Google AdSense: Forum optimization study

I got this email from Google Adsense Today...


We are currently working to improve the performance of Google ads on forum
sites and would like to invite you to participate in our forums
optimization study. Participation in the study will include the option for
you to customize your AdSense code to rotate through several positions in
your forums in order to find the best performing position(s). If you are
comfortable with modifying your forum code, you can also extend the
rotation to another targeting option we are beta testing for forums.


I am taking part in the study, so I'll keep you advised as to whats involved. It looks like google is recognizing the traffic and earning power of forums!

That´s great, keep us posted of your development.

Wow! Good news! Definitely keep us posted, Joey.

Ok - I have begun the test. I have decided to use this forum.


There are 2 parts of the beta. I only think I will be testing one.

The first part of the Google Adsense Beta Test involves rotating the ads. You specify locations for the ads; Header, below nav bar, after first post, after last post. You ad some specific code and what happens is that with every pageview, the ad only appears in one of the allotted spots. In the case of this site, I have chosen 4 spots, The header, below the navbar, after the first post and after the last post.

Google suggests several possible spots: See attached image below

Header
Under the navbar,
Under the first post
Under the last post
Footer
Right Column
Left Column
The Second part of the Google Adsense Beta Test is called page section targeting. Using page section targeting, you can designate which section of your pages to target an ad unit toward. An example of usage would be to target the ads near the top of the page to the first post and/or the post title, while targeting ads near the bottom of the page towards the last post on the page.


My first observations...

It seems clear to me at first glance, that Google has not actually looked at how vBulletin works. I say vBulletin only because I dont know how other forum software is coded, but this may also apply to other forum software. I say this because the easiest way to code Google Adsense into vBulletin is to place it in the showthread, forumhome and forumdisplay templates. This makes it so you dont have to use conditionals. (conditionals have to be used to remain within compliance of Google Adsense TOS so ads are not displayed on registration pages, error pages, etc)

However, by coding it this way you will notice on forumhome, the ad is not going to show under the first post or under the last post, because that template has no first post.

In addition, most forums dont have a right or left columns. Here again, that would require conditionals be placed because the vBulletin hack that gives you side columns causes the columns to appear on every page in its stock format.

I also dont like it because it only shows 1 ad per page. The reason I see this as a problem is because it is my thinking that if the ad on top of the page (under the navbar, in the header or under the first post) doesnt catch a reader's eye, then the last one will. (after the last post or in the footer) Under this situation Im stuck with only 1 Google Adsense Ad per page, there is no second chance.

The Google Adsense Section Targeting Beta Test, in my opinion, would only work on websites which have several different topics on a page, such as a newspaper. I cannot see how that would work effectively on a forum.

These are suppose to be beta programs designed for forums specifically. The idea is if the ads move around, then users wont become blind to the ads.

What is everyone else's opinion? Lets hear some other thoughts. I have also sent the Google Adsense Team a link to this thread, so they can see what other forum owners think of this beta.

aaah, so that's what's going on over at DP too. it'd be fine if they were actual ads rather than just taking me to a list of ads. Like I'd want that. If they were predefined search pages with actual listings on the left, and ads on the right it'd be ok. But I won't click through to a list of ads.

That's nothing against the use of them here, just my opinion on those wee ads in general :)

Sarah

It's very interesting that Google is taking such a specific interest in ads on forum pages...

The only thing I feel pretty sure about is that the actual CPC seems to be set lower when the software recognizes forum pages....

I'm not too crazy about modifing placement. I have been quite happy with VB's forum home, showthread and forumdisplay template mods for adsence purposes....

I'll be looking to stick with that but ofcourse will keep an open mind to what turns out....

All-in-all, excellent news Joey :)

No Sarah - thats not whats going on at DP - Shawn just added a different style of Adsense ad - the Adsense Text Links -- that has nothing to do with theis beta test.

Ask them why the ads scroll in FF. In my forums, this are the clients biggest complaints.

PS: Cool that you are able to take part...

what do you mean scroll? please be more specific.... Are they doing it here?

Joey, this is a great study and good research, but I don't want ads in each post, nor do I want to see them on each side of my screen.

People left SEOchat because of all the ads there, they come to forums because of the content in the threads, not to wait for ads to download.

I think if folks want to look at ads they can look at links that lead to ads, for example, your special directories that link to web host listings or vB services directory.

I don't mind an ad between posts as the Google link ads, but when there are skyscrapers and banners everywhere, that takes away from the enjoyment of the readers and they will leave as they did at SEOchat because of page load factors.

