Forum Moderators: martinibuster
The problem is what some are referring to as the "Adsense Penalty". Others, not sure what has happened to their traffic, only know it happened right after adding Adsense code, or right after site renovation.
In looking at the source of some of those pages, have identified a problem, that is really simple to correct, but it is becomes complex if you let it get it out of hand.
The problems is the misplaced Adsense code. When adding the Adsense code to the page, Adsense advises, and rightly so, that you place the code near the top of the page, above the fold, for best CTR. I agree completely. However, when you do that, if the code is above the code for your content, Google is now upstaging your content.
The Adsense code, in part, has the following snipet:
google_ad_client = "pub-";
google_ad_width = 728;
google_ad_height = 90;
google_ad_format = "728x90_as";
google_ad_type = "text_image";
google_ad_channel ="";
That snippet has the keyword 'google' in it 6 times, and is often seen right below the publishers meta tag, within the page code. The keyword 'google' now six times within the top 50 lines of your html has become a very important keyword, and is doing nothing at for your page seo. In fact, it is hindering it, by loading irrelevant keywords above the keywords in your content.
Marginal webmasters may not notice this, and marginal websites will be severely affected by it. If the site already has more longevity and higher traffic levels than their competitors, the Adsense Offense, as I call it, may be minimal. However, if the field is very competitive, the site is marginally seo, and the traffic begins to decline, the percentage of affect the AO will have, will increase. There is less to absorb the offensive keyword impact. Short pages, on small newer sites, with lower traffic will feel the hit more than large sites with longevity, with longer pages, and more traffic.
The problem is this. Relevancy is decided by many criteria. One major factor is placement in the code. The rule is this. The most relevant keywords are placed at the top of the page. This is why the meta title has such strong weight. Also seen at the top of the page, is the meta description. I have seen on occassion the meta title, placed in the code, below the meta decription. It should actually be above the description, in the code. When it is below it, the webmaster will have to work harder at seo, over their competitors.
After the meta title and meta description, if the next thing the bot sees is google six times, you have now given 6 votes of confidence for google, in that very critical top 200 lines of code. That is a mistake. It actually helps Google, and hinders you.
They have their #1 keyword on every Adsense publishers page, 6 times. If it is in the top of the code, you have made them very popular. Why do you think that many sites with feeds offer them free? You might say it is because they will get traffic from you. More importantly, look at their code. How many times do you see their brand keywords in the code? I have seen feeds and java script that other sites offer webmaster free of charge, have their brand keywords in the code as many as 20 times. Any more than twice is excessive.
The Adsense code must be coded into the page at the bottom of the page, to minimize the impact it has on your seo, while the code is actually displayed at the top of the page. This keeps the Adsense ads visually at the top of the page, but not letting it upstage your keywords, within the balance of the content.
While the Adsense code may have always been there, and had no effect on the site, other things may have pushed your seo ratio down, and the Adsense code contributed to it.
Example:
The Adsense code was placed on the page a year ago. Traffic was at 10,000 per day. Burbon came along, traffic declined, and Burbon is blamed Was it Burbon? Maybe yes, maybe no.
The Bourbon update coincided with another web scenario, the 3 month average. Traffic after winter went into decline for many sites, beginning in Feb as things warmed up and more people began to get outside. Traffic is not only calulated on the web daily, but also monthly, and more importantly, quarterly... every 3 month period. So, when Burbon came along, it was calculating traffic to the sites for March, April, and May. That was incorporated within ranking algo. This means, alot of sites ranked lower, because their site traffic declined more than their competitors.
If your site traffic declines by only 10 hits a day, but your competitors only declined by 8, 7, 6, and 5 a day,while other competitor's traffic increased, when all other things are equal, you could be hit very hard, even though the decline was minimal, when considered over the 3 months.
Further, if your traffic did not decline, but did not increase as much as your competitors, for 3 consequetive months, your ranking may also drop, considerably. The 3 month average change has a larger impact on serps, than monthly. You are actually given the benefit of the doubt, and not immediately hit hard, on a monthly basis, for slight declines in traffic, while you are rewarded monthly for traffic increases, the 3 month average of traffic increase can have a much larger impact on your traffic.
