|Denis at eVR|
We have had static statistics for the last couple of weeks (impressions, clicks, income). No sign of any trend, either up or down. I think some webmasters are looking too hard at the numbers - maybe can't believe that they won't go down, so finding every scrap of evidence, however insignificant, that they are. Don't do it, guys. Sit back, relax and enjoy the ride!
Probably a seasonal variation thing. My main site is a cctld domain information site and traditionally, the main traffic is during the weekdays. During the weekends, traffic drops off considerably. Also the fact that today is a public holiday here in Ireland has meant that the figures have been low as well. As has been said elsewhere in this forum, some advertisers reduce their spend over the weekends. A single day's stats is too small a set to make any decisions on as any number of factors (net problems, routing, sunny weather getting prospective viewers to the beach) could affect it.
You make an excellent point. For three years, I have seen this "the sky is falling" over at the Amazon Associates board. At any point in time, some group of participants in such programs will experience results below their previous average.
From July 31 thru early Aug 4 (4+ days), with less than 200 clicks in that time, my CPC has dropped almost 50% from its level during the three previous weeks. CTR remains about the same. At no time in the previous three weeks did my CPC fall to this level for more than one day at a time.
Since many webmasters expected Adsense CPC to fall, I am wondering if others experienced the same. Or not? If anyone has experienced NO declines in CPC over the past 4 days, please post. That would be VERY useful information.
I saw up and down but no more than 2-digit variation. It should be normal.
Again, I do not think Goolge has made any change on the payout rate, at lease from our site's view. Any change could be resulted from advertisers' decisions and visitors' "taste" change.
No need to concentrate on daily variation, even weekly. This is not the trend. Focus on your "core" business, no matter content development or e-store/commerce.
no discernable trend here. Only to suggest that i dount very much that google would introduce simple "precentage" reductions but that changes could depend on a variety of factors (and possibly do already) say (just out of the air) - 1st 2nd or 3rd click for the same IP/user within a certain time from same site or network wide, sliding scale of percentages (higher the CPC, lower the percentage payout), type of ad (depending on whether adwords has low or high unmet demand for that category, or wishes to encourage/disencourage sites to provide space for them), and how much they like you site and think it is good for advertisers and adsense based on a multitude of indicies including person-review, CTR, etc etc.
No trend here, either. My effective CPM yesterday (Sunday) was quite a bit higher than normal after a slow Friday and Saturday.
This topic has already been discussed ad nauseum, so I'll just quickly repeat what has been said in other threads:
1) The lower your traffic, the more variation you're likely to see in your daily numbers.
2) You'll have good days and bad days for any number of reasons that vary from time of day or month (which will affect ad competition and bids) to seasonal factors.
3) Over the long haul, the good days and bad days will average out, and you'll be able to get a better handle on trends.
|...depend on a variety of factors (and possibly do already) say (just out of the air) - 1st 2nd or 3rd click for the same IP/user within a certain time from same site or network wide... |
I agreed with some other webmasters that it would be great if Google could "tell" the impressions (I have to emphasize that it should not be the "clicks") from publishers/webmasters to avoid the "fraudulent impressions".
I just could not imagine that Google could set up so "sophisticated" methodology to track/identify the "1st 2nd or 3rd click for the same IP/user". If those clicks would be really from "decent" visitors without "fraudulent", why discount them?
As a publisher/webmaster, I personally focus on impressions and overall revenue, not CPC & CTR. Let Google and advertisers track CTR, let advertisers decide th CPC.
I'm seeing exactly the same as you - for the last 7-8 days now, my CPC has been averaging around 50% of what it was - however, impressions and CTR remain at about the same levels
In the 2 weeks prior to this period (since I started), I NEVER had more than 1-2 days with a CPC this low (I would subsequently get a couple of days of very high CPC to take the average back up)
Yes, I know the old thing about don't look at 1 days stats - only a fool does that - but the question is, how many days do you have to wait before you say things have changed for good - at the moment, all I'm saying is I haven't seen a 'dry spell' last this long before.
I'm still being patient and waiting for the good times to come back - let's keep our fingers crossed!
|I'm seeing exactly the same as you - for the last 7-8 days now, my CPC has been averaging around 50% of what it was - impressions and CTR about the same |
You simply have no idea which ads your visitors clicked. So, if I were you, I just could not tell whether the CPC were lowered or that your visitors just simply clicked different "cheap" ads in different category/keyword (I assume that your site just like ours would have different types of ads).