The deal here is that Google wants more units displayed on pages and this is their way to do it, we can't fall into a trap of each page becoming a Google page, the ads pay very little now (.02-.05 cents a click) and there are better ways to make money today.

It seems that ADSENSE payouts continue to drop while Google records record profits, seems like they are taking bigger cuts of the cash flow and paying less and less everyday to us.

Why should we give more and more space for lower payouts?

There is something wrong with that picture.

actually anthony - the ads are less now then they were. THere was 2 ads per page - one below the navbar and one below the last post. Now, there is only 1 on the page at a time, it just changes location.

No one ever had any problems the the two ads, so why reduce it?

What we don't like are ads on both sides of the screen, I don't spend much time in forums that have layouts like that.

I can PM you some links to forums that have this sort of layout if you want to see them.

Joey, this is a great study and good research, but I don't want ads in each post, nor do I want to see them on each side of my screen.

People left SEOchat because of all the ads there, they come to forums because of the content in the threads, not to wait for ads to download.


I agree..... I don't feel that the ads should play toomuch of a role in the forum... You start to get that feeling when the ads get more sophsticated....

I wouldn't want to feel like the ads are following me around.... I like them set in one place where everone knowws they are and if they want thenm they are there.....

I may be missing something here - I thought the point of the Google study was to find out what placements work best on forums, not to have them permanently moving around. If the goal is the latter, I agree - no way I want it that way on my forum.

I experimented myself with different ad placement and it seems that in the page header (with conditionals) works best for me.

Joey,

What I mean is, if you are a FF user, and you hit a site with 728x90 ads on it, and you scroll over the ads you will see the ads actually scroll within the page. This often leaves a white bar at the bottom of the Google ads and that is an asthetic no-no...

I am glad to hear that you got into this beta program and I think that the power of forums these days is so great and it's great that they are now taking advantage of it. Also, that diagram there is a perfect way to setup your ads and I use that too and I have been very sucessful with that layout.

No one ever had any problems the the two ads, so why reduce it?

What we don't like are ads on both sides of the screen, I don't spend much time in forums that have layouts like that.


Maybe your not understanding --- I implemented the BETA program here to test it and to get feedack from a variety of forum owners on the program;

Additionally, I said in the beginning I chose to use 4 possible locations, none of which are left or right side. I am using the header, below the navbar, below the first post and below the last post. Only one ad appears per pageview. I also said I wasnt impressed so far, but im trying it out so Google has a chance to gain feedback on this concept. THATS why the change. If not for the aspect of participating in a beta program and trying to help Google Adsense improve, I wouldnt change from what we had.


I may be missing something here - I thought the point of the Google study was to find out what placements work best on forums, not to have them permanently moving around. If the goal is the latter, I agree - no way I want it that way on my forum.

I experimented myself with different ad placement and it seems that in the page header (with conditionals) works best for me.


The goal is the latter from my understanding. The idea being that if the ads are always in one place, regular users of a forum will become 'blind' to them. But if they move around then people wont become blind to them.

Well, if that's the goal and the AdSense people are reading these responses, I can tell you right now I'm not interested. Ads moving around may get more attention -- so do mosquitoes. But not positive attention.

.... ads that move positions, I think something more innovative like ads that rotate without having to refresh the page and ads that disappear would be much better than ads that change positions.

I also think that Google should hire me if they want more ideas :wave:

I've done this specifically for my own forum with the following results:

1) moving ads around annoys the members.
2) ad placement after the last or first post perform the best. After the last post is slightly better than after the first post, but since I use the adsense revenue sharing hack as a way to reward the staff, they have to go after the 1st post.
3) a lot of members actually enjoy content directed ads, there's a few hardliners opposed to any ads, but they'll just use ad blocking software, or become a sponsor so they don't have to view them.
4) vBulletin specific: ad placement in the archive is highly effective. People directed to your site via search engines are frequently directed via an archive link.
See http://hfboards.com/archive/index.php for example
You can do pretty much anything there without compromising the layout of your forums.

Yes, ads after the first and the last post are the way to go.
Vertical banners are horrible and everyone hates it.

I would concur that Google does not yet have an understanding of the dynamics that drive communities. I own and operate one of the largest and fastest growing communities on the net. In addition, my new company will soon be offering something to online communities that will catalyze their growth and offer new opportunities to owners, members, and prospective members.

