Does anyone know if css plays a role in seo? When I redifine my <H1> tags in a stylesheet does that play a role in the ranking? Any info or experience is welcomed and appreciated.
Thanks

Are you asking if styling your <h1>'s with css will negatively affect you as far as seo is concerned?

The answer is definitely not. The search engines will still see an <h1> as a top level heading and treat it the same. I think the only thing the search engines will frown upon is you are using css to try to hide text so as to present keywords to the robots and and hide them from visitors.

If anything I think using css to drive a site design will produce less code and make it easier for search engines to spider your site.

Yes, CSS can have an impact on SEO.

Search engines often pay attention to H1 tags, as you mentioned. H1 tags generally have slightly more weight than normal text with SEs, so if you can put your keywords in a main title, do.

Another example is bold text. Font size also has an impact - if you have size 1 text, it will be ignored by modern search engines, which is the same for same colour as background or "white-text".

I believe you can define your H1 tags to be pretty much the same as any other text - it does not have to be the default size, which is bold and huge!

Yes, CSS can have an impact on SEO.

No it doesn't. A header tag is still a header tag no matter what size you make it with or without CSS.
CSS may well have an effect on your marketing efforts by making your site more attractive but it will have zero effect on search engines

SEs don't read CSS, SEs don't render CSS. It wouldn't make a difference to how the page appears (to a SE indexer) if your <hn> sizes were 200em or 2px they are still <hn> tags.

However if you use CSS as a deceptive means of hiding text etc it only takes a look over by a real person to decide if a penalty is in order.
There is unlikely to ever be a fully automatic penalty for hiding text by any means. It is simply way beyond a text based analysing algorithm, there are too many combinations of stacked elements using different backgrounds to make it possible.
What if a dark image is used as a background with text set to white? The indexer has absolutely no way of knowing what the image colour is, so in this case that would fall into the penalty trap.

The only way that anything like that could be accomplished is to use scanning techniques such as OCR then compare against the source code, even this is not a foolproof or 100% reliable method so even though Google have in the past aquired a company that has this kind of technology (the name escapes me), it is doubtful that it will become an automatic process.
That of course is not to say that it is not currently in use and throws up pages for manual inspection. Because things like this http://www.mattcutts.com/blog/seo-mi...king-your-site appear from time to time.

Originally Posted by chrishirst

No it doesn't. A header tag is still a header tag no matter what size you make it with or without CSS.

Yup, that's what I said:

Originally Posted by Harvey

I believe you can define your H1 tags to be pretty much the same as any other text - it does not have to be the default size, which is bold and huge!

Originally Posted by chrishirst

here is unlikely to ever be a fully automatic penalty for hiding text by any means. It is simply way beyond a text based analysing algorithm, there are too many combinations of stacked elements using different backgrounds to make it possible.

Really? Surely it can't be that hard for a script to notice that the text colour, #FFFFFF, is the same as the background colour, #FFFFFF?

Originally Posted by chrishirst

The only way that anything like that could be accomplished is to use scanning techniques such as OCR then compare against the source code, even this is not a foolproof or 100% reliable method so even though Google have in the past aquired a company that has this kind of technology (the name escapes me)

Interesting. Oh and I believe the one you're referring to is Riya.

Anyway, thanks for correcting my errors

Really? Surely it can't be that hard for a script to notice that the text colour, #FFFFFF, is the same as the background colour, #FFFFFF

You're quite correct, something that simple is easy to spot and action. However, how would something like;

<body style="background:#ffffff;color:#ffffff">
<p style="background:#000000;">A lot of text in here</p>
</body>

be handled? to the simple algo the text is the same colour as the backgound, but we can tell that it actually isn't, because there is an overlayed element with a different background colour. And of course it gets even more complex from there on.

Don't think it was Riya, their technology is more image recognition, the one that Google aquired was OCR, maybe in preparation for the Google Books project.

Anyway, thanks for correcting my errors

Not picking on you personally Just in the all too widespread idea that CSS directly affects SEO.
Good design affects SEO and CSS is certainly part of good design, but pages created using every deprecated tag and out-moded method available will still do well in the SERPs provided they have the right content and links.

I'm with Chris in that I don't think you'll see major seo benefits or penalties from using css. From what I've noticed though sites that use css for their layout instead of tables tend to use less images in general. This seems to be true espeically for text image once developers see they can often create effects using css that they thought needed an image. I see this alot with rollovers for navigation.

Using text instead of images will help you with seo, but that's not really caused by the css, just a byproduct of developers using more css.

I suppose it's possible that using css will leave your site with less overall code and thus increase your keyword density somewhat. That's assuming the code was counted as part of the overall text of the page, which it probably wasn't. Even if it is the benefits are probably minor at best.

Even if css isn't going to give you a lot of seo benefits there are still many other good reasons to be using it.

There is unlikely to ever be a fully automatic penalty for hiding text by any means. It is simply way beyond a text based analysing algorithm, there are too many combinations of stacked elements using different backgrounds to make it possible.

If a browser can render it, why not a search engine? It can theoreticly psudo-render it in memory and then check the colour values of a certain co-ordinate of a pixel (i.e. where a bit of text was rendered) and compare it to the same pixel without the text. If the colour values are the same, or within a certain degree of bad contrast, then it might be an attemt to hide text. Not difficult at all.

(Although it's very unlikely a search engine would do this on every page, because the trade off of processing power verses websites that might actually try and pull it off would be too small.)

Using CSS won't directly benifit your site directly, although it does have benifits - for example, when a search engine is first crawling your site, it only downloads the first few Kb of each page. If that first few Kb is filled with HTML that is just there to style, it'll have less of the 'good stuff' that search engines look for than the same page in CSS (cos the CSS will most likely be in a seperate file, and the markup will be just markup, and not markup + style)

Originally Posted by Minaki

If a browser can render it, why not a search engine?

Absolutely, and something that you can be sure that the SE experts will be working towards as one of their spam fighting tools even if not for general use.
There are quite a few issues to overcome, the main one being is to "see" the intent behind using a particular technique. Text that is hidden may be part of a design using collapsing elements for navigation or "flyover" descriptions etc, to have a 'bot decide this is "spam" would throw out many pages that are being creative or trying to "tease" visitors whilst very likely missing the real spammers.

I have recently gone through this article. As I have seen this thread, I thought to share this. I think you should also visit this article.
http://www.seochat.com/c/a/Search-En...ation-and-CSS/

I read the article. Do spiders really give up if the website is to "hard"?

In a way, yes. If your code is badly formed, it might run into problems parsing it. They can cope with most things though, but do tend to favour structured, well formed code.

There are so many misconceptions and misinformation in that article it is untrue!
just on the basics,
1/ SE spiders do not even read the code. A spider requests the page and streams the output straight into the datacentre it is attached to.
The author seems to have this "cute" idea of "spiders" roaming the internet like ants looking for food. Nope, A "crawler" has the same functionality as a page scraper and simply grabs the binary source. This is then analysed by a separate software agent called the indexer.
So there is no waiting for tables to load and render, no waiting for images to stream out.
Ther are very many good reasons to use standards compliant code, XHTML, CSS, external javascript etc SEO isn't one of them.

chrishirst is spot on

Spiders read a version of the page that you or I would. It has no access to code nor databases, it gets content from a page and deals with that. Nothing more nothing less.

Ibbo