June 2010 Tag Rankings

[ipopbb]

Also, I don't see blockquote in your data above. Last month, it was pretty high. How did it shake out this month?

Just Happens. Most of the tags are pretty fluid. Used to change a lot slower before the Caffeine/MayDay updates (2-3 months)... Whatever they are doing they do it faster now... You'd need you're own realtime system in place to play this game now... significant changes happen every 10-15 days.

Fortunately my source is available in another thread. You just have to edit it for your secret word and consider getting the word as a domain, but you don't have to for just the tags... its the url tests that need it.

 

[ipopbb]

Ipopbb,
I've been following your threads with great interest. I hope others realize what you've provided here and take some action. (not my competitors of course). If nothing else, do some testing. Which I will be doing soon, and would be happy to share my data.

I have a quick question. I have an ecommerce store with 35K products. Should I ignore the page rank leak from having 35K outbound links,and actually link to each of the manufacturers websites from each product page? I've done some linking to places like wikipedia from some of my major keyword pages. I'm just a little nervous about linking out to widgetmanufacturer.com from each product page when i sell widgets. Obviously, I'm trying to build pages that are the best for my customers, not google, but it's hard to ignore the math. Your thoughts?

Niksa

WOW! 35K SKUs... affiliate, dropship or do you fulfill orders?

First... relevant does not mean external... they can be your own pages that are tuned for the topic.

Too much payload for me to gamble with... I'm trying changes on my current product pages next week... PM me in 6 weeks and remind me to share how its working out. Without solid numbers, I'm not touching your request with a ten foot poll...

I will give you some pointers to look into between now and then:

Compare your search engine traffic from March 2010 and May 2010 notice anything shocking? If so then read the threads on web master world (We are not allowed to link to other forums here, sorry) pertaining to Google Caffeine and Google MayDay... you are not alone.

Canonical Tags! Know them. love them! Use them on every page. Even if you don't think you have dupes. Trust me you do.

Make sure you have robots.txt, apple-touch-icon, favicon, proper doctype and content encoding.

Compare your content to the content that beats you in the serps... anything obvious jumping out at you like amount, or number of exact matches, more or less focus on a particular phrase... etc...

Do they have dmoz and yahoo links and you don't? (this is the hardest nut to crack)

Is your splash page less than PR3? If so link build.

Usually a retail site pops to life when you get those things sorted out and they are low cost and low risk.

 

[niksanyl]

We fulfill some, but mostly dropship. I've been doing more internal linking, and this should help some here.

For your keyword, type this command in google. . .
site:yourdomain.com keyword

This will tell you what pages on your site are the most relevant to google for your keyword. build a few contextual links from these pages, to the page you want to rank.

I've been doing this with much success.

I'll take a look at my rankings over the last couple of months, but our traffic is actually up over that time period.

I'll take a look at the canonical tags. I've tried to do a good job eliminating the duplicate content by writing unique title tags and meta descriptions. I have a couple of tools to help me find the possible duplicates. But like a lot of other things, this often gets overlooked. I need to eliminate more duplicate content which should help get more pages indexed.

I do have a good robots.txt, favicon, doctype and content encoding, and all pages validate xhtml strict.

12 year old domain and I do have both yahoo and dmoz listings, so no problems there.

I think one of our biggest problems is link building. It needs to be a much bigger focus, and has been over the last few months. Home page is a pr3. was 5, then 4, now 3.

Thank you for taking the time to answer. You've been a big help reminding me of some of the things I need to look at, that when you stare at them all day long, you tend to forget about. I'll look this over again when i have a few more minutes. I'm sure i'll want to pick your brain a bit more.

Thanks Again!
Niksa

 

[albertb]

Have you tried testing with different data sets of different lengths? For example google might expect data within a H1 tag to be significantly shorter than data within a p tag

 

[ipopbb]

Have you tried testing with different data sets of different lengths? For example google might expect data within a H1 tag to be significantly shorter than data within a p tag

Interesting. I have not tested that.

I want to test next if it behaves like a currency...been wanting to do that for a long time now... if only there were 86 hours in a day. What I mean by currency is:
Does <b>keyword</b><b>keyword</b> rank better than <u>keyword keyword</u> rank better than <b>keyword keyword</b> and stuff like that... thats a game changer in content tuning if it does.

 

[MisterGemini]

Interesting. I have not tested that.

I want to test next if it behaves like a currency...been wanting to do that for a long time now... if only there were 86 hours in a day. What I mean by currency is:
Does <b>keyword</b><b>keyword</b> rank better than <u>keyword keyword</u> rank better than <b>keyword keyword</b> and stuff like that... thats a game changer in content tuning if it does.

And what about <b><u>keyword</u></b>

 

[ipopbb]

And what about <b><u>keyword</u></b>

Exactly! You read my mind!

 

[digitalworldz]

This is good information. Never seens this from this side

 

[MisterGemini]

Exactly! You read my mind!

I can't remember where I picked this up. Somewhere I once learned that if you use <ul> in combination with H tags, it was the most heavily weighted keyword terms you could generate. I remember this from a few years ago mind you. They pointed out something about university papers and the like, and the organization of data as examples of how Google prefers to organize data, and the more in line you are with it, the better they rank your data.

