4. Site Autopsy–Checking the Homepage 12


If a site has been over-optimized, chances are you’ll spot it on the homepage.  For that reason, my #2 check is to look at the homepage for problems.  If I suspect there are problems, I might then check another one or two pages on the site.

IMPORTANT: Before making ANY changes to a site you are trying to fix, make a full backup so you can restore/compare versions of pages if things go wrong.

Now, I know that the owner of the site has made changes since the penalty, and some very recent ones at that.  For this reason, I prefer to look at a copy of the homepage at the time of the penalty, or shortly after.  Looking at the data the owner sent me, I can see the penalty occurred around 24th April 2012.  You can see the clear drop off in traffic after that month.

image

 

On 24th April 2012, Matt Cutts announced a new algorithm was being released. It didn’t have an official name, but we know it was Penguin.  Therefore it is clear that the Cayenne pepper site was hit by Penguin, after getting through the initial Panda updates unscathed.  Since Penguin is all about over-optimization and backlinks, we probably need to focus on those for this autopsy.  Let’s start be looking at the homepage as it was back in May 2012.

You can do this by going to the Wayback Machine website.  Most sites will show up using this tool.  The notable exception is a site that uses “noarchive” tags on its pages.

Let’s search for the Cayenne site.

Wayback has records for this site going back to the end of 2007.  I can check out individual records by firstly selecting the year:

image

.. and the clicking on one of the days that has a circle over it:

image

This opens up the homepage, as it was on that date in the browser:

image

This site was ranking for a wide variety of cayenne related words and phrases, but the owner told me that the main keyword for the site was “cayenne pepper”.  His site used to rank #1 or #2 before the penalty, but dropped down to #7 or #8 immediately.  It then continued to drop over the months that followed.  The owner told me:

“I do remember well that by June or July 2012, it was no longer on Google's first page at all for "cayenne pepper" but had completely fell off Google's first page. In fact, it is not even in the top 1,000”.

Today, when I search Google, I see his site at #44 for this phrase.

image

Quite often when new algorithms are released, they are harsh.  Over the weeks that follow, Google adjusts them, and we can often see pages that dropped out, making a return. 

Two other phrases the site use to dominate for are also shown above.  They have dropped a little.

The fact that the “cayenne pepper” rankings seems to have fallen a lot more than the others, and that it was the main phrase for the site, lead me to suspect over-optimization of that phrase.

If I look at the cached page from the Wayback Machine, I can see the phrase is found 54 times according to the SEO Quake chrome extension:.

image

That’s a density of over 6%, which is very high.  I don’t often think in terms of density on my own sites, as I just make sure it is well written for humans and density takes care of itself.  It will be whatever it is after the article is naturally written.  However, if I was to try to put figures on acceptable densities, I’d recommend looking carefully at any 2 or more word phrase with a density of   1% or more.  Single word densities are naturally higher. 

From that screenshot above, you can also see the phrase appears in the Title, Description and Meta Keyword tags.  Here is what the HTML looks like:

image

The title tag is simple “Cayenne Pepper”.  That’s great pre-Panda & Penguin, but not something I would recommend in 2014 as it was a technique used by “spammers” to get a page to rank for a single term.  Google are going to be suspicious of web pages using that technique.  I’d put the phrase cayenne pepper (or perhaps the plural version) in the title, but not on it’s own.    Make the title a little more descriptive, but still keep it concise.  You can add to the description with the meta description tag (which I do recommend you use by the way).

To me, the meta description tag reads as a keyword optimized description.  It has a long list of keywords in there, including cayenne pepper twice, and the keyword phrase “health benefits of cayenne pepper”.

The keyword tag is one I don’t recommend you use, but if you do, only add 4 or 5 related phrases and don’t repeat words.  I am sure Google may use the keyword tag, but not to help rank your pages positively.  I suspect they use this tag to spot spammers.  After all, spammers load the keyword tag with keyword phrases, the more the better, so a spammy keyword tag may raise a red flag.

This keyword tag has:

  • Cayenne Pepper
  • Cayenne Pepper Benefits
  • Health Benefits of Cayenne Pepper
  • Cayenne Pepper Diet
  • Cayenne

Way too spammy for my liking.  The word “cayenne” appears 5 times, and the phrases “cayenne pepper” is there 4 times.  Something like “cayenne pepper, health, diet” would be much more natural.  I am sure a couple of other words like “capsaicin”, “chili” and “capsicum” could also be added.  Doesn’t this look less spammer:

  • cayenne pepper
  • chili
  • capsaicin
  • capsicum
  • health
  • diet

With this revised list of meta keyword tags, you can immediately see what the page is about.

This over-optimization of the title and meta tags appears throughout the penalized site in May 2012.  Here they are for another page called “Cayenne Pepper Hemorrhoid Cure”:

image

Again the title looks like a keyword phrase.  The meta description is heavily optimized for the target phrases:

It include:

  • cayenne pepper twice
  • hemorrhoid cure twice
  • hemorrhoid three times.

