CSS Destroyed My Rankings

[pebblez]

I had used a table based design for years, and I had good rankings on all of my keywords. Recently i changed over to a pure css design including the use of h1 tags, divs instead of cells, the whole works.

When the site was finished I uploaded it to the web and waited. A week later my rankings started dropping. First they dropped in google. Then they started dropping in yahoo. Msn has been fine, but who cares.

Can anyone tell me what I've done wrong?
The new site can be seen at http://pebblez.com
One page of the old site can be seen at http://pebblez.com/index2.html

I really appreciate any advice you can give me. Thanks

[Dubbya]

Changing your site from table-based layout to pure CSS is a MAJOR change.

Google seems not to appreciate major site-wide changes as they're usually indicative of spam. If your text to code ratio has changed significantly, Google might be worried that your domain has been overtaken by yet another spammy advertiser.

Most often, they'll penalize these sites to discourage this sort of thing, so this might be what you're experiencing. Resubmit your site map and your rankings should pick up after a few weeks.

My site went AWOL for six weeks until I resubmitted my sitemap and sent an email requesting reinclusion in the Google index, so that's something else you can do.

Give it some time. It takes a few weeks to be reincluded, so use that time to fine tune your SEO efforts and get everything ship-shape.

Here's a handy article that might be of some help:
http://www.mattcutts.com/blog/reinclusion-request-howto/

[incrediblehelp]

Besides the table and CSS code nothing else has changed?

1. Content?
2. URLs?
3. New competitors?

Google has been updating (http://www.seobook.com/archives/002070.shtml) in the past few weeks as well.

I have changed plenty of websites over. The only time I really see major drops is when URLs switching is not handled properly.

[incrediblehelp]

As most know here I am not a big fan of text to code being that important in the overall SEO side of things. Here is recent post from WPN's Joe Lewis that discusses it more:

Can Coding Affect Your Search Ranking (http://www.webpronews.com/insiderreports/2007/03/05/how-coding-can-affect-your-search-ranking)

[pebblez]

First thank you both for your help.

Thats whats so weird! I did not change any of the words, or pictures. The URL's are all exactly the same, and its definitely not added competition that is doing this to me, some of my keywords im losing 200 spots. I even added a few pages of content to fill out spots that were lacking

and I think my code to content ratio should be better now...although I could be analyzing it wrong

I just can't see why doing everything right, making the site more accessible, faster loading, and giving it cleaner code, should cause a significant drop in rankings. Stupid me I thought it might even help...a little

[incrediblehelp]

Do you know how much your link totals and quantity have changed in the past 6-12 months?

[Gary Golden]

It has not effected my rankings, checked your site seems alright, noticed that you do not have a site-map, I link to my site map on all pages.

> Noticed that you do link to your main page on all pages.

I have been off google for months now with-in the last several day I am back though I spent the last couple of weeks making sure that every page linked back to my main page and whola, at the top of GOOGLE again though do not know how long that will last and I also use CSS on both of my sites.

[pebblez]

IBL's? I gain about 100 a month these days according to yahoo. As far as my sites linking structure I didn't really change it, although I did make a standard nav bar for every section of my site, which has made things a little more uniform.

As far as a sitemap I have a google site map (am looking for a decent ROR sitemap generator) I also link to my homepage from every page.

I do have a theory and if someone could take a look at this I would really appreciate it. In my header I have an H1 tag around the first text on the page, and its just about the first thing that shows up in the code. At the top of the header div I have a black picture background and at the bottom I have set the background to just be white. The text in the H1 is black and would show up against that black background, but I have added margins to the H1 to move it down.

Could the search engines inability to read the css margins on my H1 cause them to think I'm hiding black text against a black picture in an H1 tag right at the top of my page? Can google even tell the picture is black?

I would really appreciate any feedback on this.

[Seopro]

This took me about 2 minutes to find your problem. Doing your website in CSS is not enough. Go to the W3C and validate your code. You have 36 errors on your front page. Fix those and I bet your rankings will improve
Don
Seopro

[incrediblehelp]

This took me about 2 minutes to find your problem.

Care to share exactly what that is? Simply saying validate code is pretty vague.

[Seopro]

The W3C has an html validator on their site. Bring that up and type in or paste your URL in the top box and it will show you how many errors on the page. I do this with every page I upload. You can also check the code by uploading a file to the validator before it is uploaded.
Don

[incrediblehelp]

Yeah Ii know, but you said you identified their problem. Having non-valid code doesnt stop you from being indexed or being ranked at the top of the SERPs.

[Seopro]

I think all search engines are members of the W3C and I believe they will place a site written in standards above a site that is not with all things being equal. I have never failed to put a site in the top 10 that had valid code. That has just been my observation. On one key phrase affordable seo company oregon I have #1 #2 #3 #4 #5 #10 even though there is hardly any competition. On affordable web design company (yahoo) I have been #1 for 14 months with valid code. And on google I have a #1 with a PR2 ahead of a PR6 because I have valid code and he does not. I have notice this several times with not just myself but with other designers that use web standards.
We all have our own ideas which is great and that is why I joined this forum. I will learn something from others if I take part in the discussions and hopefully I have some knowledge to pass on to others.

[incrediblehelp]

I agree seopro, valid code is better than not. My point was only that you said the reason why pebblez rankings are tanking was because of in-valid code. I dont believe that to be true.

[Seopro]

That brings up another subject that Matt Cutts did not answer in his blog but it is a possibility. Does a search engine skip over some or all invalid code so it is not indexed? My search engine spider did but I do not think that is proof because other times it will index things. Will google's spider skip over id="row" when it shows up more than once? If you use class="row" then it will read it correctly. I don't think there is a way to find that question out other than a wild guess. LOL now you have me thinking and I am not use to doing that.
Don
Seopro

[BigOakSEO]

I just did a check in Google for a term on your site, Glowing Stone Onyx, and you were number one in Google. Maybe during this latest update period you are seeing a datacenter that doesn't have you in mix. I'm on the east coast and your rankings look okay. Was there a particular term you were checking?

[krisidious]

my site was sitting about 57th in the rankings for the term "House Plans" and about 30th for the term "Custom House Plans" I was recently convinced to rebuild by the folks here at WPW...

with some key help from a certain "justtryingtohelp" I switched my site from imagemap navagation to CSS...

I also fixed a lot of small html errors...

now I sit at 47 for the term "House Plans"
and I rank 4th for the term "Custom House Plans"

I would look elsewhere for the loss in rank...

[corvettehunt]

1 - There are many errors that need to be fixed. Use the W3C validator:
http://validator.w3.org/check?uri=http://pebblez.com

You jumped your error count from 7 to 36.

2 - Trim your spaces. You got text that is broken by linebreaks all over the place, that in your original page where ok. I.E:
"Rustic
Stone Home Decor"
Instead of:
"Rustic Stone Home Decor"

Basically you are losing a lot of keywords because of that, for you won't rank anymore for "rustic stone", "rustic stone home", or any derivatives. The same happens all over the place!

---------
Fix those 2 points (which is fairly simple to do BTW), and wait for the next reindexation. You should do a lot better... good luck ;)

[iany]

There has to be something other than changing to tableless design at work here. I have changed plenty sites from table design to CSS over last year and have had no problems with Google rankings. Quite the opposite. Have tended to gain a few places - unless already no 1 :0).

Having had a look at the source code, I am not surprised the html doesn't validate. You need to read up on use of divs in CSS. You cannot have more than one div id of the same name. (div class yes)

Ian

[weslinda]

Maybe Google decided that they didn't like any web sites using Yahoo Hosting. Pointing all over the place in your site to a Yahoo cart could make google mad. If you use Google Checkout, maybe you'll be first for all of your terms.

Okay, all kidding aside. I would recommend a cleanup of code. This is just good design, may or may not help in regards to your rankings.

Second, I would recommend trying to find your own cart for all checkouts. This simply will allow all content and visitors to stay on your site.

Third, I would review matching up your title's to your <h1> to your linking structure, as this should help as well.

Finally, I would look at privacy and security issues. Make sure your site has a privacy policy. P3P if you can put the time into it.

Oops, one more thing, make sure your pages have plenty of content, major changes to a site can spell trouble if there isn't a marked improvement in the site.

[dux]

It looks like you are using div id tags on multiple occassions in the same document, such as <div id="general2of3">

i believe you can only use a div id tag once, whereas you can use div class tags as many times as you like. you need to validate your page using a html validator. the returning document will list the errors you need to fix.

i changed from tables to css and my rankings increased, but the site validated.

[SemAdvance]

Design styling will have no effect on search rankings.

Poor html code however will trip up a search robot faster than any other issue involved with your site.

Instead of focusing on which issue caused the drop look at all issues and fix them.

Poor coding?? then validate, no sitemap? add one, poor interpage linking? fix it...there are over 100 parts to the algorithms not 1.

Whatever you do to improve your site will most often have a resultant improvement to your bottom line.

Also check domain registration dates...This one issue alone tells Google a great deal about how serious you are in building your business.

My dos centavos

[thebigdishman]

Hi

I have personally never liked CSS, and I have always prefered a traditional table design.

By the way your new site does not work at all well using Firefox.

Hope that helps a tiny bit at least.

[SemAdvance]

Maybe Google decided that they didn't like any web sites using Yahoo Hosting. Pointing all over the place in your site to a Yahoo cart could make google mad. If you use Google Checkout, maybe you'll be first for all of your terms.

Okay, all kidding aside. I would recommend a cleanup of code. This is just good design, may or may not help in regards to your rankings.

Second, I would recommend trying to find your own cart for all checkouts. This simply will allow all content and visitors to stay on your site.

Third, I would review matching up your title's to your <h1> to your linking structure, as this should help as well.

Finally, I would look at privacy and security issues. Make sure your site has a privacy policy. P3P if you can put the time into it.

Oops, one more thing, make sure your pages have plenty of content, major changes to a site can spell trouble if there isn't a marked improvement in the site.

?????

Most of the issues you have raised have little impact on rankings.

Google doesn't like Yahoo hosted sites????

That is just too far of a stretch .....point two has no bearing on rankings...matching title h1 and links does nothing of signifigance...plenty of sites ranked # 1 with no privacy policy posted.

