Your point being that by blocking it you increased the variety of ads? Or that other similar ads showed up, that paid better? Or...? Interesting story but I am wondering what got you from A to B.

Yeah it was all in the "Long story short" bit..

This advertiser was hogging many of my pages alone in a text leaderboard, my visitors were probably tired of seeing that ad again and again, my ctr and epc were sliding, my overall earnings were in the drain, now without that that ad I am seeing the best performance ever in all metrics, amazing how the 'smart' algo gets stuck.

good idea. I do see some advertisers "hogging" space in my site. If they're there forever, repeat visitors might be inclined to ignore the ads, so filtering one or two of them might force a bit of variety. I'll try it.

Thanks, Hobbs. Makes sense. This reminds me of a post from several months ago from someone who had been seeing the same advertisers all over their site--maybe 7 or 8 in all. And to increase variety they were doing rolling blocks--block 2 or 3 for a week, then unblock them and block a few others instead. They said it helped.

on topic and very related to my site

I've run into similar situations on my art reference site where the ads are appropriate but I just KNOW it isn't what the visitor would likely to be interested in. As an example, a very pretigious gallery site offering original Picasso's vs. a print & poster site selling framed reproductions.

Many, I'm sure, have a big budget (high value clicks) but it doesn't make any difference if no one ever clicks them! After filtering those out and having the "probably most appealing" ads surface... earnings always go up.

Takes a lot of time and is really only effective for your local geographic zone.

Chapman

Hunderdown, that makes good sense if there are a lot of advertisers in that particular niche.

Especially if the top 10 or 12 are bidding almost as much! Earnings should rise especially on forums etc with lots of repeat traffic.

But in some niches the top couple might be bidding a dollar, but the next 10 could be bidding very low cents... Its hard to know but removing a bunch that includes the top ones would be a little disapointing! The overture tool will give you a good idea though for each page/site?

Actually if you knew the spread of cpc you could decide if one single add or 3 blocks would work the best on each page!

No point in 3 blocks if from 3rd ad down on block one they pay 2cents... This is why on some sites one block or a single ad can pay better.

[edited by: Genuine1 at 6:07 pm (utc) on Oct. 25, 2006]

if you knew the spread of cpc

Wow... that could be useful! ;-)

Chapman

Genuine1, yes, depends on the site and topic. Personally I wouldn't block a genuine advertiser except in the kind of situation Hobbs described, where one advertiser had completely taken over most of the space, and clicks and earnings had dropped signficantly.

Sorry to hear about your slow week Hobbs - good to see you found the problem.

Hey - some familiar names in this thread what?

Hey - some familiar names in this thread what?

"Never heard of them. That's not exactly true. We were like brothers." [Airplane II, Shattner]

Been having a slow week, earnings down by 40%

My waters tell me that AdSense has a serious problem and they haven't a clue how to resolve it at the moment. The last couple of days I've earned less than a weekend which has never happened before...ok, ok, there's a first time for everything however this is one of my trade's busiest time of the year...go figure.

Logging-in and logging-off problems, huge drops in EPC in the second 12 hour Google days, usually around 30-40% less, how accurate this is I really do not know since metric updates are very sporadic, plus unannounced maintenance updates and plenty of other questionable occurrences really is beginning to test my patience (that was polite!).

They really do need to get this thing sorted out, I think we should rename it hydrasense:-)

Umm, but Optirex, the last ten days for me have been very good, after a rocky start to the month.....

Me thinks Hobbs really has it right. I notice that the 24th to 29th of each month, clicks remain relatively the same but EPC goes down and there are a number of advertisers hogging space. Even if the ads are showing the highest paying, visitors would probably be "blind" to the regular ads. Maybe even with advertisers budgets geting depleted towards month-end, throwing a few into the filter list might just increase clicks instead and even things out.

the last ten days for me have been very good, after a rocky start to the month.....

Obviously everyone's mileage varies however I am getting very noticeable overnight trends with substantial EPC reductions.

There could be several reasons for this with the most obvious being:

1. The Middle East and European trade is relatively quiet with surplus capacity therefore they are advertising more.

2. The USA/Canadian market is buoyant and they do not find the necessity to advertise so much hence the ad bids are reduced. This I have actually experienced to a specific extent with my AdWords promotion.

One thing I do not know about AdSense is whether it is trade or public clicking on the ads...oh boy would I like to know that statistic!

The frustration is that my bricks and mortar business is absolutely pedal to the metal with substantial enquiries every day and production sold out for many months ahead however trying to seek even slightly predictable trends with AdSense(!) is obviously an impossibility.

Maybe I ought to just leave my AdSense to run without regular analysis and keep smiling:-))))

Opti-

So then the beer didn't work tonight, huh? ;-)

Chapman

So then the beer didn't work tonight, huh? ;-)

Steady 4 pints of Tetleys extra cold smoothflow tonight:-)) Worked a treat however looking forward to the Waggle Dance tomorrow:-)

Opti I would study things closer. Because my income, epc and ecpm is up from the start of this month by a fair bit over the previous 3 or 4. About 1k dollars better for this month.