The description looks like padding around keyword phrases, and not something a visitor would find particularly useful in describing the page on the site.

One thing I would do if this was my site, would be to rework all titles and meta descriptions.  I’d probably also ditch the meta keywords.  As an extra measure, I’d make sure that my opening header on the page was not the same as the filename for the page (this site was created as HTML, and is not a WordPress site).

image

This is something of a pattern on this site.  The easiest thing to change here would be the opening header, since changing filenames would mean setting up a lot of redirects.  For the headers, I’d add a more “newspaper” style headline.  One that catches the reader’s imagination, and not one that is purely based on the keyword phrase the webmaster wants to rank for.

OK, that’s it for this part of the autopsy.  In the next part I’ll be looking at another issue on site.

If you have any comments, please leave them below.


Leave a comment

Your email address will not be published. Required fields are marked *

12 thoughts on “4. Site Autopsy–Checking the Homepage

  • John.N

    Hi Andy – Actually, I think an inner page ranks at #16 for that main keyword doesn’t it? Out of interest, what software you are using that shows the ranking of those three keyphrases?

    John

    • Andy Williams Post author

      John
      Yes, I got #17 when I initially checked Google. However I use Rank Tracker to check rankings because I can specify that it checks the US Google as if I was searching within the US. When I used a proxy at home, pretending to be in the US, the position was in the 40s. However, all of this highlights how difficult it is to accurately check rankings these days. Are Google trying to make it difficult for us?

  • Kay

    Very helpful article Andy, thanks. Can you explain what prominence means? Is there a %age that you should avoid?

    • Andy Williams Post author

      I am not sure what SEO Quake are doing to get their prominence score. You’d need to check their site. Generally though it means how close it is to the top of a page.

  • Norm

    Hello Dr. Andy,

    Not so many would stick with this article, I think. Why? Because this is really getting way out in the weeds of SEO.

    For the rest of us your insight is pure gold. I’m thinking that somewhere out there in ivory towers there resides teams of Ivy League SEO experts who get paid the big bucks for this sort of analysis for very big advertising firms. They sell dearly, and the info is held close to the vest.

    You are digging down deep into the soil of SEO and giving your findings away for free.

    Just my way of giving you an extra special compliment. You know I love this stuff!

    Norm

  • Howard Lee Harkness

    “Are Google trying to make it difficult for us?”

    Depends on how you define “difficult.” Two things to keep in mind: 1) You are not a Google customer; you are their property, to do with as they choose, and 2) the goal of Google is to own the whole world and rent out pieces of it to the serfs.

    I don’t have any problem with #2 as long as there are at least 10 major players in the arena, but unfortunately, that’s not how the internet works. Everybody knows who #1 is, about 50% know who #2 is. 2% are aware of anything past #2.

    As for the first, Google is basically corrupt (absolute power, and all that). That corruption is easy to verify — Google doesn’t have to follow it’s own rules, and if you are a Really Big Customer, neither do you. The rest of us? Well, we’ve been tricked into providing Google a lot of really important information (Google Analytics) for free, which they can then sell. Like the corner drug dealer, they used to be fairly generous in sharing their stash… But now, all of a sudden, you have to pay.

    So yes, Google wants to make things difficult. Unless you pay.

    I’m inclined to do what the unfairly-penalized Ann Smarty of MyBlogGuest did, and disallow the googlebot for all of my sites just to see if it makes any difference at all. I’m guessing the googlebot will still crawl my sites after being told not to (after all, they *sell* that data), but my site(s) won’t appear anywhere in the search results for any important keyword. Thereby providing addition evidence of corruption at Google. Unless they find a way to disguise the googlebot visits, I will see them. Come to think of it, I wonder if any of the dozen or so bots I see in my logs are alternate names for Google.

    I would venture to guess that won’t change my income at all.

    • Andy Williams Post author

      Hi Howard. Thanks for posting your thoughts, I appreciate the time spent. One thing interests me though. Would you mind explaining why you think Ann was unfairly penalized?

  • Norm

    Hello Friend Andy,

    I’d say things in Cowchip town are mixed right now. I just successfully talked a potential client out of paying me $1200 to build him a website. Two reasons. First, after studying the SEO potential for him actually making any money selling his vastly overpriced fantastical real estate PDF primer, I told him he’d be wasting his dough. He was kinda let down, especially since he didn’t actually ask for my opinion about that. Lost a customer, but maybe gained a friend.

    The other reason was that I’m having all the fun with my new book series coming out sometime in this lifetime. Ha. My poor characters have no clue at what’s coming their way.

    Oh…and I finally got my hands on a Kindle HDX 7 last week. Finally! Now I can read your books like a real pro! I’m still trying to figger out what else it is good for.

    BTW, I did advise my new friend to drop the price and turn his legacy work into a book. See you on our Kindling group until the next issue!

    Your friend and pupil,
    Norm
    Cowchip Alabama

  • David

    Hi Andy,

    Is it necessary to set the proxy settings in Rank Tracker itself or is it sufficient to be connected to a proxy through a VPN such as Hide My Ass?