Our CPC went down around the end of July/very beginning of August. Adsense was running ads for smaller companies that don't normally show up. I think the big guys used up their advertising budget for the month, because they're showing up again and our CPC went back to normal.
You both make good points.
More important than the number of days is the number of clicks (if the topic is CPC). Thus I quantified my last 4+ days as "under 200."
EFV, you imply higher volumes than that. I hope so, because then your so-far-stable results would be more significant than my drop-off.
On the topic of CPC's, some rough indication of number of clicks as well as days could help others evaluate.
Here's hoping it's just random variation.
>>As a publisher/webmaster, I personally focus on impressions and overall revenue, not CPC & CTR.
You would be short sighted if you only concentrated on Impressions and overall revenue :(
You can improve both CPC and CTR (in other words CPM) by getting better targeted ads from Google. In other words look at your page targeting.
Also look at the type of positioning you have on your page. By adjusting it you can improve CTR.
Increasing impressions is another skill altogether!
I have seen a sudden and statistcally relevant drop which lasted 3-4 days - its fixed itself now though, (I didn't panic though, I just created a load more content pages ;))
Can these sudden drops be caused by one or more advertisers pulling out?
Anyone seen this whilst monitoring whose adverts are appearing?
Seen more than normal PSAs appearing?
>>"More important than the number of days is the number of clicks (if the topic is CPC). Thus I quantified my last 4+ days as "under 200.
EFV, you imply higher volumes than that. I hope so, because then your so-far-stable results would be more significant than my drop-off. "
Getting more clicks that your cut off figure, I can concur with EuropeforVisitors that figures have been, and continue to be, stable and if anything improving.
My CPM figure is, with the exception of last Friday, continuing to push upwards
Folks - don't forget that the adwords sponsors may be adjusting their campaigns - realizing that paying $15 per click through for "widget" is not giving them ROI.
I have had wildly variable revenue in my adsense I have been paid one day something like .02 per click through on one day, and then the next been paid an average of $.60 per click through.
I would blame the advertisers (and their budgets) and not the adsense program.
Thanks. That is welcome news.
Please bear with me if I keep this thread alive, until my CPC's return to previous levels.
In the meantime, those with significant clicks (10+ per day?), and stable CPC's, please keep posting sush results.
Well - since posting my last message I checked stats for Monday, and I'm seeing VERY juicy CPC!
Of course, today may just be a wild fluke and I will plummet back down again - it will take at least 7 days of 'back to normal' CPC before I relax again :-)
I think Rearden may have it - it may be a case of ad budgets for the month being used up - today is the first Monday of a new month and the cash is rolling in again!
Anyway, here's hoping Rearden's correct!
Analyst - any improvement at your place today?
Our daily clicks are in the three digits. I can confirm there was a sharp drop in revenue on Sunday. However we had a small drop in number of clicks as well. Perhaps everyone's out getting a tan. Today seems about 20% down on last Monday.
|You would be short sighted if you only concentrated on Impressions and overall revenue :( |
I do not care your comment and we just do what we think right and follow Adsense's TOS!
|You can improve both CPC and CTR (in other words CPM) by getting better targeted ads from Google. In other words look at your page targeting. |
Our site has already been well targeted! I do not see what we could do further to "grab" those high CPC ads!?
Advertisers/Google make those ads. You could not beg your visitors to "click" them. I have no idea and doubt the effectiveness you "improve" the CTR.
|Also look at the type of positioning you have on your page. By adjusting it you can improve CTR. |
No interest to spend more time on this issue. We have our own guidence to structure our site. Adsense is just extra "gift" for us and we do not live on it.
Good for you. And since several seem to be "up," that would strongly argue against a system-wide drop in rates.
It took me a while to figure out the patttern of CPC's, on my site. (I obsessively watched the data change on my Adsense Report and did the subtraction.) But I think it's helpful.
Since I can't use real CPC's, I'll use a currency called Pfennigs (PF), not equal to one cent.
80% of my clicks pay 2 or 3 PF. Most of the rest, say 18%, pay 10 PF. Together, these average out to 5 PF. Call 5PF the "base rate" CPC. In any short time period, I can expect a CPC of 5 PF.