The 3 months average gives you a chance to fix things, before you drop considerably. Unfortunately, most webmasters neglect their logs, and don't notice anything until 3 months has passed, and the update slaps them in the face. By, then, the downward wheels of destruction are in full motion, and to stop the freefall, will require much more work. The problem could have been avoided, by advertising in the spring, to maintain traffic levels.
Since Burbon happened at the exact same time as the 2nd largest decline in website traffic, on an annual basis, for any consequetive 3 month period, and included those past 3 month spring averages, it was easy to think that Burbon was the culprit. The culprit was you. You didn't maintain your traffic.
Then, because ranking was reduced, traffic declined again, and greated a compounded effect, each consequtive indexing.
Now, with less traffic, the Adsense code has a larger impact. Before, you had the ranking and traffic to offset the minimalization of your seo, now you do not, and the impact from that code above your content is having on your site, is more severe.
It is important to remember...
It is not always that your traffic has decline, but that your competitors may have increased. Don't let that confuse the issue of ranking loss. You can have an increase in traffic, but loss in ranking, simply because others have traffic increases.
Today, I have the #1 ranking. It might be because my competitors are typical summer slackers. If come fall, they advertise heavily, and I do not, their traffic will out pace mine, even though mine continues to increase, and they will snatch my #1. That will reduce my traffic, reduce my quarterly average, and kill or weakend my winter...if I let it.
Popularity, the part of it that relates to how many visitors you have, not inbound links, is the single most important factor in ranking of a properly seo site. When you let it slide, you put wheels in motion that may not have immediately effect, usually don't, but will surface for a fact, 3 months down the road. Then, if during that time you did site renovations, changed things, added Adsense code, the problem will become very complex to troubleshoot.
The first line of defense is watching logs, site stats.
The second line of defense is watching for any sign that things are going south with Alexa. Alexa will help you see signs of future problems, by watching the DAILY traffic stats for competitors. If you see their ranking is 4,000 for 3 months, their daily is 2,145, and the weekly average is 3,890, you need to hunker down. 3 months from now they are going to kick you in the head and leave you wondering what happened. The writing is on the wall. Look at their increases and decreases in ranking over the day, week, and quarter. How does it relate to yours? You need to match them, or out pace them, in order to maintain ranking.
Finally, get that Adsense code out of the top of your html. It is a silent traffic killer.
The Adsense code, in part, has the following snipet:google_ad_client = "pub-";
google_ad_width = 728;
google_ad_height = 90;
google_ad_format = "728x90_as";
google_ad_type = "text_image";
google_ad_channel ="";That snippet has the keyword 'google' in it 6 times, and is often seen right below the publishers meta tag, within the page code. The keyword 'google' now six times within the top 50 lines of your html has become a very important keyword, and is doing nothing at for your page seo. In fact, it is hindering it, by loading irrelevant keywords above the keywords in your content.
Can you tell me which search engines count Javascript variable names embedded in HTML comments as on-page keywords?
I would question the following especially:
1. Does Google index the adsense code? I do not believe so - it may note it, but I doubt if it indexes it.
2. Does Alexa really matter that much?
3.
Popularity, the part of it that relates to how many visitors you have, not inbound links, is the single most important factor in ranking of a properly seo site.
Ranking where? Visitor numbers are absolutely crucial but I am not sure they have, directly at least, anything much to do with your SE rankings. Of course lots of visitors may lead to more inbound links, which is another story.
Firstly search engines don't understand javascript. So possibly Google just skips everything between the script tags.
Google also claims (somewhere on their website) that joining AdSense will not influence your rankings. So it's possible that they just skip all those AdSense Javascript lines. Processing it would just a waste of bandwidth and processing power. Google doesn't need an extra vote for itself.
But while the AdSense codes may not influence your Google rankings it might alter your rankings in other search engines as they use other algos.
You also speak about maintaining your traffic. Search engines do not have a reliable way to check how much traffic your site. Google (and others) simply can't know how much traffic your site gets. They can make a guess but I doubt they would use it to rank sites.
