Possible Keyword Level Penalty?

[SEOPanda]

Long time listener, first time caller here :)

We have an ecommerce site that's a PR7 and about a decade old. It's very strong in it's niche, and very dominant in the rankings for just about all of it's main keywords. We pride ourselves on having a clean white hat SEO approach.

We launched a new product line a few years ago and have been working on having the main term rank. The ranking page is a PR5, has a good backlink profile, optimized just as well as the rest of the site, but it simply doesn't rank for it's intended keyword.

The keyword is extremely competitive, but we currently rank anywhere on average page 15 for this term. For a site of our authority, a page 15 ranking just doesn't make sense.

We've been doing optimization on this for the past year, and are not seeing any movement besides a few pages up and then back down... total see-saw effect. We have a backlink profile that ranges from edu sites to military sites to this page, but that doesn't seem to have made a difference.

Our site was not affected in any version of Panda.

I was thinking there might be some kind of keyword level penalty for our site. The ranking page does well for long tail, but simply doesn't for it's main term.

Does anyone have any suggestions? I've been pulling my hair out trying to diagnose this, but keep on coming up short.

Our brand + keyword is #1 BTW (if that makes a difference).

Thanks in advance for your help on my first post! :)

[goodroi]

Welcome!

Have you made sure you have not over optimized the page and its anchor text?

Have you analyzed the page text to see how it compares to the competition? Sometimes Google will not rank a page if it does not include or does include certain keywords.

[Marketing Guy]

What is the context of the EDU and military links? Why do they link to your ecom?

[suggy]

I have a similar situation on some keywords; won't rank for widgets (after a long time trying) and outranked by minnows.

Can I ask if the keyword is a bit of a digression/ tangent from the rest of your site?

Also, could it be over SEO'd? Did you become a little over zealous in repeating your winning formula?

[edited by: goodroi at 5:59 pm (utc) on Dec 16, 2011]
[edit reason] Widgetized [/edit]

[santapaws]

is it a bigger money term that your other terms?

[Planet13]

What is the context of the EDU and military links? Why do they link to your ecom?

I second the question; Do you sell educational or military supplies?

I know I am in the minority when it comes to link development, but I believe that you can easily have too MUCH diversity in your backlinks, and it is just as bad as having too little diversity.

Over-diversity of backlinks would probably be easier for google to spot than for them to identify individual paid links.

[netmeg]

I would probably be doing some stealth spying (er... competitive research) on the people who are ranked #1-10 for what you want. If you haven't already.

[Planet13]

One other thing:

I am assuming that you checked, double checked, and triple-checked for any technical errors, right?

[SEOPanda]

Good morning everyone! Thanks for your replies. Let me answer each one:

Have you made sure you have not over optimized the page and its anchor text?

The page is optimized a bit much (about 25 repetitions vs 10-15), but I can't change it more than I already have because of other departments in the company. However, it's over optimization is very slight, and some other pages on the site are optimized more for other terms, and hold #1 spots. Even then, it would still rank better than page 15+.

Have you analyzed the page text to see how it compares to the competition? Sometimes Google will not rank a page if it does not include or does include certain keywords.

The page ranks well for long tail and brand + keyword, but not for the main term. What exactly do you mean by analyzed the page text? I think I know what you mean, but double checking.

What is the context of the EDU and military links? Why do they link to your ecom?

They link to us for various reasons, much of it because we offer products that they use.

Can I ask if the keyword is a bit of a digression/ tangent from the rest of your site?

It is a bit of a tangent from the rest of our site, even though it still relates. An example would be if we sold nothing but TVs for 7 years, then for the last 3 also sold DVD players. We'd have a hard time ranking for the top level DVD keywords. Does that make sense?

is it a bigger money term that your other terms?

Not bigger, but it is a high volume term that can bring in significant revenue if we can rank for it.

I know I am in the minority when it comes to link development, but I believe that you can easily have too MUCH diversity in your backlinks, and it is just as bad as having too little diversity.

They are all related though. We sell a very wide range of products, so we get relevant links from many sources, including media.

I am assuming that you checked, double checked, and triple-checked for any technical errors, right?

There are none... the site is solid, and everything ranks well except for this particular term.

Thanks so much for your help guys... any other ideas?

[deadsea]

I've been in a similar situation where we added huge new sections to our PR 7 site, put tons of resources behind it, targeted an entire new set of keywords and found that we just can't rank for them despite having better offerings than the established competition. The new section was a logical extension to our business. A related area that had historically been handled by specialty sites.

I believe that Google classifies sites into niches. This makes it easy for the site to rank for keywords in the niche, but harder to rank outside the niche. I believe they populate these niches based on categories of keywords that people spend adwords money on and software they built that parses the web and tries to make some sort of semantic sense out of it.

I'm not sure if there is a human involved in the site classification (it wouldn't totally surprise me) but my guess is it is all algorithmic. The good news is that the new section of our site grew over time. The bad news is that the time frame has been 2 years and we are still playing catch up with the competition, but at least our site is on page one now for almost all the terms only slightly behind the established competition.

Like you we put plenty of internal links into the section and did lots of link building. It was a wakeup call to me that pagerank and on-site factors are no longer sufficient. Niches and a history of positive user experience within the niche are required for competitive rankings.

[Planet13]

Don't take this as a personal attack, deadsea. This is directed at google:

...and a history of positive user experience within the niche are required for competitive rankings.

I have to ask; how is google determining the level of positive user experience?

I got a post card in the mail the other day from a customer who LOVED the items in their order. How is google going to measure that?

If a customer likes something, they might rate it online somewhere. If they like it a lot, they might write you an email saying how much they like it. But if they LOVE it, then they are going to send a post card / letter to you about it.

All I am saying is that the "positive user experience" metrics that google is / will be using can easily be gamed. It's easy to buy FaceBook likes and positive reviews on some of the different review sites.

[deadsea]

I believe that Google's user experience metrics are mostly based on the "long click". If users are happy with the search result, they don't come back and click on something else and they refine their search. Based on this particular experience, I suspect that you need to build a long click history in a niche to be able to rank for fat middle and head terms.

Google could do a lot more to measure user experience better. I have a five year old site that is now competing very favorably with large established sites with huge backlink profiles that were launched in 1995. I know from usability studies and from emails from our users that the site is better than the competition. However, Google doesn't see that. In the last two years we have made huge gains into fat middle rankings, but the head terms all go to the huge old established sites still. Without improvements from the competition, it should be a matter of time before Google is able to measure our site as the better one. I just wish they could do it more directly and faster.

[bumpski]

You didn't mention Google's Webmaster Tools assessment of your site, keywords, meta-descriptions, performance, titles, duplicate content, etc.

Anything there?

[Yellow_Sun]

Another thought,

When you created the new page did you optimize/over-optimize and build the back-links too quickly? Seems G usually likes to see things set in place on a consistent basis over time and not all at once.

Example: Preferably getting 20 back-links over 20 weeks, rather than 20 back-links over 20 days.

Hope that makes sense.

[Planet13]

@ deadsea:

I suspect that you need to build a long click history in a niche to be able to rank for fat middle and head terms.

Thanks for the explanation. That helps.

so basically what you are saying is the more "sticky" a page is, the better, right?

[deadsea]

I don't like the term "sticky". My biggest site is anything but sticky. It has 1.2 page views per session average, and users spend less than 30 seconds on the site on average. But it satisfies them with the answer the wanted. For your site, sticky might be a good indication of user satisfaction, but that certainly isn't true for my site. I rely on other metrics.