[ran_dizolph]

@ seopro

google failed validation (http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.google.ca%2Fsearch%3Fhl%3Den% 26q%3Dgraphix%2Bplus%26btnG%3DGoogle%2BSearch%26me ta%3D)

as much as i agree that building validating websites is important, i don't agree with your theory that it has much to do with page rank.

[iany]

Yeh, meant to say that.
When designing using CSS, you must make sure it works on Firefox first and then make any amendments to sort out he non compliance of IE6. Fortunately IE7 is much better (not perfect).

That is why many folks still stick to tables. Tables are meant for data, not layout.

[marketing-helpersdotcom]

Yeah Ii know, but you said you identified their problem. Having non-valid code doesnt stop you from being indexed or being ranked at the top of the SERPs.

I agree, check out ESPN.com, Yahoo.com, CNN.com. They all rank well and they all have about 100 validation errors, and they all rank REALLY well.

Validation does matter in the sense that you want to make sure the spiders can access the entire page of all the pages in your site. Beyond that I wouldn't worry about it too much.

The time you spend cleaning up your code can be better used building quality inbound links.

[HHI_Golf_Guy]

Google seems not to appreciate major site-wide changes as they're usually indicative of spam. If your text to code ratio has changed significantly, Google might be worried that your domain has been overtaken by yet another spammy advertiser.

What a load of malarky! If you change all of your page names and don't use 301's, then it could cause some problems. But the poster stated he kept the page names and most of the content the same.

If you read what the poster said, he also stated that he included H1 tags in his new layout. Chances are, he added other text content or H tags. He may have to re-work some of his on page SEO.

A bigger question would be if the rankings drop is consistent on all datacenters (check the cache to see if each datacenter has picked up the new layout).

Even if all of the cached pages are the new pages, that might not give us a clue as to what is going on. DC results are all over the place right now. I'm not sure if it's an update, or if it's another infamous "data push" problem. For example, I see a competitor site #1 on some DC's, and #135 on others. That's a big difference.

My advice? Check across a number of DC's over the next week or so. If rankings on all DC's are still tanked at that time, take another look at the changes you made to the on-page SEO (H tags, text, etc.). The structure change of tables to CSS didn't cause the problem.

[marketing-helpersdotcom]

Yeh, meant to say that.
When designing using CSS, you must make sure it works on Firefox first and then make any amendments to sort out he non compliance of IE6. Fortunately IE7 is much better (not perfect).

You can use a loose DOCTYPE on all your HTML pages:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">

This will help with some of the differences in Firefox and IE.

Also, try to avoid using absolute values in your CSS. They can really mess up your layout at different resolutions and text sizes.

[schachin]

I noticed you have added a lot of words to your site and removed some others while removed things like 's' from the word tiles (Google sees tile and tiles as two different words).

Ok so by looking at the SEO toolbar in Firefox I can see that your kw percentages let's say for 'stone' were higher on the old page. Also you have no H1/H2 or H3 tags and fewer kw's in bold. One of your percentages (bold for the kw stone) is 2% lower than before. This would affect your ranking and depending on competitiveness it could do so dramatically.

I would start there. Hope this helps :)

[marketing-helpersdotcom]

What a load of malarky!


In case you all don't know what malarky is....LOL!

ma·lar·key /məˈlɑrki/ –noun Informal.
speech or writing designed to obscure, mislead, or impress; bunkum:



A bigger question would be if the rankings drop is consistent on all datacenters (check the cache to see if each datacenter has picked up the new layout).


I agree, it's possible it's a drop from an update. I wouldn't panic.

Do you have a Google sitemap installed?

The info Google gives you in the Webmaster tools may help you figure this all out. If everything checks out OK in the Google Webmaster tools then I would attribute it to an update and re-check your rankings in a couple days.

[Seopro]

Common sense tells me that since the search engines belong to the W3c they will give higher marks to a site with valid code. That is only one area out of 100 or 1000 areas that a search engine will rank for. I thing an interesting question is does any search engine skip over invalid code? The validator points out errors, does a search engine do the same thing? I always write valid code but that is my preference the same as I code by hand with very few mistakes and those are easy to fix. This is a subject that can go on forever because we are not engineers programing the spiders. It is just my opinion but what do I know I spent 15 years in the 3rd grade.
Don
Seopro

[southplatte]

Going on personal experience from a huge site redesign I recently performed, we went from a table based layout using a cgi script to build the data from the back end to an complete CSS based layout using php/mysql in the back end. Not only did we retain our #1 spot for our most prized keyword, in a matter of weeks we were picking up for all of our product names in searches and are number 1 for many of them accross the board. We not only changed page names, site layout, but web hosts also because we moved the site to our own in-house server. Also, much of the content changed, but not all.

So go figure. Figuring out Google is like figuring out the preciseness that the universe was created with. It just isn't going to happen any time soon.

I would follow some recommendation put forth already, regardless if you use tables or css for layout control:

1. Proper valid html code is necessary
2. Proper valid CSS is necessary
3. Proper content saturation with keywords is necessary
4. Site Maps are a large key, or at least proper navigation, and make the navigation show after the content (or so I have been told)
5. Make proper use of the Divs and their IDs.

If you have ten Divs with the same ID, why would Google, not get confused and think it was the same content repeated? It may just not be picking up all the content in the repeated Divs thinking it is duplicate, or it could be confusing the spider in some other way.

Make sure by using H1 tags and calling out some more keyword attention you have not over done it to the point that Google thinks there may be keyword stuffing happening, especially if it is getting hung up on the duplicate Div idea.

Anyways, best of luck to you and the site rankings.

[Refiner]

It looks like you are using div id tags on multiple occassions in the same document, such as <div id="general2of3">

i believe you can only use a div id tag once, whereas you can use div class tags as many times as you like. you need to validate your page using a html validator. the returning document will list the errors you need to fix.

i changed from tables to css and my rankings increased, but the site validated.

You are correct in saying that... if you assign a div id to 2 different elements in your document, you will have errors.

it is a good practice to use validators for checking your code as you can fix a lot of mistakes


here are some places to check for validation and compliance guidelines:
http://jigsaw.w3.org/css-validator/
http://validator.w3.org/
http://www.outfront.net/tutorials_02/adv_tech/compliance1.htm

[marketing-helpersdotcom]

I thing an interesting question is does any search engine skip over invalid code? The validator points out errors, does a search engine do the same thing?

I know that a spider will not skip over the code just because it is invalid. However, if the code is so improper that the spider can't access it (ie. page cannot be found in IE when your set it to show friendly HTTP error messages) then that is a different story. If that is the case the spider may skip over your Web site entirely.

I should also point out that browsers like IE and FireFox are more forgiving and will still display a Web page even though the code is poor. This is not the case with spiders. So I would advise installing a Google sitemap or using a spider simulator to check your Web pages, to ensure they all load up OK for the spiders.

[weslinda]

SEMAdvance...first, regarding the Yahoo hosting piece, did you not see it was a joke?

Second, regarding having your own cart, it most certainly would affect rankings as content pages are content pages, keeping content on your domain vs. Yahoo's is quite important. It also adds a bit of credibility to your visitors.

I'm not sure why you would think your title, description and H1 shouldn't match. They should all be on topic for your page and definately would help for search results. If nothing more than being more on topic for those reading the results and increasing your click count.

Finally, that's true, a lot do that without a privacy policy, but there isn't one merchant company with any ethics that will allow you to setup payment processing without one. Also, it gives more credence to your site and confirms it is a non-spam, non-phishing site.

Google recommends doing it, no reason you shouldn't.

A site can be done very well with CSS and DIV's and perform nicely even in IE6 if you simply don't try and cut corners.

Valid code is only part of the ranking process, other items will overrun that issue. If you have 47,900 IBL that www.espn.com has, then you can get by with screwing a few things up.

Here we're talking about small online shop, that doesn't have that kind of reputation yet.

Why is everyone so afraid of valid code and standards, why do people feel they are worthless to web site design?

Oh well, guess we should just tell everyone that shoddy work is okay, who cares if you get sued for not complying with US Section 508 accessibility laws?

[marketing-helpersdotcom]

Oh well, guess we should just tell everyone that shoddy work is okay, who cares if you get sued for not complying with US Section 508 accessibility laws?

I don't think anyone here disagrees with you weslinda. Good code is important, but in perspective to IBL's it's obviously not as much of a factor. Good code is certainly not going to rocket you to the top 10.

By the way, Section 508 on applies to federal agencies, not the private sector. I am an expert with section 508, I helped the DoD develop their 508 policies.

In 1998, Congress amended the Rehabilitation Act to require Federal agencies to make their electronic and information technology accessible to people with disabilities. Inaccessible technology interferes with an individual's ability to obtain and use information quickly and easily. Section 508 was enacted to eliminate barriers in information technology, to make available new opportunities for people with disabilities, and to encourage development of technologies that will help achieve these goals. The law applies to all Federal agencies when they develop, procure, maintain, or use electronic and information technology. Under Section 508 (29 U.S.C. ‘ 794d), agencies must give disabled employees and members of the public access to information that is comparable to the access available to others. It is recommended that you review the laws and regulations listed below to further your understanding about Section 508 and how you can support implementation.

[southplatte]

If you have 47,900 IBL that www.espn.com has, then you can get by with screwing a few things up.

That is all too true of statement. Many times these large sites that have so many errors, poor code are written from a complex back end that it is hard to comprehend what actually goes into creating the site and having it be 100% proper valid code. The large market, traffic, links and shear strength of the brand I feel factors into the search engine rankings as much as good code and content does sometimes.


By the way, Section 508 on applies to federal agencies, not the private sector.

While I do not doubt the intent of the purpose of Section 508, you must take into consideration the lawsuit against Target.com for not being accessible. It is a mere matter of time that it will be held mandatory (or something very similar) for the private sector as well. why not, if the tools, standards and practices exist, employ them now instead of needing to re-code/re-write an entire site to meet these needs of the consumer base that is more and more comprised of different individuals needing different tools, presentations and access?

Not to be off topic on that.

Back to the discussion: IBLs can help generate search rankings, but the spider must be able to read the entire page of content and if there is something that is hindering this, then it is not good.

Has anyone suggested the use of the Google web master tools to see exactly how and what the spiders are seeing on the web site? Might be a good set of tools to use to see from Google what is going wrong, though it may not be the definitive answer needed.

[activeco]

Design styling will have no effect on search rankings.

