EPC, eCPM down ~6.5%.
It's good to remember that it's not really a "normal" week:
- President's day
- Valentine's day
- (Ski, etc) holiday week in many countries
Yesterday was my best day this month... so far.
Also, everyday has been above my baseline.
[edited by: Lame_Wolf at 2:48 pm (utc) on Feb. 18, 2009]
|It's good to remember that it's not really a "normal" week |
No criticism of you alephh however just how many weeks of a year are considered "normal" in US terminology.
I've never known so many holidays, they seem to be every other week! In the UK, apart from Easter, Xmas and New Year, we have 3 extra days...2 in May and 1 in August, that's it!
Another question has to be why would the CPC paid by AdWorders be reduced simply because it is a "holiday" time? Surely for many widget sectors it would be a sensible time to increase their advertising spend IF extra traffic is going to those sites or is this another American "norm" with either too much or too little proven analysis?
Me? I know nothing and I haven't known that for very long :-)
Yes the eCPM has gone too low today for no reason, the traffic is up everything up but the eCPM gone!
HuskyPup, I am seeing the same as you, since the reporting blip my eCPM is about 20% - 30% down.
And being though Sun - Mon - Tues are my best days it dosenít bode well for the rest of the week.
I hate it when I see these patterns; it brings out the conspiracy theorist in me!
[edited by: dawnstar at 3:46 pm (utc) on Feb. 18, 2009]
|my eCPM, has been reduced by just over 15.5% |
My eCPM for the same period is down 35%
Valentine's Day is typically the end of the shopping season for many Americans that starts around Halloween. This time due to the Depression things are even worse -- basically Americans are not shopping and therefore clicking less often.
HuskyPup's post was about EPC, not eCPM, so it might be better to focus on that topic.
I haven't noticed any significant drop or increase in average EPC over the past few days (just the usual minor ups and downs), and EPC for February to date has been pretty decent. For my site, clickthrough rate has been the big drag on earnings. People just don't seem to be clicking as readily as they used to, probably because so many of them are reluctant to translate "That sounds cool" into "That sounds cool, I want to buy it" in the current economy.
Here's a hint to y'all. The best way to understand adsense is to actively understand adwords and how advertisers work.
|Another question has to be why would the CPC paid by AdWorders be reduced simply because it is a "holiday" time? Surely for many widget sectors it would be a sensible time to increase their advertising spend IF extra traffic is going to those sites or is this another American "norm" with either too much or too little proven analysis? |
Lots of reasons. First, extra traffic isn't important. It's the "right kind" of traffic. Second, it's anticipation of less competition, so drop the bids to make sure. Third, it's anticipation about the value of the traffic.
You should be able to take it from there.
...and always remember, it's PERCEPTIONS, and not objective reality that determines behavior.
|is to actively understand adwords and how advertisers work. |
|It's the "right kind" of traffic. |
Your points are totally valid however have advertisers the actual "facts" that holiday periods do not attract spending visitors or is it a reverse psychology game played by some pretending the traffic is of less value when it may actually be the opposite?
Honestly I don't know however my EPC usually recovers very quickly from a low spell yet this is now the fifth day of much lower than 2009's average...and straight after last Saturday's debacle.
Conspiracy, no, something broken for some of us...more likely.
I feel that people underestimate the seasonal/holiday variations. Sure, some sectors are inaffected by them, but some peak heavily, most obvious example being the American tax-period, when EPC jumps from 0.1$ to 50$ per click, and then collapses back to the normal after advertisers have spent their budjet and tax deadlines have passed.
Sure, there are lot of holidays, most of them without any affect to publishers (depending on niche, traffic/advertsers/continents, etc) but I have come to notice that some of the holidays will explain/fit data variations pretty nicely - year after year.
Why not wait a few days and see if things improve? If there is a "holiday effect," it should go away--at least until the next holiday.
Husky, why do you think I mentioned that it's PERCEPTIONS that count?
|Your points are totally valid however have advertisers the actual "facts" that holiday periods do not attract spending visitors or is it a reverse psychology game played by some pretending the traffic is of less value when it may actually be the opposite? |
|For my site, clickthrough rate has been the big drag on earnings. |
Me too, the average CTR of my sites has decreased substantially from the long term average (past years) since september, however, some increase of EPC has somewhat help my revenues stay above water.
The past 2-3 days have also been pretty weak.
I'm from old "UPS Club". Very active domain top in my country, very well estabilished, long date AdSense client. Same CTR, but loss of 40% of EPC since 12/feb.
