More Empirical Data From MY SEO Lab

naviivan said:

Hmm, I am not sure about this one, I have heard and read a lot about the difference in weightage of link juice coming from different sections of the page. Now I don't run any tests of my own, but I know the engines can differentiate http://www.seobythesea.com/?p=1093. I have even heard of sites getting rejected from Yahoo due to having too many footer links.

Also, as you know many people create themes for WordPress and insert their money site links in the footer for getting backlinks. So I decided to analyze one of them, it's called www.bromoney.com. Most of the themes they create have links to this particular site and it's inner pages, they target terms like "internet banking" etc. But I never see the site ranking on the first page. I have analyzed the top 10 sites for the keyword term and even though this site seems to have a better backlink portfolio with higher PR, mozRank, domain authority etc. than few of the to the top 10, it does not rank in top 10. And 95% of the backlinks to this site are footer links which were embedded in the themes that they distributed and this leads me to believe they are being treated differently. Now this is only one example, but I have seen this often. Footer links are surely not worthless, just that they may have differences in how they pass link juice (maybe relevancy plays a bigger role etc.), atleast that's what I think.

If I may ask, what setups do you have to measure such a thing and what sort of tests do you run?

Click to expand...

Regarding your example:

http://www.bromoney.com/

cd rates 26 1.75%
cd > 100
rates > 100
checking account 16 1.08%
checking 62
account 72

adding every related phrase IS NOT tuning for a keyword


#1 for checking account
wiseGEEK.com
http://www.wisegeek.com/what-is-a-checking-account.htm

checking account 32
checking 32
account 50

#1 for cd rates
bankaholic.com
http://cdrates.bankaholic.com/

cd rates 70
cd > 100
rates > 100


Your example doesn't stand a chance even if it had the best backlinks in the world. Tuning for everything is the same as tuning for nothing. A handful of LSI keywords with a great deal of density is the formula...not every LSI keyword you can find.

ipopbb said:

I link to a website from the splash page of my PR8 at the top. I link to another one from the splash page footer. They both attain PR3 in the same time for the same keywords without any additional link building. I don't see a difference. Like I said... maybe my tools aren't sensitive enough. There might be value in being at the top of a buzzillion PR0 pages, I haven't tested that. I don't see it in my pages with weight. Footer is my playground, it has always delivered for me.

Click to expand...

If you are just measuring PageRank then of course both those pages will get the same PageRank. I am not doubting it's ability to pass PageRank, I am doubting it's ability in the total value of the link.

Is the keyword that you linked to both of the sites a unique keyword? Have you checked the rankings for that keyword? Which page comes first in the results?

ipopbb said:

Regarding your example:

http://www.bromoney.com/

cd rates 26 1.75%
cd > 100
rates > 100
checking account 16 1.08%
checking 62
account 72

adding every related phrase IS NOT tuning for a keyword


#1 for checking account
wiseGEEK.com
http://www.wisegeek.com/what-is-a-checking-account.htm

checking account 32
checking 32
account 50

#1 for cd rates
bankaholic.com
http://cdrates.bankaholic.com/

cd rates 70
cd > 100
rates > 100


Your example doesn't stand a chance even if it had the best backlinks in the world. Tuning for everything is the same as tuning for nothing. A handful of LSI keywords with a great deal of density is the formula...not every LSI keyword you can find.

Click to expand...

To be fair, you would need to look at http://www.bromoney.com/cd-rates which is more targeted to that keyword and has a higher density. Compare that page, it ranks around 25th for the term.

But what is surprising to me is that you are placing so much weightage on on-page optimization when trying to rank for competitive terms when everyone in the top spots are link building like crazy. Again this is something that I have heard and read at numerous places, but it's interesting to hear a totally opposite view.

naviivan said:

If you are just measuring PageRank then of course both those pages will get the same PageRank. I am not doubting it's ability to pass PageRank, I am doubting it's ability in the total value of the link.

Is the keyword that you linked to both of the sites a unique keyword? Have you checked the rankings for that keyword? Which page comes first in the results?

Click to expand...

