| 5:59 pm on Sep 15, 2004 (gmt 0)|
I don't calculate percentages and movements, I look at flat figures.
Revenue went sharply down when I implemented a second ad unit. I tested it for a week and stopped it abruptly.
I may try it again on another one of my sites, but for my highest traffic site, it was a flop.
| 6:06 pm on Sep 15, 2004 (gmt 0)|
I am dont want test multiple ads for now, can be paranoia but everytime that makes big changes in CTR/PV the EPC drop alot
| 6:30 pm on Sep 15, 2004 (gmt 0)|
My revenue went down a little.
Interestingly, I put two ad units on every page instead of one, but my impressions only went up 50% instead of doubling. I think this is because I'm not in a very competitive sector so there are not a lot of adwords, so I'm probably showing more PSAs. Doubling my ad units might have over-supplied my keyword market.
| 6:38 pm on Sep 15, 2004 (gmt 0)|
I believe but not so sure that page impression is counted one either for single or multiple ad units.
Does anyone have some insiders on this?
| 7:06 pm on Sep 15, 2004 (gmt 0)|
It seems the general concensus since September 1st and the implementation of multiple AD units that EPC and CPM have dropped considerably.
I can definitely say this holds true for my sites.
With Google's recent ipo and their first quarter reporting coming soon as a public company, one has to wonder if there is a correlation.
The long term test for Google will be balancing the payout to web publishers while maintaining an acceptable profit margin for shareholders.
just a thought.
| 7:21 pm on Sep 15, 2004 (gmt 0)|
Well after a couple weeks of trying multiple ads my revenue is at an all time low. Impressions are up naturally, CTR is a little higher but my CPM is an all time low.
I'm frankly at a loss as to whether to return to single ad blocks, I noticed the CPM drop prior to adding multiple ad blocks so don't see any advantage in going back.
Whats going on?
Thankfully my affiliate income is rising!
| 7:40 pm on Sep 15, 2004 (gmt 0)|
I think just adding more of the same type of ad unit is a mistake. You'll just bring in more low-cpc ads or psa's.
I went from 1-wide sky with 5 links to small square (1 link), large rec (4 links), and leaderboard (4 links). So I tripled my ad units while less than doubling the number of links.
Hopefully, by featuring the single link square prominately I will be able to increase CTR on the highest paid ads and raise the overall average cpc.
| 7:44 pm on Sep 15, 2004 (gmt 0)|
I began a thread on this subject a week ago when it was looking obvious that this new offering from google was not the holy grail. A number of people gave input along the lines of "every time a webmaster has a dip in earnings, they blame google. perhaps it's seasonal, or your advertisers, or your site".
All of which could have been true. But, at this point, I think the causation for this epc dip lies squarely with google. Perhaps smart pricing has been altered?
Websearch was supposed to help publishers and people jumped on it. It turned out to be a bust for most. Multiple ad panels were made available and the consensus is largely that it, too, is a bust. But each time, webmasters expended considerable effort in implementing one of google's new "angles". To what end?
I keep thinking of a mouse on a treadmill. Turn the treadmill speed up and the mouse runs faster, but the mouse doesn't really get anywhere.
How many people here have noticed that almost as soon as they start to register more clicks, they likewise see a dip in epc? I'll bet many have.
| 7:54 pm on Sep 15, 2004 (gmt 0)|
However, I'm not bashing adsense. It's still pretty good. Just has lots of room for improvement
| 8:00 pm on Sep 15, 2004 (gmt 0)|
Let's try to explain the EPC decrease beyond the vague concept of "smart pricing" manipulation:
The multiple ad panels were introduced mainly to allow more advertisers into the content network pool.
However, instead of competing for the 4-5 top spots with the highest click, they are now able to go for the number 6-15 spots without the high bids required for their ads to be shown in the first panel. Advertisers are supposed to be happy: their ads are shown in the content network, at a lower price.
If you are a publisher showing multiple ad panels, your clicks may go up. But if most of those clicks are coming from the lower-priced ad panels, your EPC will naturally fall.