Thats over 12 or so unrelated sites and it seems to be pretty evenly spread.

There must be somethig? Type of traffic? (search update somewhere)or something!

And its not regonal because I live about 16 miles from you!

All these drops/gains seem to have started at the beginning of this month? Something changed but what?

eg [webmasterworld.com...]
And others like it.

And some say the opposite. While this always happens it seems odd that there are so many for this month and that there are a bunch that say the opposite.

[edited by: Genuine1 at 1:36 am (utc) on Oct. 26, 2006]

There must be somethig? Type of traffic? (search update somewhere)or something!

SERPs are fine, traffic is "as expected" however I have had some very, very unusually low CTR days when there seem to have been "AdSense reporting issues" plus these overnight large reductions in EPC.

Quite simply consider for every 100 clicks USD 0.10 less is USD 10.00 in earnings, say this happens for 500 clicks and all of a sudden there is USD 50.00 less per day with no real explanation as to why.

Ah well, I suppose it beats me having to employ an ad agency!

Something changed but what?

Whilst there had been fluctuations something definitely happend on 17th October and, for myself, nothing has been "normal" since then.

There have just been too many "outtages" and maintenance updates for this to be a coincidence insofar as I am concerned.

Whatever changed on the 17th, give me more of it! eCPM up about 30% from the first part of the month....

Opti-

What this says to me is that you, like others in this forum, are way too intelligent do be interacting with a financial model of this sort where you have absolutely no information as to how it really functions or the ability to affect it's true course of operation... AND it drives you bloody crazy because you are not in control of your own destiny!

The wine is obviously working for me! ;-)

Chapman

but you can not know the price of related AD.

Wine for you, Chapman. AdSense Zen for me. Let go, let go....

Let go, let go....

And that I will do!

See you tomorrow! Here and there!

Shucks, one more thing... I DID get an email off to G about the site targeting "opt-out". I look forward to a response from them and/or different stats tomorrow.

Many thanks for the advice!

Chapman

What we really need (maybe it's already there?) are channels which don't just tell you which block unit (general) is generating revenue, but which block unit *ad* (specific) is making money. I'm sure Google knows this and could let us know in channels. Yes?

This would make everything in stats much more interesting. I think it would be a win-win-win situation. We can remove (or keep) the ads which are doing well; the advertisers don't waste their time if their ads on our site aren't producing; and Google then makes more money.

Can we get beyond colors, sizes, and placement? :/

Regarding ad blindness, I'd like to see some research data. Don't know if it's the same online, but tv advertisers spend millions for the exact same spot run for months. Don't they know about ad blindness? Is it really something to worry about?

p/g

Blocking sites immediately dropped my income by 10-15%. So I chose to dump my filterlist and earnings are since then back to where it was. Reducing ad exposure and staying with 1 large rectagle box and one link unit worked best for me to maximise my revenues. Any other combination always eventualy pulled my revenues down.

This morning I've been doing some serious in-depth with my figures and possibly I maybe over-reacting to what is occurring!

Comparing my historical data for October 2005, 2004 cannot be compared since I was not sites wide, I see the following:

Impressions +37.92%
Clicks +14.91%
CTR -16.56%
eCPM +8.16%
Earnings +49.28%

All in all then I do not appear to be doing so badly, it would seem to be those overnight clicks that are frustrating the hell out of me!

Certainly I won't mind if next year's October is nigh on another 50%:-)

Hobbs - Have you done any in-depth comparisons?

Potentialgeek - I can't see this happening, as this would encourage the MFA sites just to filter those advertisers that pay them less - not necessarily the ads that don't perform. If you look at the broader picture, adwords advertisers are rewarded for good sites, landing pages etc that Google deem to have a good QS by lowering the cost they pay.

On the other side of this thread - I have had a couple of fantastic days this month, and everything seems to be looking good for me. I sometimes think that if they are going to initiate a big change, they do it on a sample audience first - see what effect it has etc before they introduce it to everyone. It could perhaps be this scenario we are seeing - on the other hand it could just be the normal fluctuations we all get. Hope things start to pick up for you guys soon.

Nice to see a thread with all the old names in it! Smooth-flow Tetley - mmmm- rather have the Speckled Hen!

Hobbs - Have you done any in-depth comparisons?

Sorry Opti, I'm still out of office
But to answer your question, no need to do any kind of analysis when you are seeing half your normal earnings as well as very low ctr and a dead eCPM, a quick block of one site as if set all metrics on fire almost instantaneous and to date the slump is a distant memory.

What is curious is Google's claim that their Algo automatically selects and places advertisers most likely to earn us more, and repeatedly it is proven to be either flawed, or easily manipulated by savvy advertisers.

I was hoping to stay out of work and enjoy the beach with no eCPM or CTR to worry about, but here I am!