I wouldn't discard views that this could mean a major change.
I have checked the two provided links (old & new) through lynx and besides errors on the new one, the two look as completely different pages.
Besides, the content in the browser IS different. You didn't change the coding only, you changed the content too, with a pretty high ratio (from a robot's view).
This doesn't mean you'll not come back in SERPs soon.
You should have done it just more carefully; coding first and then eventually content.

[Seopro]

In 2000 the Olympics was sued by a blind man and he won. There is a lawsuit against Target Stores filed by the National Federation of the Blind. I didn't mean to get anyone upset over the ADA but it is illegal to discriminate against anyone. Even if there was only one blind person in the world designing for him or her would only be the right thing to do.
Don
Seopro

[Seopro]

Oh and while I am at it section 508 was part of a bill from congress that was passed in 1973 and the DoD did not put it into their directories until 2001. In 1998 all they did was reaffirm the law and that was the wage and rehab committee of congress. It is still against the law to discriminate no matter when it went into effect.
seopro

[mattk@nexsens.com]

@ seopro

google failed validation (http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.google.ca%2Fsearch%3Fhl%3Den% 26q%3Dgraphix%2Bplus%26btnG%3DGoogle%2BSearch%26me ta%3D)

as much as i agree that building validating websites is important, i don't agree with your theory that it has much to do with page rank.

Wow. Their main page has 66 errors http://validator.w3.org/check?uri=http%3A%2F%2Fwww.google.com

[Seopro]

I think we are all learning something from this discussion and there will always be 2 sides. I dont think PR has anything to do with how google places sites on a page but that is my opinion. I am baseing my opinion on valid code and designing for the handicaped because it is the right thing to do and also the biggest system of blind readers are the search engines. Now that one statement you can't argue with the rest we will have no way of knowing but we will always wonder about it. For every example leaning one way a person can produce 10 sites leaning the other way. We all do things different and that is what makes the web so great.
12% of all americians have some degree of a disability and I am a disabled viet nam vet but can still wat and talk and see. I think this discussion actually came into existance with sub-titles on TV and the movies.
Seopro

[janeth]

This took me about 2 minutes to find your problem. Doing your website in CSS is not enough. Go to the W3C and validate your code. You have 36 errors on your front page. Fix those and I bet your rankings will improve
Don
Seopro

If this was the case then someone needs to let Google know that most of the top ranked sites do not have code that validates. The code needs to be clean so the bot can get through it but it does not have to validate. (;

[Seopro]

What I was basically telling him to do was to validate his code so it will clean it up. You are right though very few have valid code but I know that is the way I code and I think you do the same thing. The only difference between us is you use tables and I use pure CSS in a tableless design and usually put my content first. (I looked)

[janeth]

What I was basically telling him to do was to validate his code so it will clean it up. You are right though very few have valid code but I know that is the way I code and I think you do the same thing. The only difference between us is you use tables and I use pure CSS in a tableless design and usually put my content first. (I looked)

I have used both tabeless designs and tables but never saw any ranking difference between the two and most the guys here are better with tables.

The last redesign we lost the CSS. I agree if there is a problem it does not hurt to validate the code and move forward from there.

[Seopro]

I think it would be great to know if it does make a difference but it is one of many pieces of the puzzle that makes up the search engine placement. Is it 1% or 0.1%, I dont know but I wish it was 25%.

[CarolineBogart3]

@ seopro

google failed validation (http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.google.ca%2Fsearch%3Fhl%3Den% 26q%3Dgraphix%2Bplus%26btnG%3DGoogle%2BSearch%26me ta%3D)

as much as i agree that building validating websites is important, i don't agree with your theory that it has much to do with page rank.

brilliant.

The css is a red herring. Something else changed.

[janeth]

@ seopro

google failed validation (http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.google.ca%2Fsearch%3Fhl%3Den% 26q%3Dgraphix%2Bplus%26btnG%3DGoogle%2BSearch%26me ta%3D)

as much as i agree that building validating websites is important, i don't agree with your theory that it has much to do with page rank.

brilliant.

The css is a red herring. Something else changed.

The css counts very little towards SEO but the pagerank is the count of the number of links and quality of those links that are pointing to your site.

If the anchor text is being put to the best use it can have major impacts on your ranking.

[Seopro]

It is one big puzzle and everything fits in somewhere we just don't know the size of the pieces.

[janeth]

It is one big puzzle and everything fits in somewhere we just don't know the size of the pieces.

I think the title is suppose to be the biggest piece but there are a lot more of the link pieces than any other piece of the puzzle. Making them more important when added together.

[Seopro]

I don't know Janeth but can a PR2 with 15 links be ahead of a PR5 with 102 links in the seo areas on Google? Big difference is standards and validation.

[Peter (IMC)]

I think all search engines are members of the W3C and I believe they will place a site written in standards above a site that is not with all things being equal. I have never failed to put a site in the top 10 that had valid code. That has just been my observation. On one key phrase affordable seo company oregon I have #1 #2 #3 #4 #5 #10 even though there is hardly any competition. On affordable web design company (yahoo) I have been #1 for 14 months with valid code. And on google I have a #1 with a PR2 ahead of a PR6 because I have valid code and he does not. I have notice this several times with not just myself but with other designers that use web standards.
We all have our own ideas which is great and that is why I joined this forum. I will learn something from others if I take part in the discussions and hopefully I have some knowledge to pass on to others. There's no competition what so ever, so it really has nothing to do with the code being validated. Try to validate the page he gave with tables, it also has HTML errors... If your logic was correct, he shouldn't have had high rankings with the old site either. So the problem is definitely not related to how valid the HTML code is.

Valid code is of no concern what so ever to a search engine. A website definitely does NOT gain points for having valid code. Neither does it lose points for having invalid code. However, valid code does make it a lot easier for a search engine to understand the content of the page and in that respect it does help.

But does valid code mean that you use h1 tags and p tags for example? AFAIK, it doesnt. Nor those the validation look at what was written in the title, metas and the actual content of the page. This is SEO related and not code related.

Pebbles,.... In case the URL's of your pages changed, did you use 301 redirects to the new URL's?

Did you change anchor texts in the links in your website? If you have, change the link texts back to what they were when you did have better rankings. That will probably do a lot of good.

In any way,.. just time will resolve this problem. A major site change like this doesn't happen without some ripples in the water,.. :)

[Seopro]

I don't think anyone knows the answer to this but we all have our own opinions which is great. I can say that I have been #1 on Yahoo out of 20 million for 15 months and everytime I check I am the only one that validated my code. The fact that I use CSS and my content is 60% compared to the others at 20% or less has an effect on this.
Also I dont think it makes a difference how much competition there is, it could be 300,000 in this case or 3 million or 30 million to have 6 places in the top 10 and those are the only sites that use valid code has to say something. I took the 6 places in the last 4 months when I added Portland, Oregon to my web site. Before that none of the top 10 validated.
We have all agreed to disagree which is what makes a forum so interesting and informative.

[Peter (IMC)]

Valid code is only part of the ranking process, other items will overrun that issue. If you have 47,900 IBL that www.espn.com has, then you can get by with screwing a few things up.

Here we're talking about small online shop, that doesn't have that kind of reputation yet.

Why is everyone so afraid of valid code and standards, why do people feel they are worthless to web site design?

Oh well, guess we should just tell everyone that shoddy work is okay, who cares if you get sued for not complying with US Section 508 accessibility laws?Valid code is absolutely not part of the ranking process. It can't be because it is completely unrelated to the actual content. There is of course an indirect relation between valid code and rankings.

I don't think people feel like valid code is worthless to webdesign. But you have to keep in mind that making sure all your code is valid is a lot of work with (from a business point of view) very little to no benefit. That´s just the way it is.


It is one big puzzle and everything fits in somewhere we just don't know the size of the pieces.Those pieces aren't so unknown and their sizes are also not unknown. They´re pretty easy to figure out actually because search engines have just one obvious goal,.. to put the best pages at the top. And it is just common sense that a title has a lot of information about what a page is about. The same logic applies to other "pieces".


can a PR2 with 15 links be ahead of a PR5 with 102 links in the seo areas on Google? Big difference is standards and validation. That simply depends on the keyword phrase that was used to search with... :) so the answer is yes, that is possible. At the same time, validation has little to do with it.

[Peter (IMC)]

I don't think anyone knows the answer to this but we all have our own opinions which is great. I can say that I have been #1 on Yahoo out of 20 million for 15 months and everytime I check I am the only one that validated my code. The fact that I use CSS and my content is 60% compared to the others at 20% or less has an effect on this.
Also I dont think it makes a difference how much competition there is, it could be 300,000 in this case or 3 million or 30 million to have 6 places in the top 10 and those are the only sites that use valid code has to say something. I took the 6 places in the last 4 months when I added Portland, Oregon to my web site. Before that none of the top 10 validated.
We have all agreed to disagree which is what makes a forum so interesting and informative.

That huge number of results is caused by individual words. If you want to know your real competition, search with quotes. For your phrase "affordable seo company oregon" there are 9 results, including some of your own posts in various forums ;) There's no competition for that phrase.

Also, you said you gained rankings by adding "portland, oregon" to your site. That alone is already prove that your valid codes didn't put you at the top. The content did!!!

Your conclusion that you rank #1 because of valid code is wrong. Test it by adding a couple of errors to your code and see if your rankings change,. :)

Agreeing to disagree is nice, but it is also being plain old stubborn and you´re definitely not learning anything from sticking to your own opinions without even the consideration there might be more to it than just valid codes,. :)

The error often made is the logic that:

"I did A and B happened, therefore A is responsible for B happening and this always applies in every situation."

If you live by that rule you´re going to lose a lot of time on a lot of stuff... :)

[Seopro]

With the ever changing algos I dont thing anyone knows for sure the size of the pieces but we can all take a wild guess and each one of us will guess different.
For 3 years I could not put a web site in the top 100 and after I learned CSS the last 4 years I have put every site I optimize in the top 10. Valid code is just as easy to write as invalid code and it is just my preference to valadite every page through the w3c. It seems to work for me so I am happy.
Don
Seopro

[Peter (IMC)]

With the ever changing algos I dont thing anyone knows for sure the size of the pieces but we can all take a wild guess and each one of us will guess different.
For 3 years I could not put a web site in the top 100 and after I learned CSS the last 4 years I have put every site I optimize in the top 10. Valid code is just as easy to write as invalid code and it is just my preference to valadite every page through the w3c. It seems to work for me so I am happy.
Don
Seopro

You learned CSS, which resulted in clean code,.. :)

And don't get me wrong, I appreciate webmasters that makes a point out of validating code. For me however, going as far that the code 100% validates is commercially simply not a smart thing to do. It takes too much time to prevent every little error that can happen.

