Welcome to the SEO Test for Startnipol Pin
If you found this website, you undoubtedly know that Startnipol Pin is a nonsense keyword that we are using for the sake of an SEO experiment. The goal of this experiement is to finally prove whether valid HTML has any impact on a website's final ranking. As such we chose two junk keywords (this website and one other website focus on Startnipol Pin), will create two websites for each junk keyword, try to isolate every other variable that could impact the ranking of these sites, and finally see whether the site with valid HTML ranks higher than the website with invalid HTML.
There are actually four websites that we are using. Two of them focus on the word startnipol pin while the other two focus on a different keyword. We do not want to taint our experiment by including the other keyword at all or linking between the four total sites. The reasoning behind four sites is explained below.
Making Everything Else Constant
In order to isolate the effect of valid HTML on this website's ranking, it is necessary that we make every other aspect of SEO consistent across the competiting websites. The aspects of SEO that we are trying to make constant are:
- Keyword Density - both websites focusing on startnipol pin are identical in their keyword density.
- Page Size - both websites optimized for startnipol pin are exactly the same size.
- Link Source - Both links came from The Lazy Wife using identical anchor text
- Page Content - As mentioned above, the content of both pages are identical in size, content, etc. The only difference between the two pages are minor changes in the HTML to invalidate one of the websites.
Invalidating One Website While Keeping the Other Valid
One of the foremost challenges of this project is making one website with invalid HTML while the other website has valid HTML. The problem that we run into is that by changing the HTML we inevitably would change the page size. Although there is no reason to necessarily believe that small differences in page size plays a role in a website's ranking, we do not want any outside factors affecting our ranking for startnipol pin.
The solution to this was fairly straightforward: change only minor aspects of the page without changing the character length. To do this we made the following changes in the invalid website:
- Doctype Declaration - The correct Doctype Declaration at the beginning of this page should read:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">To invalidate it, we changed it to:<!DOCTYPE html PUBLIC "-//W6C//DTD XHTML 7.0 Strict//EN" "http://www.w2.org/TR/xhtml1/DTD/xhtml1-strict.dta">Although this passes Firefox's Tidy program, the validator at W3C.org freaks out at this change. - Opening Up Div Tags - In a few places we changed a closing div tag to be an invalid tag. Rather than
use
</div>
, we used</dir>
. This not only introduced an invalid tag, but similates the common problem of having an open div tag. - Repeat with Paragraph Tags - We repeated the same process as above with a few paragraph tags.
One Final Variable
Since we are using The Lazy Wife to link over to our competiting websites using the exact same anchor text of Startnipol Pin, it stands to reason that each link is given the same weight. However, one variable that might factor into the overall ranking equation is which link comes first on the linking page. In other words, it is conceivable that Google will look at the first link for Startnipol Pin as being more important, especially since we have two links for Startnipol Pin.
To prevent this from being an issue, we will be duplicating this experiment with another nonsense keyword. Whereas with startnipol pin, we first link to the site with valid HTML and then to the site with invalid HTML, with the other nonsense keyword we will link over to the site with invalid HTML first, and valid HTML second.
This should neutralize any effect that linking order should have on a website's position.
Looking for the First Snapshot
It is our belief that since there is no competition for Startnipol pin and that there will only be one initial linking source to the pages that we created for startnipol pin, that Google will have to fall back on the various on-page ranking factors. Furthermore, since we have made all the on-page factors for startnipol pin constant with the exception of the validity of HTML code, we should get a good indication as to whether valid HTML impacts a website's ranking positively or not.
Obviously this will break down as we publish our reports and new links come into our websites optimized for startnipol pin. The idea here is to see how Google initially ranks these websites and make our determination from there.
Where to See the Results
We will be publishing our results in Site Reference and will be adding a link to each of these pages to the published report once it is out.
Heading Level 3
Many Thanks Go To...
The Lazy Wife - The Lazy Wife generously donated four temporary links to
this project so that we could get these four websites indexed quickly. Many, many thanks to the Lazy Wife.
Gordon Mac - The template you see here was provided for free by Gordon Mac. Its a great template and we were happy to have found it!