But it's those remaining 2% of the clicks, that pay 50PF, 100PF, or more. Just these few clicks DOUBLE the average CPC from 5PF to 10PF. (I emphasize these numbers are NOT the real numbers in cents, but ARE accurate, relative to each other.)
So In any one day, I might not get any of the "biggies" and only get my "base rate" of 5PF. Other days, when the biggies occur, the rate goes to 10PF or higher. Overall average = 10PF.
There is LESS variation than the simple daily average CPC's might suggest. 98% of the clicks are within a realtively narrow range, and they average out consistently. It's only that last 1% or 2% that cause fluctuations.
There are three statistics that AdSense provides us with:
1. Impressions (down to me and my content-rich pages, unless Google starts serving up loads of charity ads)
2. Clicks (How relevant are the ads that Google serves up)
3. Earnings (see  below)
From these, we can derive
1. Earnings per click (EPC)- how do our customers (the advertisers) perceive the value of AdSense)
2. Click Through Rate (CTR)- see  above
3. Earnings per 1000 impressions (CPM) which we can use to compare with other advertising models
As many other people have pointed out on various threads, daily stats are meaningless as far as predicting trends are concerned. Currently, I am tracking the various stats on a 7-day moving average basis. But that's only because I've been in the AdSense program since the middle of July.
A 28-day moving average would be slightly more insightful, and I'll move to that when I have enough data for this figure to be meaningful.
But hey, guys, we surely have more productive ways of spending our time than worrying over minute daily fluctuations in these various stats.
Yes, I check my stats a few times each day, and I've got a spreadsheet to monitor stuff.
But I don't get frantic if today's stats (or even the last three days stats) dip dramatically.
This is a long-term program we're in.
Well, I wouldnT' call 10+ a significant saqmpl of clicks.
Personally I'm doing around 20-50 clicks per day, and Usually only one or 2 clicks almost DOUBLE the CPC for that day.
as long as a single click has the same statistical significance as 80% of your daily clicks, each click is not statistically significant.
Since a sinlge click can give up to $20 (I once had one of those, yipee). you must have several magnitudes more income per day to be able to get a smooth average. Until then, each click really is a statistical anomaly.
Right you are. I just thought it better to set a low threshhold, to solicit the most input.
without being too specific, let's just say we have a very good number of clicks per day with which to judge. Clicks, and click-thru rates are up (and have been steadily increasing) but the CPC is down. It could be that the cheapy topics are getting the newer clicks, but overall revenue is still down.
Basically, I just wanted to know if anybody else was seeing a trend. It sounds like the answer is no.
I think I would like Google to be more up front with their revenue share. Other programs spell it out.. we pay you 60% of revenue, etc. Even if Google were to say we pay out 80% but will change that to 50% gradually over the next year, I'd still be happy. It is the "not knowing".
|I think I would like Google to be more up front with their revenue share. Other programs spell it out.. we pay you 60% of revenue, etc. Even if Google were to say we pay out 80% but will change that to 50% gradually over the next year, I'd still be happy. It is the "not knowing". |
Google may not be using a simple percentage split. For example, they could be using a tiered structure (not unlike other royalty or commission schemes) that takes various factors into account.
Just as important--perhaps more important--being secretive about the payout formula makes it harder for competitors to come along and tell publishers, "Google pays you 50%, but we'll pay you 60%." (This often happens with affiliate programs, which will cut special deals when it makes sense to do so.)
>> I think I would like Google to be more up front with their revenue share
Perhaps their ability to pay decent commissions relies on not disclosing too much. Would you prefer to get your current commissions and not know percentages.... or get half of your current earnings but know the exact percentage they are paying you? :-)
Trends we find in our own data are probably not generalizable. On the mix of pages I'm running AdSense, I see weekend dips and by-day surges, changes in ctr etc. But I've seen nothing to make me start to worry.
Taking into consideration that my site is very small, I did not think that I had enough clicks until today to compare and the result is that avg. price per click during the first 4 1/2 days of August is only 2.2% lower than the avg. price per click during July. This does not mean anything since the program is very very very wild.
So my take is that in my case there is no material change.
I have actually seen my CPC go up in the last week. It was about 32cents, and now its upto about 40cents. My CTR has gone down considerably, but thats probably because I made the ads available on every page of my site, where as before they were only on a few select pages. Of course my revenue is up though.... :D:D:D
| This 50 message thread spans 2 pages: 50 (  2 ) > > |