But be careful with the asumption that valid code gives a website some kind of real advantage over sites that do not have valid code.

[Seopro]

There are more than 9 companies in Portland that use the same keywords. I can name at least 30 that I know personally.
The words "portland oregon" just targeted local at the same time I was #1 on yahoo for almost a year out of 20 million. Those two words have a minimal effect on content. And I don't worry about how much time it takes because it is not that much more time to create valid code. I bid by the job and my customers pay me well for the results I produce for them.
Also I am a disabled vet, all I have is time so I dont see any reason not to do things according to the W3C.
Being stubborn is a trade mark of mine. I tell my customers it will be done right or I will not do it and web standards are there for a reason. I have more customers than I can handle. I work with the PGA Tour and Turner Enterprises as a consultant plus I work with PGA star Phil Mickelson as his web designer. I guess that is doing pretty good.
Don
Seopro

[Seopro]

I have sat back many times and thought about what if I am wrong. But what if I am right? You have not even considered that and I have been almost every day for the last 4 years.
Don
Seopro

[k001]

I think youll regain your rankings.

[Clint1]

I've heard others say similar things about them going to CSS then getting their sites penalized. This is why I never have changed to CSS. If it ain't broke, don't fix it. It's pretty stupid though that an SE would penalize for it, but we all know they do stupid things.

Can anyone please tell me if there would be any negative affects of me just going to a CSS nav menu on my pages?
Thanks.

[danners02]

Validate your code:
http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fpebblez.com%2F

I did not find any H1, H2, H3 tags on the homepage?
Get an H1 tag in early on in the source code with your keyphrase(s) in.

Get rid of the comments, eg:


Clean up any errant spaces in code, eg:
[b] Decorative Wall Clocks Made

Get rid of underlines in spans:
<span class="navbig1">____________________</span>
use CSS to put a border in instead

try to use CSS for spacing, not a series of
tags

Major code improvements/amendments do sometimes have a short term negative effect on SERPS but implement the above advive and it should perform better than previous site

[brucet]

Our web site throws up all kinds of W3C errors, and the code is a honestly a bit of a mess. But that hasn't stopped us holding top positions on Google for some pretty competitive keywords over a sustained period. We have been planning to clean it up, as it does seem to affect our rankings in Yahoo and MSN. But after reading this thread I am having second thoughts - our Google rankings are too precious to take risks with.

There may well be 'invalid' code that makes a difference to search engines, but there is clearly also invalid code that doesn't. I have posted on this subject here before, and got some quite confusing responses which had very little in common with each other! I did some research elsewhere as well, but I have never been able to find any real consensus about which aspects of code validation are important from an SEO POV, and which are not.

[hasit]

You said it right. You are trying to create a new website (Which Google is going to screen it throughly ), then you are trying to hide some text.

I would HIGHLY recommend you should submit the sitemaps again and place a reinclusion request.

btw: is your website found by site:www.domainname.com?


IBL's? I gain about 100 a month these days according to yahoo. As far as my sites linking structure I didn't really change it, although I did make a standard nav bar for every section of my site, which has made things a little more uniform.

As far as a sitemap I have a google site map (am looking for a decent ROR sitemap generator) I also link to my homepage from every page.

I do have a theory and if someone could take a look at this I would really appreciate it. In my header I have an H1 tag around the first text on the page, and its just about the first thing that shows up in the code. At the top of the header div I have a black picture background and at the bottom I have set the background to just be white. The text in the H1 is black and would show up against that black background, but I have added margins to the H1 to move it down.

Could the search engines inability to read the css margins on my H1 cause them to think I'm hiding black text against a black picture in an H1 tag right at the top of my page? Can google even tell the picture is black?

I would really appreciate any feedback on this.

[lala]

danners02:
"Get rid of the comments, eg:
"

But how would you do this if you use Dreamweaver's templates?

lala

[dux]

I was wondering what had happened to pebblez as there has been no input recently, but it looks like he/she has been busy cleaning the code up! Hope it works for you....!

As a side issue to this, how do people feel about restyling a h1 tag using css so that it blends easily into the page content? By this I mean changing the standard h1 font size to say, 12px or something similar? Is this black, grey or white hat seo?

[SemAdvance]

weslinda

Since most cart pages are behind https:// most search engines cannot willnot crawl those pages so this content will not add or detract from search positioning.

Why you would not want to always match titles to h1 tags???? Search engines are data mining robots that store information and can detect patterns and in detecting a pattern it can determine things that it may or may not like about your pattern....

patterns can hurt you for example if you use a pattern of link building from link farms, or use the exact same words in your title. meta description, meta keywords.....it then becomes apparent you are trying to gain search engine positioning and are not doing the best work for the people who visit your site.

acttiveco

Yes you are right major changes could have an effect...I should have been more clear in my post.

seopro


I dont think PR has anything to do with how google places sites on a page but that is my opinion.

This is fact. I have a PR1 site ranked continually month to month at #1 on Google for a moderately competitive keyword term and nobody can move it from it's position, and the 9 other sites all have PR3 and higher.

Getting to the front pages of Google is like solving a puzzle or cooking a great chili. There are many ingredients needed... or most of the puzzle pieces put together... to give the search engines a clear picture or a taste of the some really great chili.

Lastly...if you are counting kw% percentages on pages you are really wasting your time....the scoring factor for this is of very little import. You are much better off writing for people than machines.. LSI eliminates most of the density and weighting scoring in body content.

Dux

Using the h1 tag and applying styling to it to make it look good is fine....Google says build websites for users not build ugly websites for users.

The problem arises when a webmaster uses 10 h1 tags on one page with the same keyword term, that search engines would likely have a problem.

Alot of it comes down to intent at times...

[SemAdvance]

danners02:
"Get rid of the comments, eg:
"

But how would you do this if you use Dreamweaver's templates?

lala

No need to.

What could be referenced here is when people stuff comments with keyword terms which you are not doing.

Google and other engines know Dreamweaver templates use commenting quite heavily and has no problems with this.

The employees at Google are CS engineers and know a great deal of what they are doing. It would not be much of a company if it could not ascertain what is normal operating procedures for other software companies.

[weslinda]

Only two comments. First, it's not more difficult to create a valid site, in my opinion it's quite easy if you start correctly. Do you work right, don't keep f'ing around.

Matt Cutts recently did a video based on what makes a good site. #1 issue is that your site is browseable by a text browser, all the way through your site.

Valid code, with clean layout and proper navigation makes this happen very nicely. The better a site can be navigated, the better it will do.

Everyone here is discussing how valid code can't affect search engine results, and for that, I have to disagree. I recently took over the design of a major medical web site. 100k unique visitors each month. The site had begun plummeting in the rankings over the last 6-12 months. I have begun working on the site, adding only the basic content that was already being added which was 2-3 news articles each month and nothing else.

I have begun validating the code, and replacing image based link menus with text based CSS UL menus. Within 30 days, we've seen a jump of almost 100% in the number of referrals coming from Google to our site. And that numbers grows each and every day.

None of us know everything that makes a site rank in the search engines. Anyone that tells you they do, run.

I'm not saying that the valid code is an important factor just cause I like to spout the cause of doing things the right way. I discuss this as part of the focus of a site because it is the right way to do things and I've seen it positively affect a web sites ranking in Google and other search engines. The most noticeable change was Google.

[pebblez]

First of all, I can't believe all of the helpful advice I have recieved from so many well respected and brilliant individuals in our industry. Its feirce discussion such as this that make this the best forum around for SEO and SEM

There are a few things specific to my website which may have bearing on this discussion. First, when my homepage wasn't validating it was doing so for a few specific reasons. The major problems arose from a Google search box at the very bottom of the page, and an email form which I took directly from my yahoo hosting provider.

The other problems were caused by a bit of code which my webhost yahoo inserts at the bottom of my homepage. I have completely cleaned up my homepage code and validated it, and I still get 6 errors because of this snippet which is added by Yahoo on the server. Is it possible this bit of code is used to give a boost in yahoo for sites which host with them? Do you think google would use this code to penalize yahoo hosted sites? Probably not but who knows...

Either way its time to consider a new host.

The rest of the errors were the stupid duplicate div id's. I'm new to CSS. I will learn.

The fact that most of the errors were at the very bottom of the page and were caused by things which existed in my old site, didn't hurt it, and were provided by the SE's I am reluctant to declare my sloppy code as the cause of the problem.

And, I will be honest, I have always been one of those who didn't think valid code helped in rankings.

However....despite this I respect all of those who have said it helps so much that I can't risk it with this site, and as I said I quickly validated my homepage as much as I could, and cleaned up the other pages as well. I am using dreamweaver templates on this site, so it has been a little bit quicker than otherwise.

I will keep this post updated with the results of my valid coding. Unfortunately I wont be a good scientific model as I am also tweaking other things on page including my headings, in order to fix this.

I would say that the question could be answered pretty easily by putting up a number of websites all going after the same set of keywords, all with the same amount of content, and with the only variable being whether the code was validated and how badly it wasn't. Even if its a low competition keyword, or if all of the sites remain back in the 500's of google, you can still compare them relative to one another. You might even be able to do it with one site and multiple subdomains.

I cant tell you all how much I appreciate your expert help, and I will keep this updated with the results of my validation efforts.

[Seopro]

I agree with linda that it is easier to write valid code than non-valid code by starting right. Common sense tells me that since the search engines belong to the W3C they would give just a little more weight to web standards. My content to code ratio is higher and everyone knows it is about content not how much code you put in. Pure css gives me more control over the look and feel of a site. And bottom line, traffic has increased.
Lot of good information on both sides of the fence.
Don
Seopro

[Peter (IMC)]

I would HIGHLY recommend you should submit the sitemaps again and place a reinclusion request. Reinclusion is for sites that have been banned completely (ie. completely removed from the index.) That´s not the case here so a reinclusion request is absolutely not necessary and won't have any effect.