Link Juice is about the hypothetical apportioning of PageRank to the page you link to. There is only one real way to measure anything in SEO... and I'll clarify that statement with:

No, it wasn't a unique word by any stretch of the imagination. I rank 5-6 with both most of the year and 2-3 when the words are in season. They occasionally flip. I don't care who is first and who is second. I care about $$$MONEY$$$ , so for all practical purposes they perform the same as each other. You probably could buy lunch with the difference sometimes.

I knew i'd start a fight with that. All I can recommend is that you only believe what you can prove to yourself. There is simply no other way it can work. And letting go of certainty is really hard to do. I'm wrong more often than I'm right. I even question my own beliefs regularly. A couple years ago a believed the same as you on this matter. I just don't anymore. All I am telling you is that I can't prove to myself that where a link is on a page matters in any meaningful way I can measure.

I would imagine that if it did work as you propose then the best exploit in the world is owned by you. Modify your WP themes to put the footer at the top and laugh all the way to the bank. If it works that way you can test it and it will be an easy one way ticket to the top in most cases. You can beat all the other WP link wheel fools. I'm happy when people find ways to make money. If I'm wrong, and I am quite often, then you have the single best undiscovered Google exploit ... hmmm we'll call it "Footer Topping" for lack of a better term. I also believe you could come up with the single best undiscovered Google exploit. I, individually, don't believe page location of links matters in a meaningful way in SEO. If you have doubt and you are invested in link location as a strategy then prove it to yourself. Better safe than sorry. If you have no doubt and you are invested in link location as a strategy then you should take a big step back, because SEO is not and should not be a faith based religion. ( Now look what you made me do! I just went from pissing off 10 people to pissing off 100 )... ok, i'm ready! flame on.

naviivan said:

To be fair, you would need to look at http://www.bromoney.com/cd-rates which is more targeted to that keyword and has a higher density. Compare that page, it ranks around 25th for the term.

But what is surprising to me is that you are placing so much weightage on on-page optimization when trying to rank for competitive terms when everyone in the top spots are link building like crazy. Again this is something that I have heard and read at numerous places, but it's interesting to hear a totally opposite view.

Click to expand...

Sure... I was looking at the page provided and what it was tuned for.

http://www.bromoney.com/cd-rates

cd rates 51

Still needs 40% more occurrences to be competitive just on number of matches.

to be fair to my point... page placement of the backlinks isn't why this site doesn't rank higher.

naviivan said:

But what is surprising to me is that you are placing so much weightage on on-page optimization when trying to rank for competitive terms when everyone in the top spots are link building like crazy. Again this is something that I have heard and read at numerous places, but it's interesting to hear a totally opposite view.

Click to expand...

You might find that desert between PR6 and 7... the one where it takes 1000 times the effort in link building to cross. Or you can tweak your content in many cases in a few hours for the same effect. There are plenty of "link build forever" recommendations. I'd be a voice in the crowd. I look for other ways because at PR6 you either go viral or you cap out and a PR3 with well tuned content can easily beat a PR6 with poorly tuned content. The difference often being... content.

I hold 5-8 links on page 1 for about 40,000 keywords... #1 for thousands of them. The people that beat me in the SERPs aren't doing it with backlinks.

After PR3 if your website isn't making money... your problem isn't backlinks. My advice, look elsewhere. If you are on the bottom half of page 1 trying for the upper half ... tuning content the right way will probably do the trick. You might have to try several times. If you are #2 trying for #1... tie breakers are the hardest in SEO... thats when google's 200 factors come into play... you don't want that... go back to content fundamentals and crush them!

Dude... add a 500 word diatribe in the comments of the cd rates page and mention cd rates 40 times in it. Not 40 related subjects to cd rates, but cd rates specifically. You will bump up and it is so much faster than link building.

Pull a loon comment "why can't these cd rates websites speak in a way I can understand. Why do they provide so few sources for cd rates? If this was my cd rates website you'd see this about cd rates and that about cd rates..."

ipopbb said:

Link Juice is about the hypothetical apportioning of PageRank to the page you link to. There is only one real way to measure anything in SEO... and I'll clarify that statement with:

No, it wasn't a unique word by any stretch of the imagination. I rank 5-6 with both most of the year and 2-3 when the words are in season. They occasionally flip. I don't care who is first and who is second. I care about $$$MONEY$$$ , so for all practical purposes they perform the same as each other. You probably could buy lunch with the difference sometimes.