AndI wouldn't count on Alexa. Many people don't use the Alexa toolbar so Alexa isn't really a good way to compare website traffic. Two years ago I used the Alexa toolbar for a few weeks and I noticed my Alexa ranking improved with more than 30,000 (I don't know the exact number anymore, it could even be more).
[edited by: Zygoot at 6:39 pm (utc) on July 18, 2005]
We've been running Adsense since it started in 2003, multiple ads at that, and our traffic is consistently growing.
Search in the engines for:
google_ad_height = 90;
You will find 44,000 pages in yahoo, and each never mentions google. You will see in the description, the entire Adsense code in many of those descriptions, if not at least a line or two of it.
You will see 74,600 results for it in Google, same as above search results of the java script codes.
You will find 4,698 at MSN. Same descriptions with java script.
See how many of those pages have the code on twice, and more relevancy, than those with one ad block, and how many have 3 ads blocks, and even more relevancy.
Though relevany is higher on Blogger, likely because of the word 'Google' within the Blogger code. However, the java snippet in text form within a forum, ranks #2 in msn.
#3 at yahoo, is not a Blogger page, and has no 'google' content, except for the 3 ads blocks of adsense, all at the top of the page.
To the post above, I didn't say anything about the Alexa tool bar. That is a visitors tool. I said use Alexa to judge the traffic ranking, and all sites are included in that, not just those who display the tool bar.
You don't think so? Try it. Drive 5,000 visitors a day to your site, extra visitors, for one month. See what happens, then come back here and post.
They can, and they do use traffic as a measure, incompared to other sites. If Alexa can do it, what makes you you think that Google can't?
I have a site right now getting 8 hits a day. It has no ranking. I will promise your right now, if I do nothing to at all to it, but advertise that site and drive an abundance of traffic to it, that site will rank within 45 days, without any increase in inbound links or changes to the pages. It will because that industry is in a slow down right now, and my increase in traffic will outpace all other sites. How do I know? I have already done it with another site.
You will see 74,600 results for it in Google.
But that's not very many, is it? There must be millions of pages with AdSense code on them. Why are only a very small number of them showing up in these searches you did? Maybe Google really is omitting Adsense code from their index, and it's only being picked up on certain kinds of pages, perhaps ones where the AdSense code has been inserted incorrectly.
One other problem I have with your theory is that I don't see how AdSense code could have a large impact, even if it does have some. So many sites have big chunks of code at the top of their pages--navigational elements, css stuff, various bits of javascript--would AdSense code have more of an impact than that stuff?
Most had made a complete mess of adding the Adsense code, either had other code in the middle of it, had pasted it into a Word doc first or otherwise destroyed it or had left out the <Javascript> tags.
Others were forums where some of the code was quoted on the page (this post will join them soon!).
One was a host offering clients some Adsense code to put on their pages (Hmm?)
A few had no obvious reason to be showing - but had messy, messy, messy code which leads me to believe that they had made errors but have no correted them.
A few more, also with very messy code, had no sign of Google ads on them at all - maybe they just could not get them to work?
I did not find one with correctly entered Google code in the source.
You have obviously thought long and hard about this, but I really think you are barking up the wrong tree using this as an explanation for anything.
To the post above, I didn't say anything about the Alexa tool bar. That is a visitors tool. I said use Alexa to judge the traffic ranking, and all sites are included in that, not just those who display the tool bar.
All Alexa data comes from the Alexa toolbars. The Alexa tooolbar tracks the surf history of its users and sends this data to Alexa.
They can, and they do use traffic as a measure, incompared to other sites. If Alexa can do it, what makes you you think that Google can't?
Maybe a very small part of the algo is based on this but I doubt it.
You will see 74,600 results for it in Google, same as above search results of the java script codes.
When I click on the cache of those results I get pages of mostly webmasters showing a text version of the AdSense code.
A few had no obvious reason to be showing - but had messy, messy, messy code which leads me to believe that they had made errors but have no correted them.