So even without introducing smart pricing into the equation, the end result is usually a fall in EPC. The question of course for the publisher is: how is the expected fall in EPC affecting your revenues?
| 8:09 pm on Sep 15, 2004 (gmt 0)|
I want to make a couple of points;
Figment88 I think has hit the nail on the head.
|I went from 1-wide sky with 5 links to small square (1 link), large rec (4 links), and leaderboard (4 links). So I tripled my ad units while less than doubling the number of links. |
This three ad unit possibility is probably best used NOT by getting what was a 5 ad unit and stuffing 3x5=15 ads in the page with ever decreasing epc, but by getting the first ad unit early on in the page text as a single or double ad unit(since it should get a better epc), followed by other ads later on in less prominent positions. (I have read somewhere in this forum that the highest epc ads are first in the page text.)
For your interest, I had a fairly regular page impressions of 2ythousand, put three units on the most popular pages and now have 3.5ythousand - and there can really be no other explanation other than 3xads=3xpage impressions. - perhaps they should be called ad impressions, not page impressions...
| 8:47 pm on Sep 15, 2004 (gmt 0)|
use multiple ads can stimulate advisers pay less per click, this is a bad thing...
| 9:04 pm on Sep 15, 2004 (gmt 0)|
With a couple days of data under our belt running a leaderboard at the top and bottom of each content page:
# clicks per day has increased by 72%
Revenue per day has increased 40%
And happily, not one of our outspoken users have complained about the additional ad units.
Summary: users are clicking on the ads, they are not annoyed by the ads, and revenue has spiked.
I could care less about epc or the (misleading) CPM stat displayed in the Adsense report if we can squeeze additional revenue out of the same traffic without annoying our users.
| 10:19 pm on Sep 15, 2004 (gmt 0)|
I think that the number of impressions you do may have some relevance to this discussion... for instance, it would take AdSense longer on a less trafficked site to find which ads are more profitable and then display the appropriate rotation of ads.
I have found this to be true for my websites running adsense... well, true in the sense that it appears to be such.
| 10:30 pm on Sep 15, 2004 (gmt 0)|
Had a large rectange in the middle and left and right sky for six days. Kept the rectangle and removed the skys and revenue shot up -- significantly higher than when all three ads were running.
| 10:33 pm on Sep 15, 2004 (gmt 0)|
I tried two identical ad blocks at the top and bottom of my pages, and the overall revenue went up a little bit and then started to go down. I wasn't sure why, but I'd decided that the experiment was a failure, so I pulled off the second ad set, so that there's just one at the top.
My total revenue is still way down, so I suspect that's because of the massive amounts of ad space that Google added in the last couple of weeks. With triple the inventory available, CPC rates had to have gone down.
I suspect the prices won't recover until this additional inventory is all used up, and the supply synchs up with the demand again.
I think I'm going to put the second set of ads down at the bottom, and just wait a few months to see if things recover.
| 3:40 am on Sep 16, 2004 (gmt 0)|
An auction based network is likely to see some price drop in CPC when it doubles it's inventory overnight/week.
Supply and Demand. I would wait a bit and let the market re-adjust. Anybody *NOT* load the second/third unit and *still* see revenue go down?
| 7:04 am on Sep 16, 2004 (gmt 0)|
|Anybody *NOT* load the second/third unit and *still* see revenue go down? |
Yep, but it's always a roller coaster ride.
| 7:35 am on Sep 16, 2004 (gmt 0)|
I think the point about extra inventory available is very valid. I'm going to hang on with multiple ads for a while and see if their is an increase in revenue when inventory starts to go.
| 8:17 am on Sep 16, 2004 (gmt 0)|
Interesting. We haven't implemented any additional units (too much work going on on other product launches).
Revenue is down to 75% of levels before the launch, even though impressions are rising very nicely.
It looks like this is a supply and demand effect, and we'll have to implement additional units to drive revenue back up.
| 11:01 am on Sep 16, 2004 (gmt 0)|
This is my first post in WebmasterWorld, so I'll mention that I was a newspaper publisher in a previous life ;-). Web publishing is a little different, as in the old days eveybody paid the same price for the same SIZE ad, i.e. larger ads cost more, so it's strange to me to think that I get paid more or less depending on the whims of the readers.
My site is evolving, so I can't say much about my revenues, except that when I started adding pages with multiple ads, pageviews and clicks went up, revenue down, equalling less $$$ per click. It's the trouble with anything done auction-style. Eventually you get to the bottom and I think Google will come to this realization when publishers jump ship.
I am hoping that Google does not become another ebay. I have recently cut back my ebay activity from Powerseller to ZERO. I decided that paying $150 a month in fees was absurd when I could spend much less developing my own site and maybe making money with it. But if Adsense continues on its current path, without significant competition, it will also become the ebay of the ad world. In other words, the only people REALLY making money will be Google.