[Peter (IMC)]

Everyone here is discussing how valid code can't affect search engine results, and for that, I have to disagree. I recently took over the design of a major medical web site. 100k unique visitors each month. The site had begun plummeting in the rankings over the last 6-12 months. I have begun working on the site, adding only the basic content that was already being added which was 2-3 news articles each month and nothing else.

I have begun validating the code, and replacing image based link menus with text based CSS UL menus. Within 30 days, we've seen a jump of almost 100% in the number of referrals coming from Google to our site. And that numbers grows each and every day.
Why do you assume it was the validation of the code? You´re talking about basic SEO techniques that have nothing to do with valid codes. You say you did the following:

* adding new content
* change imaged based links to text based links

THAT is why you gained positions in Google. Not because your codes are validated... :)

You say you did A, B and C,.. and your conclusion is that just C was responsible for your higher rankings.

You probably think I am doing the same thing by saying that A and B are responsible for higher rankings and that C had nothing to do with it. And you´re right! :) If a search engine uses valid codes as a ranking factor, it would result in many popular sites to drop in the results without any real reason for it.

What should the search engines do? Be like: "Oh this site has 10 errors and that site has 100 errors, therefor the site with 10 errors is better." That doesn't make any sense what so ever.

[Tolis]

Hi all

I am running the site easytraveller.gr and after your suggestion i try to convert from flash into css. many of my pages including the main ones are converted. The old designed site had a page rank in Google 5. The same happened with the new designed for the first time. after 2 or 3 weeks the page rank dropped to 0 and for 2 weeks the site disappeared from the search results even with the keyword being the domain name. Now the results seem to come back but the page rank is still 0. My other domain easytraveller.net that maps to the above one has the same page rank 2 as it was before. Hope for better days unless someone has a good idea.

[Peter (IMC)]

Common sense tells me that since the search engines belong to the W3C they would give just a little more weight to web standards.
validator.w3.org -->> www.google.com (http://validator.w3.org/check?uri=www.google.com) Result: Failed validation, 66 errors

validator.w3.org -->> =www.yahoo.com (http://validator.w3.org/check?uri=www.yahoo.com)Result: Failed validation, 37 errors

validator.w3.org --> search.msn.com (http://validator.w3.org/check?uri=http%3A%2F%2Fsearch.msn.com%2F&charset=%28detect+automatically%29&doctype=Inline)Result: Failed validation, 60 errors

Not in their own sites.. :)

[weslinda]

Peter, we didn't add new content. We continued the 2-3 articles a month that has been done ever since the site was begun in 1998.

Are you sharing that the only reason for our uptick could be attributed to text menus? Valid code has nothing to do with it? Come on, lets get real, I think you and I should agree to disagree.

You seem to be a staunch advocate of not validating code. Why is that? I believe a web site should be built to be functional to all, and that will have an affect on search engines. Why is that such a unbelievable concept?

If accessiblity and valid site code unimportant, why is google developing their accessible search? http://labs.google.com/accessible/

Yes, Google will rank pages with invalid code in the first position? Any idea why? Less than 1% of all web sites actually have valid code. They have to. If they only picked from sites that have valid code, then there would be problems, they couldn't show off many pages now could they?

Pebblez: I'm glad you got your answers and I think the more you work on the site, the better you'll do in your search results. There are lots of ways to get to the top of the search engine, and in my opinion, doing things the right way, will get you noticed by the search engines and make your visits by customers more enjoyable.

In the end, it is about the visitor, and not the search engine.

[Peter (IMC)]

Are you sharing that the only reason for our uptick could be attributed to text menus? Valid code has nothing to do with it? Come on, lets get real, I think you and I should agree to disagree. Maybe its that I look at it from a algorithm point of view. How valid code is, can not be a factor in the rankings because of what you explained about less than 1 % of the sites having valid code.

However, if you do turn the code in a website into Valid code, then the result will be the search engines can understand the content better. But that again is not because of how valid the code is. It is a side effect of using valid codes.


You seem to be a staunch advocate of not validating code. Why is that? I believe a web site should be built to be functional to all, and that will have an affect on search engines. Why is that such a unbelievable concept? There´s nothing wrong with using valid code. And I am not against validating code. The suggestion was that all you have to do is validate your code and the search engines will reward you for it with lots of high rankings. I believe that at least 1 person should mention that this is not correct... :)

But I will leave it at this. I'm not writing here to convince you and others of what the real importance is of valid code (as far as high rankings are concerned of course) but just to give all the readers the idea to look further than just a validator at w3.org

[weslinda]

There hasn't been anyone that has ever shared that only valid code can get you to the top of the rankings. I didn't say it, no one here has said it.

We've only said it is a part of the equation. Do you know the entire algorithm Google uses to rank a site? Do you have every piece they use to rank a web site?

[Peter (IMC)]

for some reason this was duplicated. So I deleted this post, you can read the content in the other post.

[Peter (IMC)]

There hasn't been anyone that has ever shared that only valid code can get you to the top of the rankings. I didn't say it, no one here has said it. Are you sure? This was what I replied to originally:


This took me about 2 minutes to find your problem. Doing your website in CSS is not enough. Go to the W3C and validate your code. You have 36 errors on your front page. Fix those and I bet your rankings will improve
Don
Seopro
You can argue about what that says, but to me it says that the problem of dropped rankings is because of validation errors. I don't agree with that and I never will... :)

(added: By the way, I have nothing against Seopro. He´s a member in this forum just like all of us are. He just was the one that made the statement. What ever I say in the forum is never against anybody. I´m just sharing my experiences and knowledge of SEO.)


We've only said it is a part of the equation. Do you know the entire algorithm Google uses to rank a site? Do you have every piece they use to rank a web site? Do you need to know in exact detail how an engine works in order to drive a car? No you don't. But the best drivers in the world, do know how an engine works. They need that knowledge in order to do a better job at driving the car. However, they do not know the car in detail like the top engineers that developed the car.

Now keep in mind that the pilots of the car are much better drivers than the engineers who developed it.

Understanding what gets a website high rankings is not a matter of knowing the details of the algorithms, it does how ever require the knowledge of how a search engine works. The question of wether or not Google uses coding errors them selves to determine positions can be answered by using simple common sense.

Suppose you have 4 webpages about the same subject. One has 68 code errors, the second has 29 code errors and the last only 4 and the last one has no coding errors what so ever. Now I ask you: "What does the number of code errors tell you about the relevance of the pages? The answer is: Nothing!

So why would they spend valuable processing power on a factor that tells absolutely nothing about relevance?

[Seopro]

What I was refering to was the improper use of the div id in the css code. If he validated his code he would find out exactly where those problems were and could fix them. The big question is if the search engines will skip over code they do not understand? Does improper use of CSS have the same effect as the use of JavaScript? Only the search engines know that answer for sure and they are not talking.
Don
Seopro

[pathikbd]

Dear Pebblez,

I've not used CSS before, but I guess I know what happened. I looked at your CSS file and found something that is making you rank low.

Google recognizes your important keywords between some special tags, like:

Bold Tags:
Strong Tags:
Heading Tags: <h1></h1> or <h2></h2>
Font Weight Tags: <font size="#"></font>

Astonishingly, I saw some [b] tags in your source HTML, but no [b] or Bold Tags. You don't have a single <h#> or Header Tag either. Your Font Size tag does not show more than "-1" value.

Google therefore weighs your unimportant keywords and important keywords in same scale. But, don't blame it on CSS. It's a good thing to make your site rich & dense with your important keywords. i.e. Achieving high keyword density page is not possible without CSS. Because, you can eliminate a lot of messy HTML codes in your pages and most of it will be enriched with keywords.

My suggestion would be to eliminate the following lines in your CSS file:

[font-weight: bold;] in line # 51, 61, 73, 129, 187, 190, 317, 322, 340, 347, 370, 377, 603, 607 (Just give a "Find" and "Replace All" command with nothing)

I hope you got the idea. Similarly, delete these lines in your CSS file:

[font-size: ]
[h1] / [h2] / [h3]

Finally, reformat the whole site using these tags manually and give Google some candies to lick on:

Bold Tags:
Strong Tags:
Heading Tags: <h1></h1> or <h2></h2>
Font Weight Tags: <font size="#"></font>

I think it should work out and I believe that it can give you better result than your past performance from Google. Most of all Yahoo and MSN also will like you.

[noel_x99]

I'm one of those people that don't believe that valid code affects rankings much.

I have lots of older sites out there that don't validate. Some that use tables and really bad formatting...some CSS with cleaner code - Yet a number of these rank well and convert well - even for some competitive terms.

SEs are looking to provide the answers that their "customers" are searching for. All things being equal, maybe they do rank valid code, or pages with a better code/content ratio higher. But how often are "all things equal"? One site may have a better page title or better content or better use of the keywords that people are actually typing in. It seems to me that these pages rank better than pages that have valid code.

A search engine "customer" doesn't care if the code is valid or not. They care if they find the answer they are looking for. It seems to me that's what the SEs care about too.

[Peter (IMC)]

Does improper use of CSS have the same effect as the use of JavaScript? Only the search engines know that answer for sure and they are not talking.
Again, a little bit of common sense will provide you with the answer. We know what the effect of javascript is, right? What ever is inside javascript is invisible to search engines.

If "improper use of CSS" would in fact have the same effect as javascript then 99% of the internet would be invisible to the search engines,.. :) So obviously, it doesn't have the same effect.

Bad codes reduce the ability of the search engine to destract the content from the pages. But in principle you can assume that if your browser is able to show you a website, Google is able to destract the hypertext content from it.

Ofcourse, the proper use of h tags and p tags and alt atributes, and titles and etc... will give you the SEO benefits, but if you don't use them, your code can still validate, can't it?

[pebblez]

interesting thing about the H1 tags...

I had them originally. one on each page indicating what the page was about.

However for several reasons based on searches I did trying to get inside the SE's heads, i thought the H1's might be hurting me so I removed them.

As soon as I did I saw some of my more important rankings returning. Not all of them, and it could be a coincidence, but who knows

I put them back...and a day later I was gone. So...I took them down again, and for now they are coming back a little.

It may be that me suddenly adding h1 to 700 different pages all at once caused google to think i was h1 stuffing or something, and that combined with all the other changes flagged me as a spammer

whatever it is, everything I do seams to be working opposite to how it is supposed to these days. Im ready to go back to link building already

[justicewhite]

interesting thing about the H1 tags...