Click on the "cached" link and you'll see Google say that "These terms only appear in links pointing to this page." So those sites probably have scraper sites linking to them where the scraper grapped the javascript and is displaying it or sites linking to them with broken javascript.
It's pretty obvious that Google isn't looking at the javascript except in cases where the javascript is included correctly and becoming "visible."
-- Roger
I have also found janethuggard to make some pretty insightful comments in the past.
A few months ago I was making some major alterations to a page that had a javascript script that pulled in some links and an adsense script. At one point I had uploaded the page when it was pretty much empty of everything but the two scripts. To my surprise, shortly after the upload the adsense ads were targeting the text in the other javascript script.
Try it and see for yourself if adsense targets the content of another script.
It seems there may be some confusion about how using AdSense might affect your sites SEO. To clarify, AdSense does not affect either positively or negatively your site traffic or its ranking in the Google search results.
For more information about how sites are indexed and ranked, you might be interested in the following page:
[google.com ]
This page also explains how PageRank is designed to maintain Google search results as:
a source of objective information untainted by paid placement.
Hope this helps!
-ASA
It reached out to me that all my most important keywords were at the bottom of the page, not at the top. I had noticed that before, but seemed I read something about it once. I wondered if that made the difference. So, I browse the web for info, and found numerous articles that indicated, top part of the code got the most attention in the serps.
The rule was:
If it is important, get it at the top of the page.
So, I moved the code around, without removing or adding anything. I liked my design, but it was dead in the water.
The site was designed with software, and that code was assembled based on the order I placed things on the page. So, if the first thing on the page went at the bottom of the page, it would actually show up first in the code source, below the meta.
To fix it, I had to create new code on a new page, start from scratch, and put the most important things on the page first, in order of importance, then remove the code from the index page, and paste in all the new code.
After moving the code around, getting my large bold font words seen at the top of my page, in the top of my code, Google came a few days later, and I was back in. While it sounded easy to solve, the process took us several months to correct...every page had to be redone.
Later, I made more changes, and one day forgot, and removed something important at the top, and then recreated it, and readded it. It was then in the bottom of my code again, but I didn't notice. Because that piece of code is some of my most important, I dropped again like a rock. Thinking on it, I sourced the page again, and the error was easy to see. I switched it back, replacing everything on the page again, in the correct order, most imporant at the top of my code, and have not made that mistake again.
However...
My partner has made the mistake MANY times, forgetting to get the code on the page in the correct order, and when he asks for help, I source the page, and same thing. He corrects it, everything is fine again next indexing.
On one occassion, I specifically remember he had 3 sets of Adsense code on the page, and the page lost all ranking. He asked me why. I sourced the page and seen he had redone the page, putting the Adsense code at the top of the code, and the content below it, even the logo with alt text of his site, seen at the top of the page, was below the Adsense code. I had him put every back on the page, in the correct order by importance, Adsense ads on last, and the ranking came back.
We have tested this many, many times. In another case on my partner's site, we wondered why we were ranking strongly for a keyword, but not the one intended. I notice in the search engine listing, they picked up a bunch of keywords, that we had not intended to rank, as they were minor, and didn't reflect the majority of the content. When I sourced the page, there they were, at the top of our page, the very first words indexed in our body, though they were seen visually in the lower middle. They comprised our listing, until we moved them back into the code. Next indexing, the search listing changed, as did the ranking keywords.
Another time our listing and ranking was based on a design company logo, seen back then, in the lower left hand corner of the page. When we sourced the page, it had been added to the page first, and became the first thing the bot saw within our body content. We rearranged the page again, everything was back to normal. I could go on and on with examples.
I have no doubt at all that the code at the top of the page is the most important and carries heavy weight in keyword density, along with large font and bold text. But, unlike reducing the font, or removing bold, it can drastically decrease search engine ranking especially for marginal site, who have to count heavily on that aspect of seo to compete. Again, it may have no noticeable effect on a site on the web for 10 years, with 1 million visitors a day...
Everything is a percentage, and more of everything means you can mess up more, without decline.
