, but perhaps others on here might be able to share their experiences? Does Analytica provide any time data for clicks and eCPM?

I just noticed another interesting trend. When looking at my traffic graph based on hour of the day, the curve seems to match up with the eCPM curve I see throughout the day with Adsense.

It's not nerdy. It's full of logical faults, wrong assumptions about the numbers, and wrong conclusions about what the reports you are given mean.

You cannot draw ANY conclusions about time of day based on the google reports. PERIOD.

Someone forgot their morning coffee. I forgive you.

Statistically speaking the probability that Google's reporting is a completely random system is, at the very least, highly improbable. If one were to assume that Google updates their reporting data on a scheduled timeframe (as most automated systems would do - a probable assumption), then you could take away something from the data as it relates to trends.

Yes, we all know the exact time of day cannot be pin pointed per click, but timeframes can be used as a measuring stick for looking at trends.

Assuming their system is completely random is... how can I see this politely... it would be a logical fault... ;-)