I knew i'd start a fight with that. All I can recommend is that you only believe what you can prove to yourself. There is simply no other way it can work. And letting go of certainty is really hard to do. I'm wrong more often than I'm right. I even question my own beliefs regularly. A couple years ago a believed the same as you on this matter. I just don't anymore. All I am telling you is that I can't prove to myself that where a link is on a page matters in any meaningful way I can measure.

I would imagine that if it did work as you propose then the best exploit in the world is owned by you. Modify your WP themes to put the footer at the top and laugh all the way to the bank. If it works that way you can test it and it will be an easy one way ticket to the top in most cases. You can beat all the other WP link wheel fools. I'm happy when people find ways to make money. If I'm wrong, and I am quite often, then you have the single best undiscovered Google exploit ... hmmm we'll call it "Footer Topping" for lack of a better term. I also believe you could come up with the single best undiscovered Google exploit. I, individually, don't believe page location of links matters in a meaningful way in SEO. If you have doubt and you are invested in link location as a strategy then prove it to yourself. Better safe than sorry. If you have no doubt and you are invested in link location as a strategy then you should take a big step back, because SEO is not and should not be a faith based religion. ( Now look what you made me do! I just went from pissing off 10 people to pissing off 100 )... ok, i'm ready! flame on.

Click to expand...

Well the problem is that although SEO should not be totally faith based, not everyone has the tools, knowledge, resources and money to test everything out for themselves. So you have to place some trust in others.

You follow the SERP positions for the HTML tags so seriously but not for this experiment?

Now, in my posts I never stated that a link from the top of the page would hold more value, I just said that there is a good chance that different locations have different weightage (they already have the technology to track the location within a page using Vision-based Page Segmentation). Of course I have no idea what the differences would be. There is a chance that an incontent link would have some more value because that is where a real, natural editorial link would be from and those are harder to spam.

ipopbb said:

No, it wasn't a unique word by any stretch of the imagination. I rank 5-6 with both most of the year and 2-3 when the words are in season. They occasionally flip. I don't care who is first and who is second. I care about $$$MONEY$$$ , so for all practical purposes they perform the same as each other. You probably could buy lunch with the difference sometimes.

Click to expand...

Personally, I find that click-through-rate for your keyword is a much stronger factor in SERP ranking than it is stated anywhere, if it's stated at all.

In fact, I would even go as far as to say that if you had a site with NO on-page optimization and NO backlinks, but you had people Google the query and browse to page 5 to click on your site, it would shoot up to page 1 rather quickly, outranking much more authoritative sites!

What's your opinion on that ipopbb?
Also, could you clarify what is "Citations Listed By Google (%)"

naviivan said:

You follow the SERP positions for the HTML tags so seriously but not for this experiment?

Click to expand...

Sigh... at the end of the day it all has to come back to money. It's not a video game. People want money not SERPs and you don't necessarily have to be #1... When I started I did pretty well only striving to be in top 50 (for 100 keywords) at the same time. There are niches where people will dive to page 4 and 5 to find what they are looking for... you have to understand making money to determine how much time is worth investing in a venture... When you make money you have an easier time getting to #1 because the motivation to not quit becomes real... time is the enemy for most people. I follow everything seriously when it comes to my time and my money.

Others are reproducing my experiment with a just a little of their time. The $5/ month web hosting with a free domain is optional, but needed if you want to know where URL components place on the list without extra bias existing aged and SEO'd domains would put on it.

I'm not peddling goods here. I've made my theories, experience, data, and source available for free. If you want to show the world I'm wrong... awesome! Do it with data. You'll be doing me a big favor.

Hi. ipopbb, you got some pretty amazing stuff going on here. I sat up the whole night reading this and other threads by you.