I hate to just say "same here", but same here.
I had an eCPM drop from X.XX to .XX in one day with no changes to the site.
Traffic is consistent, CTR about the same, the eCPM just went BOOM!
And usually after the 15th for some reason my eCPM climbs.
Oh yeah, and ad targeting went to heck in a handbasket as well.
It's sad when YPN is better targeted than Adsense.
But that's what I'm seeing right now.
Boing...they replace the batteries and my EPC goes back to where it was!
After Thursday being my lowest earning day of 2009, Friday's earnings were 70% higher. I don't think roller coasters are allowed to have such extremes are they? :-)
HP I was and am still seeing a drop in EPC, ECPM and overall earnings since last Sundays downtime.
It's not yet recovered, that's if it does. I'm just hoping things stabilise soon but what can I do but wait.
I don't like these huge swings since they are obviously glitches in the Stupid Pricing algo. No one will ever convince me that all of a sudden loads of advertisers in many different countries coincidentally decided to increase their bids all at the same time...this algo is flawed and I have the feeling that no one knows what to do with it therefore until they "invent" something else we're stuck with it.
I'm with you HP. It seems some troll the forums to reiterate the same old "changes in advertisers" etc. It's just too coincidental that the huge drop come after the downtime.
These "glitches" can mean a difference of thousands of dollars per month which is dramatic.
Fingers crossed this week is a better one for us all.
|I'm with you HP. It seems some troll the forums to reiterate the same old "changes in advertisers" etc. It's just too coincidental that the huge drop come after the downtime. |
But why was there a "huge drop" for some and not for others? If Google made a global change of some kind, shouldn't it be affecting publishers across the board?
|If Google made a global change of some kind, shouldn't it be affecting publishers across the board? |
The problem is that we have no idea of how Stupid Pricing is actually implemented and how Google actually applies it.
Sure they give us a brief explanation of what is supposedly more relevant for advertisers however a slight change in the algo, or whatever has been done, could mean a significant change both up or down for only a select few.
It's very noticeable that a limited number of publishers see similar changes all at the same time and to varying degrees therefore it is not a random coincidence but a specific action on Google's part.
Do they see posts here and think "Ouch, we tweaked it a bit too much, better throttle it back a bit"?
Certainly we should all expect statistical variations however some of these tweaks are a severe kick in the nuts making one wonder IF one has actually done something wrong and has been penalised only for a while later to see the metrics return to their normal patterns.
Interestingly I was re-reading 21_blue's post from 3+ years ago earlier on, it is still extremely valid, probably more so these days:
Your thinking in this case is illogical IMO. The variables involved in smart pricing and even just the basic pricing algo are probably immense and we definitely don't know all of the factors involved.
Lets say it were a bug that only effected channels that were created before February 19th 2007. It's not going to effect everybody is it, in fact only a small percentage of publishers.
Perhaps a new set of keywords, topics or something else now trigger smart pricing to a great extent. Perhaps Google decided to take a greater percentage from only UK publishers. I really have no idea but it's crazy to assume that because EVERYBODY wasn't effected, the drops in earnings, must be due to drop in advertisers or changes to their max CPC.
Their algo is utter crap. It's unstable at the best of times. A prime example is that I just made a new channel for one of my main ads and ecpm has shot up 4 times. I find that alarming. Even if it's temporary.
I'm down an average 30% on daily revenues since the 14th though site traffic remains pretty much the same. CPM has taken a big hit, CTR also less but not by that much. The ads seem to be less targeted and that may be why people aren't clicking, but WHY are the ads less appropriate to my content since the 14th? What did Google change - and would they please change it back?
|but WHY are the ads less appropriate to my content since the 14th? |
Interesting you should write that. I supply specialised construction products and normally the ads are superbly targetted however today I saw some Greek island hire car ads!
Their "Little" tweaks seem to have untested consequences at times.
Same in Spain (Financial site).
Average eCPM drop by more than 50% since Feb 14.
Same page views, similar CTR but huge drop in earnings.
The Doublecrap effect since 12th or so. Targeting screwed up. Running out of junk slots on the competitive ad filter. It is like someone took an axe to the targeting algorithm or rewrote it in VisualBasic. The Doublecrap effect seems to vary by day too.
Hey jmcc, don't be dissin VisualBasic 'tis the king of codes!
(hobbiest coder here)
Maybe the lack of advertizers in some sectors is partly responsible for the poor targeting (they are simply running out of related ads), which in turn decreases CTR.