So the question is, can we create a 'super booster' out of a certain combination of tags? You have been using your experiment to rank certain tags according to their individual elements. Suppose you put them together... take for example something like this:

<h1><ul><li><b><u>Primary Keyword Term</u></b></li></ul></h1>

Do you think the various weight scores would combine, or could they even bring an exponential weight factor to the page?

 

[ipopbb]

I can't remember where I picked this up. Somewhere I once learned that if you use <ul> in combination with H tags, it was the most heavily weighted keyword terms you could generate. I remember this from a few years ago mind you. They pointed out something about university papers and the like, and the organization of data as examples of how Google prefers to organize data, and the more in line you are with it, the better they rank your data.

So the question is, can we create a 'super booster' out of a certain combination of tags? You have been using your experiment to rank certain tags according to their individual elements. Suppose you put them together... take for example something like this:

<h1><ul><li><b><u>Primary Keyword Term</u></b></li></ul></h1>

Do you think the various weight scores would combine, or could they even bring an exponential weight factor to the page?

Its a good hypothesis... totally testable. I have not conducted a test like that.
If you post a list of what you would consider some good tests I'll add them to my test cases and we'll see where they fall in the sort order.

 

[bowcot84]

Given Google's apparent interest in categorising sites now (my interest is getting mine in the shopping category), would it be valuable to see the effect of various terms in getting into each category.

For example, does the use of terms like "item description", "price", "product name" make is more likely to appear as a shopping site. How about "powered by", "updated" meaning it ranks better as a blog. And so on.

I would have thought given your role, testing methods and resources, this might be a useful and feasible experiments. If you care to share, so much the better

While what you've done is fascinating and I've taken some lessons away, surely content visible to vistors is going to have more of an impact in determining SERPs in G's eyes?

 

[MisterGemini]

Given Google's apparent interest in categorising sites now (my interest is getting mine in the shopping category), would it be valuable to see the effect of various terms in getting into each category.

For example, does the use of terms like "item description", "price", "product name" make is more likely to appear as a shopping site. How about "powered by", "updated" meaning it ranks better as a blog. And so on.

I would have thought given your role, testing methods and resources, this might be a useful and feasible experiments. If you care to share, so much the better

While what you've done is fascinating and I've taken some lessons away, surely content visible to vistors is going to have more of an impact in determining SERPs in G's eyes?

I second, third, fourth, and fifth that! Totally for this kind of testing, though it would be quite a task to come up with them all. Some of the ones you mentioned are a few examples, but I am sure there are more.

 

[ipopbb]

Given Google's apparent interest in categorising sites now (my interest is getting mine in the shopping category), would it be valuable to see the effect of various terms in getting into each category.

For example, does the use of terms like "item description", "price", "product name" make is more likely to appear as a shopping site. How about "powered by", "updated" meaning it ranks better as a blog. And so on.

I would have thought given your role, testing methods and resources, this might be a useful and feasible experiments. If you care to share, so much the better

While what you've done is fascinating and I've taken some lessons away, surely content visible to vistors is going to have more of an impact in determining SERPs in G's eyes?

I do this in the sense of keyword densities of my competitors... Amazon, Target, Walmart, etc... but for their long tail pages. Those went away just like mine did for long tail search. I think part of googles change is stop giving online retail so much FREE traffic.

I believe this is the case because Caffeine/Mayday had 2 effects on white hat retailers.

A. Spend more to compete in adwords
B. sign up for google checkout for page 1 placement from product feeds.

Retail is being pushed into "paid traffic" channels. Lots of retail home pages still, but the product detail pages are vastly reduced in the SERPs now.

I'd say the real question isn't what words get us in... it's "what triggers are pushing us out?"

 

[Abstroose]

So let me get this straight.. according to your results and tests, if I link to a relevant pages, my ranking should improve?

 

[ipopbb]

So let me get this straight.. according to your results and tests, if I link to a relevant pages, my ranking should improve?

Yes... If the pages you link to are relevant to the keywords you tune for... They can even be other pages on your site... They don't have to be external. The test case is all internal pages.

What I don't know is if link internally or externally makes a notable difference... I don't have a test for that yet.

 

[iglow]

love your topics

 

[unzero]

any updates ?

 

[albertb]

All the pages had just one keyword so thats basically duplicate content. So this means that google does not regard content as duplicate if you change the HTML footprint. Or did any of the pages appear in the supplemental index?

 

[MisterGemini]

the supplemental index?

I haven't heard this term in years. We're talk'en old school here.

 

[caneman]

ipopbb, thanks for the info... i studied all your seo threads and studied your charts and tag rankings and it brought my site from 93 to 6 in 4 days! This site was bouncing around the first page between 8 and 12 for several months, then in June G put it at 70+... this is what I did:

-tried to get 50 keyword matches on landing page
-tuned keyword density to 3%
-added 10 html comments with keyword in the html code
-added a "resource" page and simply made outlinks to the top 30 sites in the SERP for my keyword
-took the first 5 tags I could practically add from your tops tags list and added them to my landing page
-reworked all the metrics I could based on your charts for page bytes, title bytes, meta desc bytes, keyword bytes

I did this 4 days ago and today this site is back on page 1 sitting at #6.

This stuff works.... this first site was just a test, now I am going to do it for all my sites...

Here is my theory: Obviously G wants to rid the SERPS of MFA spam... they know most MFA spammers will not use these obscure tags and give outlink love but many "quality" sites will... so G will filter you out of the SERPS if you are not using these obscure tags and not outlinking enough...