You Might Also Like:
Heading Tags & SEO
It is commonly thought that using keyword targeted headings on a web page is good SEO and can help boost ranking in search engines. Recently there has been some debate as to whether it is strictly necessary to use H1 tags or if using a large, bold font is enough. Rand Fishkin suggests that it is no longer important whilst Matt Cutts seems to indicate H1’s still carry some weight. The impact of this is fairly significant as making changes takes time out of clients budgets.
I decided to test whether H1’s had an impact on a web pages ranking in Google.
10 nonsense keywords (NKW) were selected and checked for existing results in Google. For each of these NKWs 4 files were created with the following HTML:
- <h1 style=”font-size: 20pt; font-family: arial;”>[NKW]</h1>
- <p style=”font-size: 20pt; font-family: arial;”>[NKW]</p>
In the first two examples the headings appear the same visually on the page, with the same amount of code. The only difference is the tag.
Number three is to determine whether the amount of code on the page has an impact on ranking. Number four is a control, which we should expect to rank lowest in all examples.
These files were then placed in directories named “h1styled”, “hnostyl”, “pstyles” and “pnostyl”. The logic for this naming schema was to make reporting easier and the possibility that the length of the URL may influence results (even if it doesn’t it’s easy enough to take into account).
Once uploaded to a directory on the test server named “tagtests” links were placed on the home page of cbutterworth.com pointing to the tagtest directory index page.
Two sets of results were taken. The first about 2 weeks after the initial set up. The second three months later. NB: The lower the number the higher it’s rank in Google.
- Styled H1 Tags – Average Rank: 1.4
- Unstyled H1 Tags – Average Rank: 1.6
- Styled P Tags – Average Rank: 2.7
- Unstyled P Tags – Average Rank: 3.3
The second list of results was less clear cut:
- Styled H1 Tags – Average Rank: 2
- Unstyled H1 Tags – Average Rank: 1.8
- Styled P Tags – Average Rank: 4
- Unstyled P Tags – Average Rank: 2.2
The first set of results would seem to indicate without a doubt that H1 tags have a significant impact on rankings, whilst the use of styled P tags has limited effect. The second set of results still show H1 tags as having a greater impact on ranking than styled P tags, however there is an unusual outcome in the results with styled P tags having a lower average rank than un-styled P tags.
There were a number of potentially confounding variables in this experiment.
- The link structure of the directories may have impacted up the flow of link value to various pages.
- The structure of the directories changed during the experiment, though in a uniform way which should not have impacted on individual results.
- The use of in-line styles rather than CSS placed in the head or an external stylesheet could have impacted on load times, another potential ranking factor.
Whilst this experiment was far from perfect it does seem to indicate that there is some benefit to using h1 tags for SEO. However this may be in small measure.
What Does This Mean For The Humble H1 & SEO?
Where possible using H1 tags is still a good idea. However if it takes longer than an hour to implement H1 Tags on all pages and they already have some large descriptive text on the page then it is probably not worth the time required.
You can provide feedback in the comments below or on my Google Plus Page