I had them originally. one on each page indicating what the page was about.

However for several reasons based on searches I did trying to get inside the SE's heads, i thought the H1's might be hurting me so I removed them.

As soon as I did I saw some of my more important rankings returning. Not all of them, and it could be a coincidence, but who knows

I put them back...and a day later I was gone. So...I took them down again, and for now they are coming back a little.

It may be that me suddenly adding h1 to 700 different pages all at once caused google to think i was h1 stuffing or something, and that combined with all the other changes flagged me as a spammer

whatever it is, everything I do seams to be working opposite to how it is supposed to these days. Im ready to go back to link building already

Hmmm.... I haven't heard anything like this before. Perhaps do these changes in a controlled manner and let us know how it goes. Maybe you've discovered something here.

Anyhow, good luck with your site.

[Seopro]

Peter,
I was talking about major mistakes not minor ones like forgetting to close a paragraph. Frames are invisable and so is JavaScript, if you tell the browser that the text is written in hebrew which is right to left will the browser read it and will the search engines index it. Search engines can not index flash and a few other programs. Anyway we agree to disagree and that is very informitive to others.
Don
Seopro

[weslinda]

Peter, I think we're missing each others boat. But I hope to rectify my thoughts here, and my thoughts only.

1. I've never said that Valide code is the only reason a site does poorly, or well in SERPS. I will never state that, and I will never agree with anyone that does.

2. I will and have said that I believe that using valid, and well laid out code, can improve your rankings. Be that through the engines being able to extract the information on your pages more effectively, or if the code to content ratio is lower, because I've effectively used only the amount of code I need to display the page.

3. Great search engine rankings have to do with so many things, it cannot be said that one thing will lead to great results. If you have a crappy site, with no content, with no one linking in, but you have valid code, that alone won't get you to the top. On the other hand, if you have a site, with great content, but don't use valid code, and clean content layout, I think you will do more poorly than someone who takes the exact same site and cleans it up. I've seen this happen time and time again.

In the end, web design should be done with the visitor in mind. Ensuring all visitors can access all information is what everyone should be striving for. Until designers do that, the search engines will have to rank poorly done work because they have nothing else to work with.

In this particular case, I think there are numerous issues that are out there that affected this site, and I surely don't think it has to do with the flip to CSS. I think the other side of this coin is that we have to look at long term results. I've seen sites do an initial slide only to come back in a few weeks to a higher spot than when they started.

[Seopro]

Hats off to weslinda that is exactly what I have been saying. The bulk of my business is taking sites out of tables and into css and optimizing it. Needless to say I am very successful with what I do and I will always use valid code no matter what.
I have a feeling if you read books by Eric Meyer or Jeffery Zeldman on CSS and Web standards you might have a reason to think about what we are saying. I know the search engines will drift more toward valid code than away from it.
Don
Seopro

[greeneagle]

Let's make it real simple here, we seem to be drifting off.

Stepping back to an earlier post, I find it really hard to believe that the GOOG or any of the main SEs would inflict any kind of SERP or Ranking penalty for moving to an all CSS Site from tables or a combination thereof, as long as:

1. Site structure (URLs) are maintained in the transition.
2. Each indexed page doesn't represent significant relevant topical and/or content changes.

There has never been and most probably never will be substantiable evidence to conjecture or propogate rumors otherwise.

There is absolutely no penalty for extreme content to code ratio changes under the simple conditions met above.

Surely we don't really expect extreme logical "disjoints" in maturing "logical machines", do we?

Ken

[kgun]

Valid code is just as easy to write as invalid code and it is just my preference to valadite every page through the w3c.

Fully agree. The top of laziness is doing everything correct from scratch. Also note:

Some HTML markup is being deprecated.
A few here mentioned du (n) plicating the same ID. How does the Se BOT see that lack of uniqueness? That is a type of markup we call "spagetthi coding" and should be avoided by all means. The big difference between a CSS class and CSS ID is that an ID should be unique.
An open tag in an XML document is a serious error. The DOM tree is incomplete. If you are manipulating DOM nodes, how can you find the correct node (element) if there are open tags? A DOM tree is a simple datastructure. You can look at HTML as a dialect of XML, a subset for those with a little mathematical background.
I can not understand why designers / coders in march 2007 can be so sloppy on incorrect markup and code.
Last argument. How easy is it to update and change a site with a site-wide linked or imported stylsheet and JS in external include files? How difficult can it be to update a site with lot of pages with open HTML tags? Time is money, money / time that can be used on SEO and PPC.

[kgun]

Just noted this related WPW thread posted March 07:
Designing Search Engine Friendly Websites (http://www.webproworld.com/viewtopic.php?t=72807)

Note the opening sentence:
"SEO on a website should begin during the design, redesign or template stage not as an afterthought once the site has been launched".

[Peter (IMC)]

whatever it is, everything I do seams to be working opposite to how it is supposed to these days. Im ready to go back to link building already This can be explained by the historical data algorithms. When you do something drastic to your pages it can happen that certain ranking factors are set to an age of 0 (zero). This causes your average age to go down which results in a drop in rankings. what you´re describing is very normal. You add an H1 tag and your rankings drop, you take it out and the rankings come back. That seems to be contradicting the logic that the h1 tag is about the second most important tag (after the title). But all you have to do is give it some time and your rankings will come back stronger than they were before. It just requires that painful thingy,.. patience.. :)

[Peter (IMC)]

Weslinda,

I liked your resume. Good post!

Peter

[weslinda]

My resume Peter? Please do tell...I think I missed something.

[msaqibansari]

I had used a table based design for years, and I had good rankings on all of my keywords. Recently i changed over to a pure css design including the use of h1 tags, divs instead of cells, the whole works.

When the site was finished I uploaded it to the web and waited. A week later my rankings started dropping. First they dropped in google. Then they started dropping in yahoo. Msn has been fine, but who cares.

Can anyone tell me what I've done wrong?
The new site can be seen at http://pebblez.com
One page of the old site can be seen at http://pebblez.com/index2.html

I really appreciate any advice you can give me. Thanks


Same problem with my site.
www.superdiscountshop.com (http://www.superdiscountshop.com/)

[greeneagle]

msaqibansari,

Welcome to WPW.


Same problem with my site.
www.superdiscountshop.com

Your site has about 500 different topics on the home page. The great machines in the prairie (SEs) don't know which "silos" to deposit or retrieve your information to and from.

Those "logical disjunctures" are much more excusable at "Trusted Site" levels.

That is a different issue.

If you want to open up another thread and get some help, you may want to try the "Site Review" forum.

Ken

[Peter (IMC)]

My resume Peter? Please do tell...I think I missed something.

resumé as in abridgement, abstract, summery

Not the list of previous jobs and schools and universities you attended... :)

Since english still isn't my native language, I guess I pick the wrong words sometimes.

[kgun]

resumé as in abridgement, abstract, summery

Not the list of previous jobs and schools and universities you attended... :)

Since english still isn't my native language, I guess I pick the wrong words sometimes.

Where do you have your IT education from?

When is the next time you come here and tell us how to cook spegetthi?

[greeneagle]

We have a lot of room for misunderstanding in an International forum with English not always being a first language, complicated by the fact that we hail from differing business specialties and experiences.

Ken

[weslinda]

I have a hard enough time with English and I was born and raised here in the US.

[pathikbd]

interesting thing about the H1 tags...

I had them originally. one on each page indicating what the page was about.

However for several reasons based on searches I did trying to get inside the SE's heads, i thought the H1's might be hurting me so I removed them.

As soon as I did I saw some of my more important rankings returning. Not all of them, and it could be a coincidence, but who knows

I put them back...and a day later I was gone. So...I took them down again, and for now they are coming back a little.

Like I said before (http://www.webproworld.com/viewtopic.php?p=355034&highlight=#355034), your pages had no h1 tags and most of the tags were integrated in the CSS file. Nice to know that your site is getting good ranks again. Please put emphasis on bold tags also. I have a lot of websites getting good ranks and most interesting thing is, I never cared about CSS formatting. And the W3 Validation shows horrible reports.


It may be that me suddenly adding h1 to 700 different pages all at once caused google to think i was h1 stuffing or something, and that combined with all the other changes flagged me as a spammer.

I am not quite sure about this. But limiting your keywords and SEO emphasis for large number of pages is a decent thing to do. If you have more than 700 pages, then I would request you to make the site more user friendly, than search engine friendly. Large number of pages slightly SEO optimized and mostly User optimized can get you good rank.

[kgun]

I have a lot of websites getting good ranks and most interesting thing is, I never cared about CSS formatting. And the W3 Validation shows horrible reports.


So you recommend not validating your code and writing correct CSS, see my post regarding CSS ID and CSS class above.
Could the situation have been better if your markup and styling was X(HTML) and CSS valid? In other words, do you think valid markup and code make things worse?
What about tomorrow? Do you know the future? Do you know what deprecated HTML markup means?
Do you know what a HTML and DOM inheritance tree is? Do you know how that relates to DHTML, the document object model and future technologies like AJAX?

At least you and other people that support spaghetti coding and markup should buy the following book:
Stuart Langridge (November 2005 or later): DHTML Utopia: Modern Web Design Using JavaScript & DOM (http://www.sitepoint.com/books/)
read it and come back and argument.

Note:
CSS is a central element in tomorrows technologies like DOM scripting and AJAX.




Last argument. How easy is it to update and change a site with a site-wide linked or imported stylsheet and JS in external include files? How difficult can it be to update a site with lot of pages with open HTML tags? Time is money, money / time that can be used on SEO and PPC.


Is this important?

[Seopro]

Good clean code means forward compatable and the only reasons not to use valid code is you are paid by the hour or you dont know how. The first I cant do anything about the second is easy to take care of. I have not seen any writers on web design or CSS recommend invalid, sloppy code and I think they know more than I do. I make a darn good living cleaning up after someone that calls themselves a designer. My editor of choice is notepad or a free editor that allows me to write clean code. Clicking on a button does not make one a designer. Web standards is the only way to design.
Don
Seopro

[kgun]

Clicking on a button does not make one a designer. Web standards is the only way to design.

Good advice.

Yesterdays web was mostly about online brochures, catalogs and static pages. The first version of HTML was mainly about formatting and linking commands. HTML is a markup languge and no programming language. CSS is a styling language. Even if the class concept is used, it is not a class in the object oriented programming meaning of the word class. As such another word could have been better, since someone that know what a CSS class is may be fooled to think they know the meaning of the class concept used in object oriented programming languages.