But regarding this thread:

At first I saw the charts and didn't understand them. Closed them, and read some more in this thread...
Then after reading half the thread I had to scrolled up and look at the charts again. Now I wanted to understand them.
Then I thought I had figured things out, but then I see you saying that you only measure the SERPs data... And that threw me off once more.
... but you only measure the SERPs data for some measures not all of them. Some are for on-page as well?
I've read this thread two times now, and opened the the charts several times when they were discussed in a post. And I still have questions as I don't yet understand it as it should be. Hope I'm not the only one? ... and hope you don't mind. Here goes:

This is all based on the data you published in this thread. I know you probably have some new stuff coming soon, but it would be nice to get the measurements you used clarified before that.
I had quoted some of your initial replys but this just "clumsy" up my post, so in stead I just ask about what I didn't get.

"Summary Bytes"
Thought this was the same as your "Meta Description Bytes" meassurement but it's not, is it?
This has to includ everything in the Summary on the SERP regarding the ranking page?
So if I tune my page to be around 47 bytes (the blue average line) in the "meta description bytes" I would get an average "Summary Bytes" at 160 or 170?

"Summary Matches"
This is the result of your search query, if the keyword shows in the summary on the SERP, on a given result?
As if the keyword were to be found two times in the summary on a given result it is displayd as 2 Summary match?
But you can't really controll the summary. It depends if Google chooses to use your meta description as summary or not. So the moral would be "Do your meta description right and this would show up as the summary."

"Keyword Matches"
This is in bytes right?
And this is for the result page that the SERP links to? (An on-page measurement and not a on-SERP-measurement)

"Keyword Density"
This would also be regarding the result page that the SERP links to? (An on-page measurement)

"Images Per Page"
Is this measured in bytes?
So between 80(average)-300(max) bytes of images on a page would be best? (An on-page measurement)

and then the links measurements:

"Links Per Page"
Intuitively I thought this was total links on a page. But if the average is at 400 and max is about 2000 this can't be amounts of links.
Is it total link length measured in bytes? (An on-page measurement)

"External Links Per Page"
Is this also a total link length measured in bytes?
You have the average 140 or max 600 characters.
So links in total should be at least one link, but could be more links, but not exceed the total of 600 characters on one page?
(600 characters could be for instance 4 links with 150 characters each)?

I must have misunderstood something here?

Take care.

Dude... add a 500 word diatribe in the comments of the cd rates page and mention cd rates 40 times in it. Not 40 related subjects to cd rates, but cd rates specifically. You will bump up and it is so much faster than link building.

Pull a loon comment "why can't these cd rates websites speak in a way I can understand. Why do they provide so few sources for cd rates? If this was my cd rates website you'd see this about cd rates and that about cd rates..."

Click to expand...

I agree with what you're saying here ipopbb...

In this situation however, if one was trying to rank for:

cd rates
rates on cd
cd rate
rate cd
rate on cd
rates cds
rates on cds
best cd rates
best rate cd
best rate cds
etc.

Would you be better off
1) trying to stuff one page that is essentially a "cd rate" theme page, with combinations and variations of the above (as well as corresponding anchor text on backlinks)
or
2) making a separate page stuffed with each separate term, and have corresponding anchortext on backlinks be mostly that term (with some variation for spam filters)?

Can you elaborate more on your grid?

You say you have your scrapers running round robin with three second delays... is that 5 requests per IP, then three seconds pause, then next IP?

Basically I'm trying to figure out how to best load balance a long-term Google Scrape campaign.

@ OP: Great top level SEO. Thank you for taking the time to share this with the BHW community. The fact that you back up your claims with data speaks volumes about how serious you are with regards to IM and SEO.

Even more important is that the rules appear to change from page 10-3 and page 2-1... TWO GOOGLE Algorithms. Which do you tune for? I tune long tail pages for page 3 algorithm for best revenue effects and splash pages for page one for traffic gen. I don't know why I'm the only one who sees it. Google's best kept secret is that there isn't 1 algorthm... there are 2!

Click to expand...

Great observation ipopbb - question in regards to your tags order (from another one of your threads) - if there are two algos, wouldn't this shift the accuracy of your data in some respects? (more so for the tags that are ranked 30+)

goawayplease said:

Can you elaborate more on your grid?

You say you have your scrapers running round robin with three second delays... is that 5 requests per IP, then three seconds pause, then next IP?

Basically I'm trying to figure out how to best load balance a long-term Google Scrape campaign.

