H1 text and overusuage

How many times to use H1 on a single page

[trillianjedi]

I've decided to go all out for the hell of it on a small site just as a means of experimentation with some SEO techniques.

If I want to use H1 for a news "heading", which means there are likely to be maybe 10-12 H1 headings on a single page, is that likely to cause any problems?

Or is it something that should be used once, then H2 and H3 for the subs?

Thanks,

TJ

[killroy]

yeah, do it like this:

<h1>News</h1>
<h2>Moose trashes 14-wheeler</h2>
<h2>Mongoose invade factory</h2>
<h2>Grizzly seen picking flowers</h2>

That's how I do it on a page with directory listings. The headigns of each item are h2s, the body are Ps. The H2s are links and the H1 on the target page match the H2 on the soruce page.

SN

[trillianjedi]

Thanks Killroy.

Out of interest, looking at what you have there, the word "News" is not the kind of keyword you'd be picking, so having that in H1 doesn't actually help you very much (unless you're the webmaster for CNN!).

When I say "going all out" - I mean it. I want to experiment with something for the basis of an alternative CMS system.

So, knowing the weight that google, and other engines, place on H1, it makes more sense to me to do this:-

Main heading : News <--- straight user style, no H tag

<h1>Moose trashes 14-wheeler</h1>
<h1>Mongoose invade factory</h1>
<h1>Grizzly seen picking flowers</h1>

Unless google, or the other engines actually *penalise* or somehow dilute the weight of H1 by it's use multiple times on a single page.

Any thoughts?

TJ

[heini]

TJ, my understanding is based on the assumption that weighting of emphasizing markup like <h>, <b>, <strong> etc is relative.

Coming from that I assume with each additional <h1> you basically devaluate your previous <h1>.

Or, to put it another way: If you need an additional <h1>, you need an additional page.

[edited by: heini at 10:53 am (utc) on Sep. 30, 2003]

[Nick_W]

And on top of that, it's just the wrong way to go about it markupwise. I know this is not you issue though so here's what I'd do to:

<h1>Page Title Here</h1>

<h2><a href="#">Story Title Here</a></h2>
<p>blah blah blah...</p>

<h2><a href="#">Story TItle 2 Here</a></h2>
<p>blah blah blan...</p>

Each linked h2 goes to a seperate page with the h2 text as an h1 on that page.

Nick

[trillianjedi]

Heini,

Those were partly my thoughts.

So the $1m question is - "What is the dilution ratio?"

In other words, could I have up to 5 <h1> tags and each one still says to google "this is a major heading of more significance to an H2" or is just 2 <h1> tags enough to dilute it to the next level of heading type.

<h2> presumably, using that model, would have a similar, although less drastic, dilution.

TJ

[trillianjedi]

Nick,

I agree that makes much more sense and is the way I'd do it normally.

I'm talking about all-out SEO for SEO sake here, and this is partly out of curiousity - not to say I'm actually going to go and do it.

TJ

[thewebboy]

I would say have 1 <h1> tag per page, this <h1> tag should be what the page is about.

[trillianjedi]

I would say have 1 <h1> tag per page, this <h1> tag should be what the page is about.

That's the traditional approach (and what I actually do) and the right way to do things in "mark-up" terms, but that's not the question.

The question relates to 100% undiluted SEO. H1 tags for SEO sake only.

Humour me, I'm curious. Has anyone experimented with this?

TJ

[Macguru]

Hi trillianjedi,

I did some little tests about this on sites that are quite insulated from external factors in quite static niche markets. I try to keep all <H*> combined under 15 % of total weight, and over 35 % keyword density in them. Fluff words where added after keywords not to hit proeminance too much. I repeately noticed some drop in rankings on both Google and Inktomi when the H tag was diluted. Unfortunatly, those sites are not listed in AllTheweb, Altavista and Teoma.

[trillianjedi]

Thanks Macguru - that's the kind of info I was after.

Our of interest, you noticed a drop in the rankings in google versus what?

TJ

[shasan]

kilroy,
would it have the same effect if I used h3 or h4 tags instead of h2 and have the target have h1? (h2's are too big for the list page).

[Macguru]

>>noticed a drop in the rankings in google versus what?

Ok, I will spill a couple of beans here. ;)
Since external factors are of some influence on Google, I think a good analysis is better take place in some kind of petri dish. So, I use typos from foreign language keywords from a niche market : Lets say "widgette". For such queries Google shows less than 1000 results. This way, I can test on internal factors in quiet ereas well away from all this competitive turmoils.