Tomorrow's web is more about web services and web applications. With asynchronous JavaScript calls, first via hidden frames of zero hight and with, then iframes and last the XMLHttpRequest object and AJAX we have seen the start of this evolution. Then code and markup related to this code becomes much more important.

This will also require more of the SE's (http://www.webproworld.com/viewtopic.php?t=71425).

[Seopro]

We have no idea what the future on the web holds for us and it is my opinion that web standards will help in any direction design takes us. I remember as a small child drawing animals on cave walls with bits of charcoal now I use a graphics program. Its amazing the progress we have made.
Don
Seopro

[zoreli]

Same thing here. I change my table design with CSS design, and my pages are XHTML strict 1.0

My domain http://web.goto.gr vanished from serps on goolge.gr for my main keyword web design. I was on page 3 in SERPS and now I am nowhere to be found.

All my pages are valid xhtml strict 1.0 (click on the link on the bottom on the page). So something else should be the problem, not the validation of the code. Hopefully, my site will be back in SERPS soon, since 90% of my clients come from the site.

I can't do anything except to wait and pray now...

Regards, Zoreli

[pathikbd]

So you recommend not validating your code and writing correct CSS, see my post regarding CSS ID and CSS class above.
Could the situation have been better if your markup and styling was X(HTML) and CSS valid? In other words, do you think valid markup and code make things worse?
What about tomorrow? Do you know the future? Do you know what deprecated HTML markup means?
Do you know what a HTML and DOM inheritance tree is? Do you know how that relates to DHTML, the document object model and future technologies like AJAX?

Dear Kgun,

In my previous 2 posts, I have encouraged peblez to use CSS files. The reason I have not used CSS before, because it eats up a lot of time. And after Technical & SEO copywriting, CSS formatting is require extra time & care. Therefore, I usually avoided it if I could. This time I am working with CSS for the www.colorexpertsbd.com site.

Yes situations become better using Valid CSS. And valid CSS does not make thing worse. I was actually talking about the effective way to use it.

As a business graduate I emphasis on only on marketing efforts. I don't know what DOM inheritance tree is and I am quite familiar with dynamic webpages.

Thanks for the post.

[kgun]

All my pages are valid xhtml strict 1.0 (click on the link on the bottom on the page). So something else should be the problem, not the validation of the code. Hopefully, my site will be back in SERPS soon, since 90% of my clients come from the site.

I can't do anything except to wait and pray now...

Regards, Zoreli

Yes, you can do much. You can get IBL's and write new fresh content. You can look for new business models and keep up with new technologies.



The reason I have not used CSS before, because it eats up a lot of time. And after Technical & SEO copywriting, CSS formatting is require extra time & care. Therefore, I usually avoided it if I could. This time I am working with CSS for the www.colorexpertsbd.com site.

Yes situations become better using Valid CSS. And valid CSS does not make thing worse. I was actually talking about the effective way to use it.

As a business graduate I emphasis on only on marketing efforts. I don't know what DOM inheritance tree is and I am quite familiar with dynamic webpages.

Thanks for the post.

I agree that time can be an argument and in my view the only argument, not to validate and use CSS.

The DOM tree is the datastructure for the document object model. I is the nodes of this tree you manipulate with technologies like AJAX etc.

On my site, DigitalStart.net (http://www.digitalstart.net/) you find the following cite:

"The problem is that PHP is just too easy. It tempts you to try out your ideas, and flatters you with good results. You write much of your code straight into your Web pages, because PHP is designed to support that. You add the heavier code to functions in library files, and before you know it you have a working Web application. You are well on your way to ruin. You don't realize this, of course, because your site looks fantastic. It performs well, your clients are happy, and your users are spending money."

Read more about that story there. The point is that you think it is so easy, that you do it wrong from the start. I use to say, the top of laziness is to do everything correct from the beginning.

Don't build a megastructure, that collapses sooner or later. Invest a little time in sound technologies, markup and coding before you rush ahead.

By separating, content, code and styling, it is much easier to update your site in the future, and you make life easier for the SE's.

[Peter (IMC)]

resumé as in abridgement, abstract, summery

Not the list of previous jobs and schools and universities you attended... :)

Since english still isn't my native language, I guess I pick the wrong words sometimes.

Where do you have your IT education from?

When is the next time you come here and tell us how to cook spegetthi?
Are all norwegians as polite as you are?

I guess in this thread my participation is not appreciated. I'll leave it to you all.

[kgun]

I wrote that to you because I know you stand it. I hope they are more polite than me.

I am a true amateur compared to many of them. You should hear some of them on a large OOP project and you understand what I mean.

This thread is about CSS, but the problem is the same. May be I misunderstood you, but IMO you should

Never defend invalid markup and sloppy code, even if it is OK for the SE's. If the Bots of the SE's rule your programming developement, you may soon end up with a megastructure that collapses. See my post above.
In extreme programmming XP, you even test your code before you use it (http://www.simpletest.org/). That is a sound principle.

Personal advice:
Never defend bad coding and markup.

A space ship, with or without people, worth billions of USD, can end up in the ocean if you defend / attack people instead of arguments. (http://www.webproworld.com/viewtopic.php?t=72872)

Read that thread if you have time. It is not long. Then come back and tell me if you agree or not.

[zoreli]

Hi


Yes, you can do much. You can get IBL's and write new fresh content. You can look for new business models and keep up with new technologies.


Thanks for the advices. I am working on both fields on daily basis. Hopefully that will help my site to get back in serps. Currently I am still loosing my rankings for all my keywords (greek and english).

This day looks like nightmare for me...

Regards, Zoreli

[kgun]

Writing tests before coding, is extreme programming in a nutshell (http://www.extremeprogramming.org/).

"Let's begin with a simple question:
What is XP? As you will see, it is a deliberate and disciplined approach to software development".

As I told, that is a sound principle. Once you understand and learn it, you will accelerate your programming skills.

Litterature:
See the last link in my signature (you find some of my favourites there in the upper right corner, PHP, XP and design patterns in one book)

or go to the

Addison Wesley Benjamin Cummings (http://www.aw-bc.com/)

and search for extreme programming.

[Seopro]

I don't have time????? When it gets to the point that I do not have time to do things right for a customer it is time I get out of SEO and Design. If I had to change all of the font color for a client from black to dark blue I go to the CSS file and under body I change the color from #000; to #00F;
and that takes less than a minute. How long will it take you on a 100 page site.
We have only seen a pin point of what the web holds and those of us that write clean and valid code are the ones that will prosper in the future.
Don
Seopro

[activeco]

Gentlemen,

Let me remind you that this is the "Search Engine Optimization Forum" and not a section about "Best practices of web design in general".

Coding according to standards, should be promoted, of course, but many have experienced that most search engines don't put too much emphasize on validation, unless there is some serious problem involved which could confuse bots.

The OP lost his good ranking after implementing CSS.
Why is that? Is CSS related at all?
Is his CSS layout so wrong that bots can't follow it completely?

I still believe it is a kind of major content change which CSS produced (as viewed with Lynx), supplemented with REAL small content modifications.
Add some bad coding and it should be enough for big problems.

[Seopro]

Activeco
This thread started with the question of switching from a table based layout to a CSS layout. The question proposed is does sloppy code have an effect on the spiders and therefore on the SEO of a site. My contention is that if a spider can not properly read your code does it skip over it thus eliminating some of your content giving you lower rankings. I think this is something we can not answer but why chance it or promote sloppy code.
Not haveing enough time to do it right thus effecting the SEO of your customers is something that has to be considered.
Thank You
Don
Seopro

[Webnauts]

However for several reasons based on searches I did trying to get inside the SE's heads, i thought the H1's might be hurting me so I removed them.
Are you crazy?

I would replace for example this:

<div id="header">

Rustic Stone, Home Decor

and So Much More </p>
</div>

with

<div id="header"> <h1> Rustic Stone, Home Decor

and So Much More </h1>
</div>


As soon as I did I saw some of my more important rankings returning. Not all of them, and it could be a coincidence, but who knows

I put them back...and a day later I was gone. So...I took them down again, and for now they are coming back a little.

It may be that me suddenly adding h1 to 700 different pages all at once caused google to think i was h1 stuffing or something, and that combined with all the other changes flagged me as a spammer

whatever it is, everything I do seams to be working opposite to how it is supposed to these days.

Do you expect that Google checks your changes instantly? Such changes require a while, until Google approves the, I know cases that they took several months.

I saw in this discussion that many here mentioned that valid code is important. Since when? I am surprised to hear that!!! Or did I misunderstand something?

I don't miss this article: http://xhtml.com/en/css/css-and-accessibility/

I will help you seriously in terms of SEO.

[kgun]

I saw in this discussion that many here mentioned that valid code is important. Since when?

John, so you recommend invalid code?

GoogleBot or other SE bots shall not teach me how to cook spaghetty.

Look at the bottom of that link.

W3C XHTML 1.1 Valid: Validates.
W3C CSS valid: Timeout.
W3c WAI-AA WCAG 1.0

[Webnauts]

I saw in this discussion that many here mentioned that valid code is important. Since when?

John, so you recommend invalid code?

GoogleBot or other SE bots shall not teach me how to cook spaghetty.

Look at the bottom of that link.

W3C XHTML 1.1 Valid: Validates.
W3C CSS valid: Doesn't function in IE 7.0 with my options.
W3c WAI-AA WCAG 1.0
Did I ever recommend invalid code? Or is that a joke? Don't you know my web sites man?

[kgun]

Did I ever recommend invalid code? Or is that a joke? Don't you know my web sites man?

I know, but your comment / answer could be misunderstood. I wanted you to clear any such misunderstanding.

[greeneagle]

One more attempt...

From a purely logical perspective applied toward the "Logic Machines" (SEs): Converting to CSS from tables or a combination thereof, should not "destroy" rankings (SERP), alone.

IMO - Nothing contrary to that "fact" has been shown here, and most likely cannot be shown here or anywhere else.

IMO (again) The SE's are well beyond that diminished status in their current developmental position.

Obviously, the entire "play" hasn't been represented in this scenario here.

There is a great deal of emphasis on SE maturation in the form of millions of dollars (USD primarily) constantly endeavoring to progress (R&D) by the SEs.