Like ebay, google can call all the shots in their network, even so far as to charge publishers a fee to be in the loop or to cut the rates they pay (which I think they are doing right now).
Bottom line is that you need to develop your own ad base and serve them yourself if you want to survive and prosper. Google's a nice place to start, but .03 to .10 a click isn't going to cut it for most.
Like everything else, there's competition, and the weak and the lame will get weeded out.
Keep working, stay positive and focused and good luck to all.
| 2:07 pm on Sep 16, 2004 (gmt 0)|
Just to balance out what people are saying, everything adsense has done in the last 6months - 1yr has worked out badly for my sites. When multiple ad units arrived, I expected the worst based on the bad feedback here. And they do have a point about quality reduction due to being all over the place. And that gooooooooooogle thing is a short term move that is not professional. However, I can't argue with the 40% increase in overall return that multiple ad units resulted in.
| 3:31 pm on Sep 16, 2004 (gmt 0)|
We posted multiple ad units on 9 of our sites with wide ranges of epc and verticals for testing purposes.
What we found is pretty much what we expected and have moved back to one ad unit per page for all but 1 of the sites.
1. Sites with high epc before tanked through the floor and we ran this for a full 5 business day cycle, epc down 95%, raw clicks up 80%, net effect 80% revenue drop in our highest dollar revenue sites
2. Bottom of the barrel epc sites will benefit from this because if your making 3 or 5 cents per click and you have no place to go down you will increase raw revenue, so one of the sites that had bottom end epc rose in raw revenue due to increased clicks
Our test confirms to us that if you have a high dollar epc then installing extra ads will drive that epc through the ground.
But if your already at the bottom of the barrel epc wise then multi units will make you more money
On a side not we have seen inventory levels network wide drop greatly as inventory is now being depleted at a astounding rate.
I would venture a guess that gross advert bids will go down 10-20% due to a 2 fold at the least increase in inventory.
| 3:46 pm on Sep 16, 2004 (gmt 0)|
I haven't gone with multiple ad units yet Ihave too much else to catch up on right now. However, when I heard about the option I was pleased. Not because I thought it could double or triple my revenue but because it gives increased flexibility in site design. Now instead of a tower of four ads I can have two small towers of two ads each which helps in designing a site that people are not as quick to recognize (and ignore) adsense.
In it's own way I think this could increase revenue. But I agree with some others here that slapping several ad blocks (of 4 or 5 ads each)into your page is likely not going to have the effect people are hoping for.
For what it's worth.
| 4:05 pm on Sep 16, 2004 (gmt 0)|
Yes that was our conclusion as well, good for some and bad for others, it will give many webmasters more options for site design and many will make more rev for quite a few I would imagine.
| 4:18 pm on Sep 16, 2004 (gmt 0)|
The most important thing for me in this experiment was to watch what the CTR is doing. Right now, it's a little more than half the CTR was when I only had one block of ads on the page. So theoretically, once the demand pressure builds up again, the overall revenue should be up again. I'm disappointed at the revenue right now, but it's really got to be a temporary situation.
So my question is, are you finding the overall CTR is going up with three blocks of ads (ie, is it more than 1/3 of your previous CTR)? I'm only doing 2 right now, so I'd be interested to know what other people are finding.
| 4:18 pm on Sep 16, 2004 (gmt 0)|
Thanks for posting your findings!
We took off the second ad unit and went back to one...seems that our revenue is getting back on track. My guess would be that higher paying keywords got diluted. From our stand point multi ad units did not pass the test.
| 10:17 pm on Sep 16, 2004 (gmt 0)|
This is an excellent discussion!
If I may throw my 2 cents in, you might consider measuring the results of multiple ad units by examining your overall revenue as opposed to EPC and eCPM. These figures are based on ad units, rather than page performance, so multiple ad units on a page will alter their value.
I'll pass your thoughts on to the team, so that they can come up with solutions to reporting based on the way publishers use their reports.
| 11:24 pm on Sep 16, 2004 (gmt 0)|
ASA, how come it was decided not to repeat the best ads.
Seems to me someone might not want to click a link at the top of my page because they haven't reviewed my content yet, but they might be willing to at the bottom. In this situation I want to have the best ads repeated.
| This 65 message thread spans 3 pages: 65 (  2 3 ) > > |