AdMoolah News and Views

The New York Time has an article which look at Google’s advertising system. Although the focus of the article leans more towards the advertisers side of things rather than publishers, it is still worth a read.

Here are some of the bits that I found interesting.

Because of this technology, users click ads 50 percent to 100 percent more often on Google than they do on Yahoo, Mr. Noto estimates, and that is a powerful driver of Google’s growth and profits. “Because the ads are more relevant,” he said, “they create a better return for advertisers, which causes them to spend more money, which gives Google better margins.” (Yahoo is working on its own technology to narrow that gap.)

This certainly seems to be true given what publishers are saying about YPN vs. AdSense. AdSense seems to have much better CTR than YPN, and relevancy seems to be a big part of it.

Here, as in other places, many advertisers criticize Google for being like a black box, because the company gives them less specific information and control than they would like.

Although I don’t follow the advertiser’s side of things as closely as I do the publishing side of things, this seems to reflect what publishers are feeling. The lack of information about what is happening in the advertising system can be frustrating. So far Google seems to take the approach that the system works, so there is no need to reveal any of the details.

Kevin Scaldeferri, who is a Developer at
Yahoo! Search Marketing has an interesting post.

He is theorizing that one thing that could be affecting the CTR of YPN vs. AdSense could be the length of the ads themselves. Since Yahoo ads allow for a longer descriptions than AdSense does, it is possible that people reading them get a better idea if they really are interested or not. This would lead to a lower CTR, but better conversions.

Google has posted an entry about smart pricing on the Inside AdSense blog.

A lot of it is in line with my last post about smart pricing.

Here are their main point along with some of my own commentary.

1. Many factors determine the price of an ad

This is one thing that makes it so hard to determine how much smart pricing is affecting your earnings. The algorithm for determining CPC is so cpmplex that it is almost impossible for a publisher to know that they have been “hit” by smart pricing. See my post on Why Do AdSense Earnings Vary for more details.

2. Click-through rate doesn’t affect advertiser return on investment (ROI)

This is a mistake that I have seen many people on the forums making. They think removing their low CTR ads will mean they will do better with smart pricing. It is impossible to determine conversion percentage based on CTR, there is no relation between the two numbers.

3. Google doesn’t make money from ‘smart pricing’

As I mentioned in my last post, Google doesn’t use smart pricing to pocket money on the transactions. It provides for better ROI for advertisers, making the system better. A better system for advertisers means they will spend more money, meaning there will be more money to be made for publishers.

4. Remember the old chestnut: “Content is King”

This is something I am constantly saying. Building good quality sites not only brings visitors back, but they will also be genuinely interested in the subject of your site, meaning they will be more likely to click on the ads, and also more likely to converted by advertisers.

One interesting omission from this list is the possibility that smart pricing performance on one site can affect all of the sites in an account. This leads me to think that this might be true. I think it is this fact that has most publishers upset. It doesn’t seem quite fair that they would penalize all sites across an account because of one bad apple. Since Google has all the data it is hard to tell what they are thinking. It could be that the statistics show that if one site in a publishers account does not convert clicks well that it is generally true that the other sites do not convert well either.

One possible solution to this is for publishers who are large enough to open multiple AdSense accounts under different company names. This will allow the to try to segregate poorly performing sites from better ones. The problem with this strategy is that it is still almost impossible to determine which sites convert well. Publishers just don’t have the data.

It’s good to see a little more information on this. I just wish we had some more specifics about what does go into the smart pricing algorithm. I realize this would make it easier to game the system, but it would also relieve some publisher frustration.

AdSense Smart Pricing has gotten a lot of buzz in the past few days. It all started when a thread on WebMasterworld was started by kurtpdx. In it he says Google called him and during that conversation it was revealed that:

1. Smart Pricing can affect multiple sites across an account. That is, if one site in your account get reduced payouts per click because of smart pricing, other sites in your account will also get reduced payouts.
2. Smart pricing is re-evaluated weekly
3. There is a 30-day cookie that Google uses to track conversions.

The reason I called this post Smart Pricing “Rumors” is that none of this has been confirmed. This is second hand information provided by an anonymous poster. And the information he gives is based on a phone call from a single customer service representative at AdSense. It is possible the rep got things wrong, that the poster misunderstood, or that its just a plain fabrication.

One thing that is true is that publishers know very little about smart pricing. This can lead to some wild speculation and lots of frustration. Google is very close lipped about exactly what goes into smart pricing, making it very difficult for publishers to know how to counteract the effects of smart pricing. One of the reasons I started AdMoolah was because Google supplied so little information to the publishers, and this is yet another example of that.

