Forum Moderators: martinibuster
One thing I have noticed is that lately, the morning (I am in the Eastern US time zone) is very good. CTR and eCPM usually at the highest levels of the day. By 2-3 pm, however, things really begin to slow down and continue to do so until late at night.
This is somewhat of an annoyance because I always have high hopes in the morning, only to see them dashed by day's end.
What really astounds me is how my CTR can be 4-5% with eCPM at $6-10 at 9:00 am, but by 10 pm CTR is 2.5% and eCPM is $2-3.
Are people more inclined to click on ads in the morning? Are the stats delayed, so I am getting a click dump in the AM? Are daily ad budgets depleted by 6-8 pm and this causes the .03 to .05 clicks?
Anybody noticing this pattern or am I just one of the lucky ones? ;-)
On a somewhat related note, I urge everyone to read this thread in the AdWords section:
[webmasterworld.com...]
In it, you'll discover that - according to AWA (on page 3, I believe) - Google employees are actually encouraged to have their own AdWords accounts, in direct competition with AdWords customers.
Some people on the thread are incredulous and find it a huge conflict of interest. My question - still unanswered - is whether Google employees are encouraged to have AdSense accounts. My thinking is that they are, based on the AdWords revelations.
ASA, care to respond?
I just assume it's the result of Google not updating all the data at the same time.
Here's a sample of what I'm talking about, from yesterday. Just after 4 pm, traffic was roughly half of the full day total. Clicks were 2/3 of the full day total, but earnings were 3/4 of the full day total.
And it's doing the same thing again today. Basically, if CTR and eCPM remained even CLOSE to what they are in the AM, I would more than double what I'm currently earning.
I've had my doubts and suspicions about the Google algo and the veracity of results. I think there is a certain part of that algo that automatically checks earnings and keeps them in a range.
Also could have something to do with sitemaps. I have been adding pages at a steady pace, but have not updated my sitemap.
Anyhow, as usual, I think I'm getting shorted and I'll probably keep thinking that until Google becomes more transparent about some of their operations (which is likely to be never).
Well, here we go again. I guess I should curb my enthusiasm until later in the evening.
My typical users are working US folks so I can't say that they all visited and clicked the best ads while I am sleeping :o). And the advertisers I see are mostly US based advertisers so I dunno if they are fidgeting with their budgets at 3 am EST.
So I just attribute it to differences in updating, and on days when I can't explain the metrics, just pure good or bad luck.
I see exactly the same as yourself however I have put this down to the "smart pricing", or whatever it is called, in that it does not pay me as much for an ad clicked in say Indonesia and China.
My Indian and Middle East clicks seem to be relatively average with Europe and the USA paying the most.
How right or wrong I am I have no idea however my stats have shown exactly the same forever and a day like this which is a least a consistency if nothing else!
Most of my clicks are shopping clicks, not info.
I check my stats about 12-15 times a day.
Therein lies the problem. Not in the frequent checking - it's your account, you can check as much as you want - but in the expectation that generally accompanies frequent checking.
Google makes no promises to keep stats up to date, to update all the columns at the same time, or even to credit your account with all clicks and earnings in the same week that you earned them.
It could be related to higher ad rates in Europe, although the midnight-to-eight shift is also a high period for traffic all across Asia as well, so if anything I would have thought income might be lower. I don't think this is the answer, but I guess it's testable if you want to go to the trouble of setting up channels for geotargeting.
This is a real pattern that I see every single day.
Oh, don't say that, don't say that! If, one day things start up slow we'll have threads saying it's slow today.
OTOH, we already have those threads so please ignore this post :)
When I check my stats in the morning I find my eCPM is 4 times the amount is is at the end of the day.
Also sometimes I find more clicks than impressions on channels in the morning, but never more clicks than impressions in the evening.
My answer is that clicks are cheaper in California ;-( which really doesn't make any sense.
Really, it's probably a stats updating thing more than anyone else, since more than a few posters see the same thing.
However, I also think that Google also adjusts the ads it serves to particular sites through the course of a day, in part based upon the site's performance earlier in the day.
You would think when you hit $180 by noon PST you're in for a $300 day right?
Oh no, it starts dwindling slowly into oblivion and you're lucky if you end the day with $250.
We are stats freaks to some extent, and have been analysing the time-of-day trend for a while now.
Our earnings half-way point in the day is, on average, around 8:30am PST (4:30pm in proper money). This is because the bulk of our income comes from UK business visitors, though US and Australian visitors also make a contribution.
