Forum Moderators: open
If any on page analysis is, or ever will be, a part of the SE algos, then document structure can only help. Title and H1 should, IMHO be closely aligned, as should internal anchor text. (<a href="/widget-history.html">Widget History</a>, <title>The History of Widgets</title><h1>The History of Widgets</h1><h2>Widgets in the 17th Century</h2>)
WBF
Now on the H1 compared to the title tags. I like to make them slightly different. In the title tags I make sure I have the most important key words while trying to have it describe the page to the searcher. But I like to write article titles that "invite the reader in" which means they may just have a couple of key wordsin them. I make sure I have any other important words in the H2 tags in a little snippit right under the H1.
Back to the importance of H1 tags. I wonder if the search engines are looking more at the size of lettering rather than at the h tags simply because text size is done so many ways now.
hx tags are always styled. Along with this, I have moved more towards having repetitive images (like headers/logos/etc.) in background images.
So, I might have a background image header in a <div id="header"> tag with the logo and site title (e.g. "Wonderful World of Widgets"). Then, in that header an <h1 class="title"> (e.g. "The History of Widgets"), followed by a styled <ul class="menu">, an <h2>"16th Century Widgets", etc.
Using inline styling it is quite possible to have h1 tags lead naturally into a call to action. <h1>Widget Sales </h1><p>are our specialty...</p>
Some would call the above spammy, yet it remains consistant with W3C recommendations.
All that said, I see evidence all around that on page factors just don't matter much anymore. I have literally seen pages with zero text content rank in the top two or three positions, solely on the basis of IBLs. It would appear that far more than anchor text of IBLs are at play, leading to the real possibility that on-page factors are examined more on the linking page than on the destination page.
If that is the case, then it may well be that on page optimization is really optimizing the influence of the outbound links on that page. That, though, could have influence on internal linking.
So much guesswork in this stuff, don't you know.
WBF
I wonder if the search engines are looking more at the size of lettering rather than at the h tags simply because text size is done so many ways now.
I still don't see .css files being spidered except VERY rarely -- so that alone makes it clear to me that size of any lettering is simply not a factor in an algorithm.
I also like the idea of "run-in heads" in some page layouts, and have used them with some success. h1 {display:inline;} is all it takes -- and there is a print history in print for this style of typesetting.
With regards to keywords getting an extra boost for being in an H1 tag -- algorithmic emphasis goes up and down depending on how relevant the results prove be with various scoring approaches. Search engines each have their ways of testing quality, and if one factor seems to distort relevance, then it gets devlaued for now. If future testing shows the relevance signal gets stronger, then an upwards adjustment gets made. So in a very real way, how we use the tag in our mark-up has a direct influence on whether it "helps" or not at any given time.
There was a time, for example, where H1 was as important as the page title with Google. Later on, the H1 was so abused that they couldn't use it as anything more than simple on-page text. And then, after ignoring the tag altogether for many months, it apparently came to be a better "signal of relevance" and the value was increased.
The key is making sure that H1 tags are really used for the main topic of the page, rather than as some magic bullet to rank on this or that keyword. I still see entire pages of text wrapped in H1 tags -- and that's silly. Search engine algos are a lot more sophisticated than that. I'm sure the algos now can detect this kind of tag abuse and toss out any H1 influence the page would otherwise have garnered. Search engines could even flag the domain involved for a hand check when the number of words in an H1 element goes above some threshhold.
The main idea of html is that it is a "mark-up language" -- so the root of a web page is considered to be a document, with mark-up added to describe how the various parts of the document work together. That's a foundation assumption of spidering and indexing. HTML is not a "layout" language.
Some think that search engines are punishing the H1 tag because only SEO-meisters use it.
In fact, I wonder if maybe I should remove my H1 tags from my pages and replace them with P tags. This might help boost my rankings.
How many people actually look at the title element in the web browser? Yes, the title element is what shows on the search results. But also having an h1 element on the page shows that information more prominently to people who are visiting the page. This creates a valuable "location cue" for the end user -- it tells them "yes, you got to the page you thought you would."
In general, I use shorter title elements and then the h1 is a bit more fleshed out and descriptive. For example, a title element would not be as likely to use an adjective or adverb as the h1 header would.
Whether an h1 tag is valued only at the level of regular text, or cranked up a bit at times, I think it's almost always a useful element.
If the document is basically stand-alone, for example Things to See and Do in Geneva, the top-level heading is probably the same as the title.
If it is part of a collection, for example a section on Dogs in a collection of pages about pets, then the top level heading should assume a certain amount of context; just write <h1>Dogs</h1> while the title should work in any context: Dogs - Your Guide to Pets.
WBF
The site has been around since 2001 with the site title in the h1s and it includes the two most important key words for what the whole site is about. Losing those key words could really affect the site it seems to me.
Hmmm, in fact my homepage is #4 in the Google serps and #3 in MS for the one best word I could have on the topic. It just occured to me that having around 300 H1 tag link backs to the home page with this keyword in it just might be making the difference.
I don't want to make a big change unless I'm sure.
Do H2 tags count for anything? ...I don't want to make a big change unless I'm sure.
Leave it as it is, especially if it is working.
Making major changes on a well performing site are scary indeed.
There is nothing wrong with a site title in an h1. A perfectly legit view of the document's structure, if we want to use that perspective. In theory, an h2 would carry only slightly less weight than an h1 in terms of document structure.
WBF