Click to expand...

One request per IP then sleep 3 seconds. That way it looks like 1 request every 3 seconds in their server logs but is about one request every half second in your app.

Anything faster triggers the google the automated queries Captcha. Slower is safer if you leave it running unsupervised.

Fwiffo said:

Great observation ipopbb - question in regards to your tags order (from another one of your threads) - if there are two algos, wouldn't this shift the accuracy of your data in some respects? (more so for the tags that are ranked 30+)

Click to expand...

Possibly... no one knows right now because I haven't repeated these experiments since the Caffeine/MayDay update... everything is totally different now so this data represents the way things used to be... Hoping to get a little time to repeat the experiment next month. I'll share the results when I do. I know everyone is eagerly awaiting some fresh charts that are applicable to the new Google.

Fwiffo said:

I agree with what you're saying here ipopbb...

In this situation however, if one was trying to rank for:

cd rates
rates on cd
cd rate
rate cd
rate on cd
rates cds
rates on cds
best cd rates
best rate cd
best rate cds
etc.

Would you be better off
1) trying to stuff one page that is essentially a "cd rate" theme page, with combinations and variations of the above (as well as corresponding anchor text on backlinks)
or
2) making a separate page stuffed with each separate term, and have corresponding anchortext on backlinks be mostly that term (with some variation for spam filters)?

Click to expand...

I dont have as much experience with the new Google as I had with the old. From what I am seeing in the serps for you keywords it looks like focusing a page per term is most efficient for ranking with content score. Perhaps combining it with CNAMEs so your site has exact host matches with your domain. Example CNAME Hosts: cdrates.domain.com cdrate.domain.com etc... then if you url rewrite and canonical the pages to urls like
http://cdrates.domain.com/cdrates you'll get two matches per url.


then use full URLs in your HTML source

<script src="http://cdrates.domain.com/cdrates/script/jquery.js"></script>
<link rel="stylesheet" href="http://cdrates.domain.com/cdrates/css/reset.css"/>
<img src="http://cdrates.domain.com/cdrates/images/logo.jpg"/>
<a href=http://cdrates.domain.com/cdrates/contactus">Request a CD Rate Quote</a>

This will up your search term matches per page through the roof...
be sure to stuff the title and alt attributes too and all the other SEO101 stuff!

Don't forget about the legitimate content usages... need lots of those too, but this appears to be the leading content tuning method in a lot of keyword verticals. It gets your overall page matches up and doesn't scare your visitors, AND it is not a violation of the webmaster guidelines. Its neither cloaking nor keyword stuffing. It's merely a fairly common pattern of how to structure a website.






This is part of how SEO.com takes #1 for the search SEO

ipopbb, thought I'd pass on some quick data for you and people following this thread

Just posted 2 posts to a site of mine - used some of the things you've shared in this and the tags threads.

Within 20 minutes they were indexed and ranked

Lightning quick

One post to #1 for 4 different terms each
One post to #3 for 4 different terms each

Low comp, 4 words

no backlinks, no promotion, no twitter, no linking on the main site except for category / tags, and regular WP to pingomatic

Domain is about 4-5 months old

Amazed at the speed of this

Anyways, just wanted to share as I know you like to hear / see data

My first content update this month on this site, most posts usually take 3-5 days to show up in the index, 20 minutes kind of stunned me.

that's some nifty data there ipopbb.
Something that I thought of while reading the thread is to search the google patents for traces of their algo and then do a chart for each one.
I have looked through a couple of patents and its some legal mumbo jumbo for the most part so someone would have to really think about it.
@Fwiffo - are you talking about his advice to do subdomains with filenames as the phrase? and / or doing full url's to the files inside the code?

Fwiffo said:

ipopbb, thought I'd pass on some quick data for you and people following this thread

Just posted 2 posts to a site of mine - used some of the things you've shared in this and the tags threads.

Within 20 minutes they were indexed and ranked

Lightning quick

One post to #1 for 4 different terms each
One post to #3 for 4 different terms each

Click to expand...

The timing is probably just luck and the new Google... it is supposed to be a lot faster at indexing... and as far as I can tell it is.

Thanks for sharing your results!