Many people are suggesting trying to remove AdSense ads from poorly performing pages to avoid the effects of smart pricing. I think this is a nearly impossible task for a few reasons:

1. It is impossible to know which pages on a site don’t perform well. Publishers have no visibilty into which clicks on their own sites actually convert to actions on the advertisers site.
2. If Google does use a 30 day cookie to track conversions and it takes up to 7 days to re-evaluate smart pricing on a site, it could take 37 days for any changes to take effect. Given this huge time lag it would be very difficult to ascertain if any changes are actually working. So many other things can happen in such a large time frame it would be nearly impossible to separate the signal from the noise.

Something to remember is that Google doesn’t do this simply to punish bad publishers while lining their own coffers. If smart pricing is applied to a site, Google also gets less money. Smart pricing is there to give the advertisers a better return on advertising dollars.

I think Google should try to do some damage control. The level of frustration from what people perceive to be smart pricing is getting dangerous. People are leaving AdSense to try Chitika and Yahoo Publisher Network, and many are finding success with these other programs.

Additional commentary about this can be found at JenSense and ProBlogger.

Today, high profile software development blogger, Joel Spolsky, posted an entry Something Rotten in AdSense. In it he talks about the problems of click fraud and in particular, click fraud coming from splogs. He comes to this rather startling conclusion:

I predict that you’ll see a massive expulsion of smaller AdSense sites by Google, and it better happen soon, or AdSense will ruin Google’s reputation among advertisers, something which could be deadly.

If you are a small publisher this may worry you. Don’t be worried. I think this is a very hasty conclusion. I have been a small publisher with AdSense, and I am very familiar with both AdSense ( I run AdMoolah) and splog (I run FightSplog). I’m not worried.

There is no denying that click fraud does exist. People will always try to click on they own ads, encourage or pay others to click on them, and maybe even, as Joel suggests, use zombie PCs to generate clicks. But there are reasons why this won’t become the overwhelming problem Joel worries about

1. Click fraud is hard to pull off on a large scale. If you try to use any one method to produce a lot of fraudulent clicks, it becomes easy for Google to track these.
2. Advertisers accept some level of click fraud. Pay-per-click advertising has been shown to produce a good ROI for many advertisers. The advertisers recognize that some click fraud is going to happen, but as long as the overall ROI remains good, it is an acceptable problem.
3. If it really becomes a problem, Google can start to audit publisher web sites on a site by site basis. They could use automated filters to flag any suspicious sites and not allow them into the network. Of course there will need to be an appeal process to let legitimate sites into the network after human review, but this shouldn’t be a big deal for Google to take care of.
4. Smart pricing helps control costs. Any sites that get clicks that produce poor results for advertisers will be hit by smart pricing. This reduces the cost of clicks on poor quality sites, and helps mitigate the problem.
5. Advertisers don’t need to use small publishers. If an advertiser feels that the clicks from small publishers don’t have enough ROI to justify the ads, the advertiser simply won’t use the sites. Advertisers can use site-targeting to select just the sites that they want their ads to appear.

Don’t worry, Joel, the sky is not falling. Do you thing Yahoo! and MSN would both be trying to join in on this game if it was a doomed business model?

JenSense has a great tip on displaying ads when a visitor to your site does not have JavaScript enabled. The basic technique is to use the NOSCRIPT tag like this…

<script>
... Google AdSense Code....
</script>
<noscript>
.... alternate ad here....
</noscript>

Obviously this means you need a non-JavaScript alternate ad, so this excludes programs like YPN or Chitika, but affiliate ads will work fine.

One excellent use of Chitika eMiniMalls is having them as alternates for AdSense ads. AdSense Alternate ads replace public service announcements that may appear in place of regular AdSense ads when Google is unable to serve targeted ads to your page.

The steps to use these eMiniMalls as alternate ads is fairly simple, and I’ll outline them here.

1. If you haven’t already done so, sign up for a Chitika account. The approval for your account usually happens within a day or two.
2. Make a note of the size of the layout for the AdSense ad you want to create an alternate ad for. For our example well use a 160 x 600 Wide Skyscraper ad.
3. Log into your Chitika account and choose the “eMiniMalls Code” link. From the Ad Layout selection choose the size that matches your AdSense layout. Then choose the colors you prefer for you eMiniMalls ad. Again for our example we choose the 160 x 600 Wide Skyscraper.
4. Copy your eMiniMalls code from the textbox on the Chitaka page. It should look something like this:
   