While it is good to see that Google is taking note of the power of online communities, I believe they are going about it the wrong way. Instead of telling community owners to "try what we have come up with", they should work with those of us who have already optimized every single inch of our communities to maximize revenue from AdSense. When communities are your life and career, you develop an understanding of their dynamics that nobody else can. I know how my members and guests think, I know security issues, I know ways that community owners need to be educated, and I know what works and doesn't work for various types of communities. And I have no doubt that other community owners here know the same sorts of things.

Look at www.big-boards.com -- the website that tracks the 1,235 largest communities on the net. A script one of my programmers wrote shows that in just those 1,235 largest communities, there are 94,433,851 members. Add to that the thousands of smaller communities on the 'net and suddenly you are talking about a significant customer base. Google Adsense is flawed when it comes to communities. Adsense is fantastic for normal sites, but communities are not "normal".

I believe that in order to maximize the utility of Adsense on communities, Google needs to create Google Adsense For Communities (on the principle that you don't use a screwdriver to do the job of a hammer); with over 200 million online community members - and growing - Google is missing out on a huge, relatively untapped niche if they do not adapt.

I can already see a lot that's wrong with this test. But again, I am pleased that Google is beginning to take notice of communities. Because in my opinion, communities are the true critical application of the internet.

David Chapman
dchapman@zoints.com

Just realized we were hijacking the AdSense study thread here... split to "dchapman" in "ChitChat".

Joey,

What I mean is, if you are a FF user, and you hit a site with 728x90 ads on it, and you scroll over the ads you will see the ads actually scroll within the page. This often leaves a white bar at the bottom of the Google ads and that is an asthetic no-no...

I believe it's caused by the way FF renders the frames Google uses to display the ads. It does the same thing on my site.

Ok - its only 5 days in -- but here are how things are going so far.. I plan to keep you guys posted as often as I can....

Ok - as a reminder, we are testing 4 positions - the Header, Below the Nav Bar, Below the first post and below the last post.

One thing - the header position has been getting shown a disproportionate amount of times. More then 3 times the amount of the other ad positions.

Yet - it is the least productive.

Percentage of Total Impressions

Header - 51%
Navbar - 15%
1st Post - 18%
Last Post - 17%

Percentage of Total Clicks

Header - 5%
Navbar - 30%
1st Post - 35%
Last Post - 30%


-------------------

Do they not allow you to specify display percentages? In other words, how often each ad displays in proportion to the others?

The header position is about the worst place there is. I would not be very happy if it showed three times as often as the others.

Give forum members a week. They'll be blind to every one of those ads. Community members see the same page design over and over again as they refresh and see new threads. They see the same site design exponentially more than normal website visitors. Due to this, forum members become EXTREMELY ad blind. At my largest community, 11.5% of visits by anyone to the site are for 15 minutes or more. The averge visit is 6 minutes and 6 seconds. That is HIGH compared to normal sites. Because people spend so much time on the community, they stop paying attention to ads and no longer click them. Or they just block them altogether with adblocking software. There are over 200 million forum members out there on the net, and growing. That means there are over 200 million people who are ad blind! By comparison, the population of the United States is 295,734,134. That's a lot of lost revenue.

Community owners need options that work with, rather than against the dynamics that drive a community.

Thank you for taking part in this and sharing your findings with us Joey. I live for this stuff!

I wonder about these results on two grounds:

1. the number of members and regular visitors here -- would it be different on a busier forum?

2. the type of members -- everyone here is, I think, a forum owner -- would it be different for more typical forum members?

3. the definition of "header" in this case -- it's right up there to the right of the logo -- that doesn't mean that a page top position just under the logo wouldn't be more effective.

re: #3 -- I saw a significant increase in clickthroughs when I moved my AdSense banner from forum footer to forum header. But in my case, that "header" position is below the logo and navigation links (see http://www.psychlinks.ca/phpbb/index.php).

I guess we have to face reality here guys and gals, we never made that much from the DP rev. share program, so ADSENSE on forums just does not make a hell of a lot of money.

I think that DP mentioned the most he made was $500.00 a month if that, I know I would be lucky to make $10.00 in a years time and I was the leading poster there.

CPC payouts are the main problem, not where the ads are on the page, you just can't make money with CPC payouts at .02-.04 cents a click unless you get a million clicks a year on your ads.

I wonder about these results on two grounds:

1. the number of members and regular visitors here -- would it be different on a busier forum?

2. the type of members -- everyone here is, I think, a forum owner -- would it be different for more typical forum members?