The half way point can change in response to a few factors:
* there is a fluctuation day by day, as you would expect with natural statistical variation and stop-start Google reporting
* changes in exchange rates can shift the average half way point forwards or backwards, as the relative values of the US and UK markets shift
* weekend patterns are different. Eg: it is very quiet over Saturday night/Sunday morning, but picks up when Australia business time kicks in later on Sunday
* Monday mornings in the UK have lower activity than most other days. On the other hand, Friday mornings are much higher.
* At lunchtime in the UK, things slow down a fair bit.
Another observation is that page views increase throughout the day, but click through rates decline. We suspect that some US visitors look at the UK spelling and punctuation on our site, think they are "errors", and so exit the site. This view is supported by the occasional emails we get pointing out the poor quality of some aspects of our site (when you get into the detail, "poor quality" is simply the difference in UK/US English).
We suspect that some US visitors look at the UK spelling and punctuation on our site, think they are "errors", and so exit the site. This view is supported by the occasional emails we get pointing out the poor quality of some aspects of our site (when you get into the detail, "poor quality" is simply the difference in UK/US English).
You're jesting?
Americans writing to you about your standard of English and spellings? Is it that bad? :-))
I've never had that one I must admit but I did have someone who wrote complaining about the web site colours!
When I looked at her list, there were actually 2 typos in the small print (a footnote copyright statement that we hadn't reviewed when publishing the page). The other 50 were spelling or punctuation differences - mostly S vs Z, -OR vs -OUR, or putting full stops inside or outside quotation marks.
I wrote back explaining the difference between UK and US English, and the correspondent replied with surprise. She hadn't realised (sorry, realized) that there was a difference.
We have thought about using cloaking techniques to present UK English in the UK and US English in the US. However, we think we might run foul of the search engines using such techniques, so have avoided them.
We have thought about using cloaking techniques to present UK English in the UK and US English in the US. However, we think we might run foul of the search engines using such techniques, so have avoided them.
That's not what I would consider cloaking at all -
That's localisation or 'localization' for us Americans ;)
Cloaking is when you show the search engine something entirely different than you show the visitor and I don't think localisation for spelling adjustments would be considered deceptive at all.
For what it's worth, most Americans can't spell their own name let alone figure out you have typos as our educational system is in a serious state of decline.
I weep for the future.
Respected member though you are, what we are concerned about is what the search engines define as cloaking. Or, rather, what they might define as cloaking in the future.
If there is a risk-free way of presenting different content to US and UK users, I'd be interested in it (NB: we have a lot of content, but are technically in the dark ages, using only html pages and not these new-fangled cgi scripts or php or suchlike - perhaps I should have chosen the member name of "luddite"? :-)).
I should emphasise the expression "risk free". Because our focus is on content, we will only use future-proof techniques. With anything that could be construed as cloaking, we would need to be sure that the likes of Google won't at some point in the future decide to classify the technique as worthy of penalisation (or penalization or pea null I say shun).
Clicked submit too early
Back on topic, you need to take into account that some advertisers accounts might actually run low towards the evening, lowering the price per click.
Interesting idea. Though, from an SEO point of view it might still have a negative impact - diluting the impact of incoming links (splitting them between similar pages). And I'd still be concerned about penalisation due to duplication.
>Back on topic, you need to take into account that
>some advertisers accounts might actually run low
>towards the evening, lowering the price per click
Good point. Though our stats show a decline in CTR, ostensibly unaffected by CPC, it might still be a corollary of the trend you describe: as advertisers run low, there is less choice, and perhaps the 'better ads' become less available. Hence a lower CTR.
what we are concerned about is what the search engines define as cloaking
FWIW, I don't think color vs colour is what the search engines call cloaking. Traditionally cloaking is one keyword jammed page fed to a search engine designed to raise the page in the SERPS compared to what is shown to the visitors of your site. The real issue is would you want your serps to be based on "color" or "colour" for your market and show the search engine that language version of the page.
I have content filters in my site and knock out certain words in certain context, I don't really consider it cloaking at all. Some of it is "family friendly" filtering, other words are filtered out after the fact to stop Google from running PSAs based on the current set of blocking words.
It's an ever evolving landscape and I find it best to keep all my content and display only what currently meets the guidelines of the Google mediabot.
Is it cloaking?
I don't think so as I show the search engine and visitor the same filtered page.
This is a real pattern that I see every single day. I fully expect my CPM to drop by about 25% between now (9 AM Pacific time) and close of day.
FWIW, I was almost dead on with my prediction. My CPM declined 27% through the last 2/3 of the day.
Today my CPM has started out even higher than yesterday; I love seeing the number, but it'll almost certainly decline by 40% by the end of the day, back to the regular level. I noticed that the CTR on both my channels is slightly higher than usual, but the main difference is the much-higher-than average CPC.