Given that, the further down anything is in the code, the less signifance it will have. If you place 3 sets of affilate code above your content, you have pushed your body content over 30 lines further down in the code. I believe that can reduce the ranking, for a site that is marginallly holding it's own against competitors. It may however have no effect at all on a strong site, with longevity and popularity over their competitors, by a long shot. The difference would be minimal. It is the sites that don't fit that model that can have a much more detrimental effect.
In the case of Yahoo result #3, he is ranking for the
Adsense script keyword, because the page has very little on it. It constitutes a high percentage of his page content, and explains why he ranks for it. That is not normally the case with most pages.
AdSense does not affect either positively or negatively your site traffic or its ranking in the Google search results.
I think this is clear as day ...
If architects adopted this theory, one woodpecker would decimate civilization.
[edited by: incrediBILL at 7:16 am (utc) on July 19, 2005]
I have no doubt at all that the code at the top of the page is the most important and carries heavy weight in keyword density...
Keyword density, H1, bold text matters less today than ever. That was the takeaway from Update Florida, the writing on the wall indicating that Google intended to get away from anything a webmaster can do to manipulate their ranking, while zeroing in on sites with the answer to a particular query. As far as Google, links remain important.
But the ultimate spoiler to this theory is to do any number of searches and see how many of the top ten results have Adsense in them- in any search engine.
Also, as martinibuster points out, if your theory were true, then a lot more sites would have been hit.
FWIW, I have over 90 sites with Adsense, and none of them lost their Google traffic in the Bourbon update - and most of them have 3 AdSense units, an Adlinks unit *and* a Google AdSense For Search box - so that's a heck of a lot of Google code that should have tanked these pages, according to your theory.
I realize that Bourbon has frustrated a number of people, but I think that your focus should be on improving your sites' signals of quality rather than chasing improbable "bugs" and bogeymen.
Put yourself in Google's shoes and try to think what signals of quality that most "good results" have in common, and how they could be automatically detected by a bot.
Then make a list of signals of low-quality that most "bad results" have in common, and think how they could be automatically detected by a bot.
Apply the list to your site(s) and work on increasing the former and decreasing the latter.
Google is not out to "get" you. They're just trying to improve their search quality. If your site got burned in this (or any other) update and you have a site that should be in the "good results" column, you just need to do your best to make Google's job as easy as possible... send them all the right signals and avoid the bad ones.
E.g.
filetype:js "var test"
Even though this is true, search engines does not seem to be indexing Google AdSense code for some reason. It would be logical to index it since it is indeed visible content; however the most important search engines seem to filter this out.
If someone were willing to experiment, try inserting a block of JS that was identical to an adsense ad unit, but with the src="h*p://pagead2.googlesyndication.com/pagead/show_ads.js" modified to point elsewhere (say localhost). Whether or not that violates T&Cs, I don't exactly know--it's deliberately modified to not show an ad (all it does is set the variables and occupy space, so how would it even be noticed, one wonders). If there's an effect as JH is suggesting, then this should demonstrate it.
I beg to differ. Google and other search engines have certainly made efforts to follow links within JS - there's no doubt - but they do not crawl the ads themselves.
If you follow the [pagead2.googlesyndication.com...] link with JS disabled, you'll see that the URL does not actually contain advertiser content.
At any rate, if you check the robots.txt [pagead2.googlesyndication.com...] you'll see that Google has forbidden all bots from the domain anyhow - so neither Google nor any of the legitimate SE bots (which obey robots.txt) will crawl it.
Finally, get that Adsense code out of the top of your html. It is a silent traffic killer.
As others have noted, there are many high ranking, high traffic sites with Adsense at the top of the page, so this statement is clearly not accurate.
ATTENTION ALL PUBLISHERS!
Move your AdSense code to the bottom right of all of your pages Immediately!**
*
*
*
*
*
*
*
*
**So I can get all of the high-paying clicks you'll lose
Create a directory in your site, like /JS/, then via robots disallow it. Now, put your adsense code in a JS script file and put it there, same for other adsense code like feeds, etc.
Will that protect the JS code from being read?
I've thought of that but my understanding is that it's against the TOS because you will have to alter the code.