For testing puposes, I pop 3 pages for each of those 5 keywords : widgette, widgetta, widgetti, widgetto and widgettu. Each of those pages range from 30 to 300 words of original content, but the same structure. This experiment was pretty rough. I used only H1 and H2 for heading, <P>,<B> and <I> for text. I did not make any distinction between H1 and H2 weights. I just considered both of them the same zone.

Here is the original structure :

<H1> widgette, lorem ipsum</H1>
<P>Lorem ipsum <B>widgette</B> bla bla widgette bla bla. bla bla.</P>
<H2> widgette, Proin ullamcorper lobortis pede.</H2>
<P>Lorem ipsum bla bla widgette bla bla.</P>
<P>Lorem ipsum bla bla widgette bla bla <I>widgette</I>.</P>

I gradually borrowed content from <p> to put it in <H*> on all 15 pages to increase total weight of heading zone. I maintained keyword weight to same in both <P> and <H*>
waited a month,
noted the results,
reverted back to original,
waited another month,
noted the results again to see if everything reverted.

I noticed a drop past 15 % of <H> content on all pages. The heavier the pages, the harder the drop! ;) Longuer pages seemed more affected.

There are probably a lack of scientific riguor in this test since the sampling was done on small amount of test sites. But it's good enough for me.

[petersmall]

Reading this thread has me worried.

Having designed my Web site prior to any knowledge of SEO I've used Header tags simply as containers for CSS instructions. In other words, I might use an <h6> to define a main header and an <h1> tag to define a minor heading or a color change.

Is this ruining my page rankings? Should I now go through my site and sort out which are the main headings and make sure they have <h1> tags and similarly assign all other headings to appropriate header tag numbers? Or, doesn't this make a great deal of difference?

[trillianjedi]

@MacGuru:-

Many thanks for the info - that's the kind of thing I needed. I'll digest that over the next couple of days and experiment with the donor site.

@PeterSmall:-

H tags shouldn't really be misused (although I'm trying to do just that!) - I would suggest you follow the W3C standard protocols for header tags to be safe.

Certainly H1, H2 and H3 are of importance in google, so could be affecting you in the SERPS as far as these on page factors can.

TJ

[lorax]

Macguru,
Interesting bench testing. Did you try modifying the parameters of an existing page to see if the rankings changed much. Say for instance one of the those heavier pages - trimming it down a bit?

[wkitty42]

petersmall,

yes, i would go and adjust those... think of the Hx's as outline layers...

ie:

h1 - cars
.h2 - washing
..h3 - types of soap
..h3 - types of applicators
...h4 - cloth applicators
...h4 - brush applicators
.h2 - tires
..h3 - tire ratings
.h2 - maintenance
..h3 - under the hood
...h4 - types of oil
....h5 - traditional oils
....h5 - synthetic oils

[ugh, had to use the dots to prevent the spaces from filtering out ;(]

does that make sense?

[shasan]

so what if I skip a tag like:

H1

.....H3
.......H4

Any adverse effects? Can't be that bad can it?

[wkitty42]

shasan,

in the "think of hx's like an outline of your page" mode of thought, yes, you've skipped something... whether its bad or not is another thing altogether...

g1smd is fond of posting the following...

Use the heading tags only on text that is a heading.

Run the code through [validator.w3.org...] but make sure you tick the boxes for "Show Source" and "Verbose Output", but especially tick the box for "Show Outline" as well.

On the results page, scroll down through the error list (if there is one) and look for the section marked "Document Outline". If the list there does not look like a summary of your document then you are abusing the tag.

this thread [webmasterworld.com] is one of several (13 according to google) found on this site with the following google request "site:www.webmasterworld.com abusing h tags" (sans the quotes)...

[Macguru]

>>Did you try modifying the parameters of an existing page to see if the rankings changed much.

Hi lorax,

Unfortunatly, not yet. But since lighter pages had lesser drop, I suspected word count in H tags was used somwere. I played with other stuff without checking if reducing page size on same page as an entity had any effect.

Those tests took 8 months, and are almost a year old. A few things may have changed since.

[lorax]

Macguru,
I was just curious because if you reduced the volume of words and focused on kw sets that matched your <h> tags then I would expect to see a significant increase in your SERP placement. I'd be curious to see just how far one could go with word reduction before it becomes a kw list.

[Thump]

Pardon me, but I'm new here and have looked in the glossary, and do not understand "seo" . I am only learning the ropes and tend to use H1 more than once per page at times. Can anyone point me in the right direction on learning about ranking, etc.

Thanks, also, this site is huge, getting lost is easy

Thump

[petersmall]

Thump
SEO = Search engine optimization

[petersmall]

To correct my misused header tags, I used search and replace across the site. This is easy to do if you have all the header tags defined in a common CSS document.