Definitely. Different demographics and community types have different habits. As a community owner, you have to understand the dynamics of your forum and adapt your ads based upon that. A couple examples:

1. On my busiest site, at peak times, we average more than 10 posts per minute. Many members average more than 40 posts per day. They are computer literate, 16-25 years old, and predominantly male. They are completely blind to anything in the header or footer. Move ads around on them like this test and they will get very upset (we all know that community members HATE anything along those lines). Put ads after the first post and they will either become ad blind within 24 hours or get upset. A unique resolution had to be developed (one that we are seeking a patent for so I can't comment on it at this time, sorry).

2. On another community I own that was started about 3 months ago, we average only 100 posts per day. The demographic is ALL female, 18-30 years old, and not overly computer literate. I have never heard a word out of them regarding issues with the ads. Click through rate there is much higher. They are slower browsers, read more carefully, post more content rich posts, and if an ad seems to have additional information for them on a topic, they are more apt to click it.

Every community I own responds differently to ads and their placement.

I guess we have to face reality here guys and gals, we never made that much from the DP rev. share program, so ADSENSE on forums just does not make a hell of a lot of money.
With all due respect, 5 figures a month are possible with Adsense on communities. I would even venture to say six figures if your community was large enough and the topics covered were competitive keywords and phrases.

I guess we have to face reality here guys and gals, we never made that much from the DP rev. share program, so ADSENSE on forums just does not make a hell of a lot of money.

I think that DP mentioned the most he made was $500.00 a month if that, I know I would be lucky to make $10.00 in a years time and I was the leading poster there.

CPC payouts are the main problem, not where the ads are on the page, you just can't make money with CPC payouts at .02-.04 cents a click unless you get a million clicks a year on your ads.

Actually, im convinced that the problem with DP's revenue sharing is the placement of the ads. My DP CTR is 1/5th that of my own forums overall CTR. I dont use header ads on my other forums.

The biggest trick to forums is traffic, new traffic as well as regular traffic.

dchapman sure it is possible, if you have 500-750 members online live 24/7 :o

How many forums are that large out here? You may know of them through your rating service that you linked to the other day.

There are not many in the webmaster space that come close except SitePoint and maybe some of the domain name forums.

General forums that cover popular subjects like music or the daily news might have a chance, but it takes a lot of time to get to the figure you mention and administration of that sort of forum is much more than a one man start up show.

Actually, im convinced that the problem with DP's revenue sharing is the placement of the ads. My DP CTR is 1/5th that of my own forums overall CTR. I dont use header ads on my other forums.
That was one of my points above. DP's header ads are to the right of the logo just like here -- who even sees those most of the time?

But move them down a few centimeters, just below the nav bar, and I think you'll see a difference.

Thank you very much for creating the thread discussing this test and
vBulletin forums. I've passed on the feedback to our team here, and we'll
definitely take it into account going forward.

We are currently aggregating the data from the study. Please feel free to
remove the rotating ad code, or ads from certain positions from your
forums if they are not performing well. We hope that you found the results
of the rotating ads to be useful. We should follow up shortly with the
overall results from the study.

Sincerely,

Leslie
The Google AdSense Team

Thank Joe.

I hope they will come with a solution which can improve AdSense earning on the forums as well :)

Good details on info. Thanks !

It really depends on the kind and amount of traffic and features you got in the forum to earn with community. Most members are ad-blind. They post several messages, read then leave. I even tweaked one of our forum so the adsense will look like a signature (except for the "Ads by Google") of a member but it just didn't work-looks almost the same as ads after first post but it uses 468. Also tried rotating colors, the header, left and right side skyscraper, bottom leaderboard ads, the standard banner size but still get a few clicks. Though there is still one the that I havent tried, the 336x280 text only ad unit. I even used the adlinks but i got the same result.

Now I got to try blog. :D

Ok, im discontinuing this beta......

Final results were interesting.......

The header was the worst performing, by far.

Below the first post performed slightly better then below the nav bar as far as clicks and CTR are concerned, but interestingly enough, did not perform as well as the navbar dollarwise.

Below the navbar earned almost 4 times as much as below the first post - despite having virtually the same number of clicks and same number of impressions.

Below the last post also did not earn much dollar wise despite being only slightly below navbar and first post in number of clicks and impressions.

All in all - if your Choosing one place to place ads, it seems below the navbar is the best place to go.

Try http://www.click-share.com

I am working with them from 6 mounts and have very good results with very high CPC.

They also provide products ads.

When will this be open to the public?

Try http://www.click-share.com

I am working with them from 6 mounts and have very good results with very high CPC.

They also provide products ads.

Thanks man..

But I was thinking more about free stuff i can use..

Amigo!