   <script type="text/javascript"><!--
ch_client = "example_client";
ch_width = 160;
ch_height = 600;
ch_non_contextual = 1;
var ch_queries = new Array("digital cameras", "ipod mini", "sony playstation", "dell laptop");
var ch_selected=Math.floor((Math.random()*ch_queries.length));
ch_query = ch_queries[ch_selected];
//--></script>
<script src="http://scripts.chitika.net/eminimalls/mm.js" type="text/javascript">
</script>

5. On your web server, create a file named specifically for this size ad, using the “.html” extension. (The exact details of how to create this file will depend on the specifics of you host.) In our example we choose to create a file named chitika_wide_skyscraper.html
6. Paste the code that was copied from you Chitaka code text box into this new file.
7. Save the file
8. Log into your AdSense account
9. In the “AdSense for Content” – “Ad layout code” page, create a new ad using your chosen size and colors.
10. When you get to the Alternate ad URL or color section, check the Alternate URL radio button, and type the full URL of your save Chitika ad file name into the text box.
    
11. Select the code form the “Your AdSense code” section and paste this into you web page just like normal. It should look something like this:
    

    <script type="text/javascript"><!--
google_ad_client = "pub-111111111111111";
google_alternate_ad_url = "http://example.com/chitika_wide_skyscraper.html";
google_ad_width = 160;
google_ad_height = 600;
google_ad_format = "160x600_as";
google_ad_type = "text_image";
google_ad_channel ="";
//--></script>
<script type="text/javascript"
src="http://pagead2.googlesyndication.com/pagead/show_ads.js">
</script>

That’s it! Now, when Google cannot determine what to display because your page is brand new, or there is not enough inventory, Chitika eMiniMall ads will be displayed instead.

(Note: The Chitika links in this post are referral links.)

I got a comment in my Is Rotating YPN and AdSense a Bad Idea? post, saying:

my name is paul mineiro and i’m a ypn developer.

i agree we need a blog, the miscommunication around this was pretty big.

so i started one:

http://360.yahoo.com/warezio

I think that’s fantastic. There are other bloggers who have done very well blogging while at big companies like Robert Scoble at Microsoft (or it might be this link now, he’s switching blogs) , Matt Cutts at Google, and Yahoo’s own Jeremy Zawodny. Hopefully Paul’s blog will be as useful as the aforementioned ones are.

Welcome to the blogosphere Paul! I’m looking forward to what you have to say.

There is a new post today on the inside AdSense blog. This new entry includes a letter that was written by a publisher that has recently implemented image ads and seen good results. It has some interesting nuggets of information.

We’ve seen a 34% increase in AdSense revenues by allowing image ads in our ad codes and implementing section targeting on key pages. These technologies by their nature mean that impressions may initially drop, however, we have still seen increased revenues because our eCPM has doubled.

I’m not sure I understand why impressions may drop due to implementing image ads. There should be a larger of inventory of ads to choose from, meaning better there should be fewer PSAs shown. I guess I could imagine that section targeting may reduce the impressions since it could narrow the focus of the page, increasing the chance of PSAs. But overall, I’m surprised that Google let this statement go in without any explanation.

The letter goes on to talk about reporting…

Of course Google doesn’t give us all the stats yet on image v. text v. site-targeted ads, so we can’t be sure exactly what the cause is .

and Google responds with…

… thanks for sharing this part of your AdSense success story – and how additional reporting will be useful for you. No promises, but we’ll keep doing our best to give you what you need to maximize your AdSense earnings.

This looks like good news! Yesterday I posted about how better reporting was on my AdSense Wish List.

A third curious thing about this letter is this statement:

Because our AdSense ads are super-targeted, our customers tend to view them as “additional content” rather than diversionary advertising.

I have heard that before from AdSense case studies but I never really believed it. That seems to be stretching the truth a little. I agree that AdSense ads are less annoying than banner ads, but I would never view them as “additional content”. To back up my views we see that the site that the letter refers to, Recipezaar, has a “Remove this ad” link which leads to a membership page. I seems rather odd that users would want to remove this”additional content”

Bloggers have been buzzing about Chitika eMiniMalls lately. Today Darren at ProBlogger posted some Chitika eMiniMalls Tips and Jennifer at JenSense has posted the Inside scoop on making Chitika eMiniMalls AdSense & YPN compliant.

Darren has reported making good money from eMiniMalls, and others have similar experiences. I have also been testing Chitika ads on a few sites for a several days, but my results don’t seem to be as spectacular. I think the problem is that my sites are not quite the right type of sites for eMiniMalls. They seem to do much better in consumer goods oriented web sites like gadget blogs.

Nonetheless, it is good to have alternatives when it comes to AdSense. AdSense, YPN and Chitika all need to compete for the same publishers, and different alternatives work better for some people than others.