Or do we just not understand the "laws of logical progression" at all, in any way here, and what about the "milestones" we have witnessed along the way?

LOL - It doesn't seem that some here like me to post symbolic tunes, although realisticly synchronis to 'EXACTING" positions here when I do, but if anyone here feels that the SE's are left that far behind, just go ahead here: It's a Dream" (http://www.npr.org/templates/story/story.php?storyId=4850911).

There are different "trains" of thought, and everyone has to ride their own.

Ken

[activeco]

From a purely logical perspective applied toward the "Logic Machines" (SEs): Converting to CSS from tables or a combination thereof, should not "destroy" rankings (SERP), alone.

IMO - Nothing contrary to that "fact" has been shown here, and most likely cannot be shown here or anywhere else.


I am afraid I have already shown it here, but some people are lazy or unable to try it themself or even can't read through the thread.

There is a difference in a way most robots and humans see a page. A page visually identical to us could be very different in robot's eyes.
With *nix' simple browser Lynx it is possible to experience the closest view of a robot.

Here is an example of two identical pages, one coded in pure html and another one using CSS:
Html page (http://www.digital-web.com/extras/reconstruction_with_css/before/index.html), CSS page (http://www.digital-web.com/extras/reconstruction_with_css/after/index.html)

For those unable to use Lynx, here are copies of the output:

HTML:


################################################## ###########
Digital Web Magazine The web designer's online magazine of choice.

Home Search Newsletter Contact About

tutorial

The Heading for the Article Goes Here

By Name Goeshere

Sed ut perspiciatis, unde omnis iste natus error sit voluptatem accusantium
doloremque laudantium, totam rem aperiam eaque ipsa, quae ab illo inventore
veritatis et quasi architecto beatae vitae dicta sunt, explicabo. nemo enim
ipsam voluptatem, quia voluptas sit, aspernatur aut odit aut fugit, sed quia
consequuntur magni dolores eos, qui ratione voluptatem sequi nesciunt, neque
porro quisquam est, qui dolorem ipsum, quia dolor sit, amet, consectetur,
adipisci velit, sed quia non numquam eius modi tempora incidunt, ut labore
et dolore magnam aliquam quaerat voluptatem. ut enim ad minima veniam, quis
nostrum exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid
ex ea commodi consequatur? quis autem vel eum iure reprehenderit, qui in ea
voluptate velit esse, quam nihil molestiae consequatur, vel illum, qui
dolorem eum fugiat, quo voluptas nulla pariatur?
__________________________________________________ _______________

John Smith is a Web designer, developer, author, illustrator and Flash
ninja. Be scared. Very scared.


contribute | send feedback

Back to top
Copyright © 1994 - 2004 Digital Web Magazine. All Rights Reserved.

Daily
What's New
Articles
Features
Tutorials
Interviews
Columns
Art of Interaction
Wide Open
IAnything Goes
Keep it Simple
$ and Sense of IT
Reviews
Book Reviews
Product Reviews
Miscellaneous
Profiles

Advertisement
Mindjack
Advertise Here

Cool Homepages

Advertise Here
################################################## ###########



CSS:


################################################## ###########
Digital Web Magazine

The Web designer's online magazine of choice.

Site Navigataion
* Home
* Search
* Sitemap
* Contact
* About

Inside this Issue

Daily

* What's New

Articles

* Features
* Tutorials
* Interviews

Columns

* DigiSect
* ProDotCon

Reviews

* Book Reviews
* Product Reviews

Miscellaneous

* Classifieds
* Resources
* Profiles

Features

The Heading for the Article Goes Here

By Name Goeshere

Sed ut perspiciatis, unde omnis iste natus error sit voluptatem accusantium
doloremque laudantium, totam rem aperiam eaque ipsa, quae ab illo inventore
veritatis et quasi architecto beatae vitae dicta sunt, explicabo. nemo enim
ipsam voluptatem, quia voluptas sit, aspernatur aut odit aut fugit, sed quia
consequuntur magni dolores eos, qui ratione voluptatem sequi nesciunt, neque
porro quisquam est, qui dolorem ipsum, quia dolor sit, amet, consectetur,
adipisci velit, sed quia non numquam eius modi tempora incidunt, ut labore
et dolore magnam aliquam quaerat voluptatem. ut enim ad minima veniam, quis
nostrum exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid
ex ea commodi consequatur? quis autem vel eum iure reprehenderit, qui in ea
voluptate velit esse, quam nihil molestiae consequatur, vel illum, qui
dolorem eum fugiat, quo voluptas nulla pariatur?
__________________________________________________ _______________

John Smith is a Web designer, developer, author, illustrator and Flash
ninja. Be scared. Very scared.

Participate
Contribute works Send feedback Report problems

Back to top

Copyright © 1994 - 2001 Digital Web Magazine. All Rights Reserved.
################################################## ###########



As you can see, the content changes are dramatic from a robot's perspective, enough to make a vast difference in rankings.

I would advice anyone making such a move, to try to make CSS layout closer to the robot's view and then to gradually change it 'back' to the visually equal appearance.

[Webnauts]

Did I ever recommend invalid code? Or is that a joke? Don't you know my web sites man?

I know, but your comment / answer could be misunderstood. I wanted you to clear any such misunderstanding.
I go with this article about this issue: http://www.phoenixrealm.com/2007/03/05/why-clean-markup-matters-to-your-seo/

[greeneagle]

activeco,

That is an informative post which deserves strong consideration, however there is a lot more to the algs than "pecking order" IMO.

Ken

[Seopro]

Greeneagle,
This was a very interesting subject that started out by me not explaining myself well enough. I think that anyone starting out in CSS should at least validate their code to find any major errors. My code usually validates because I do not know how to use an editor and perfer notepad anyway. I am a disabled vet so I have nothing else to do but work 7 days a week.
Sometimes my code does not validate due to a program I have on it that is not XHTML but nothing to worry about. I just feel that sometime in the future valid code will really come in handy when you are working with forward compatibility.
Everyone has their own opinion and that is what makes a forum like this worth its weight in gold.
Thank You
Don
Seopro

[greeneagle]

Seopro,

This is one of the more interesting topics here in a while IMO. It indeed deserves consideration from several perspectives.

I am in no way demeaning any stated position.

I'd like to encourage anyone else with knowledge and/or experience here to continue on.

Ken

[activeco]

however there is a lot more to the algs than "pecking order" IMO.


Definitely true.

[Seopro]

I took one of the sites I was working on that had used "class" 8 times on the page and turned them into "id" just to see what happened. Needless to say my site fell apart and my spider would not read the major mistakes. I am not talking about using a depreciated tag because the search engines will read that but does a search engine ignore a major mistake thus decreasing your content?
This was asked in Matt Cutt's blog and when he finally answered all he said is everyone should use valid code. That really did not answer the question but then again maybe it did.
For every example that shows valid code outranking sloppy code you can find and example going in the other direction. This is a discussion that could last for years and never answered with concrete proof one way or the other. The only ones that know are the engineers and they don't talk. I do think if an engineer was to have a blog on exactly the way a search engine works we would all sit back in awe. Of course as soon as he said something everything would change anyway.
Don
Seopro

[Webnauts]

This is the appropriate code structure of the body of a page, which you can achieve with CSS, for enhanced Accessibility and SEO:

1. Logo
2. Skip links
3. Main Header (h1)
4. Main Navigation
5. Breadcrumbs
6. Main Content
7. Sub navigation(s) of other site features (like ads, etc)
8. Footer

Just my two cents.

[kgun]

John, sounds like a Dreamweaver template. I know your answer. LOL

[activeco]

This is the appropriate code structure of the body of a page, which you can achieve with CSS, for enhanced Accessibility and SEO:

3. Main Header (h1)


Probably very true.
The OP stated experiencing less problems after removing H1, which could be due to H1 actually being positioned in the middle or end of the page, thus screaming "spam, spam...".

[Seopro]

My CSS content goes in a different order.
head
content
nav
footer
header
If the search engines want content why not give it to them first?
Don
Seopro
My family is from Alesund we do things backwards.

[kgun]

My family is from Alesund we do things backwards.
Ålesund Western Norway?

Those people are very clever on buiness. Some of the best in Norway com from Møre, the region of Ålesund.

And very friendly people. Great nature there, so you should visit Ålesund if you get time. I have not been there myself.

[Webnauts]

My CSS content goes in a different order.
head
content
nav
footer
header
If the search engines want content why not give it to them first?
Don
Seopro
My family is from Alesund we do things backwards.
My structure is based on studies & Web Standards for the present and the future.

[Webnauts]

John, sounds like a Dreamweaver template. I know your answer. LOL
But others do not know my answer:

I hate WYSIWYG Editors. I use a text editor for coding per hand.

I hate foreign help or code! ;)

[activeco]

My structure is based on studies & Web Standards for the present and the future.

Finally someone worth of doing business with.
If you can predict the future, the sky is the limit for us. ;)

[Webnauts]

My structure is based on studies & Web Standards for the present and the future.

Finally someone worth of doing business with.
If you can predict the future, the sky is the limit for us. ;)
No one needs to be a prophet to predict the future.
Just read for example the quality/technical guidelines of the search engines, and do exactly what they say.

There are many here and elsewhere telling that Search Engines, or lets say Google, don't care about valid code.

Maybe they cannot check that with their algos yet. But one day they will. Otherwise, why do they ask us to do all that?

Are they W3C lovers? LOL

[activeco]

Maybe they cannot check that with their algos yet. But one day they will. Otherwise, why do they ask us to do all that?

Don't get me wrong. I like clean code too, but that's not what a search engine is interested in.
A SE needs a minimum of accessibility for it's robot and a feeling that at least non-broken page is provided to visitors too, including a reasonable uptime presence.

A well designed search engine is in quest for information and authoritative pages in the first place. It is very aware that there are millions of authorities in many fields who's job is not web design and those who don't even care at all about web design and their web presence.

[Webnauts]

I am aware of all that. I just wanted to explain, how can I predict the future. That was all.

[Seopro]

Any new technology that comes out will have to follow the guidelines set by web standards or it will not work. We can predict the future by using this as common sense. That is why I am such an advocate of web standards, if used right it will make your clients happy and save you time. I might be old fashioned but this I believe. Now if you will excuse me it is a long way to the outhouse and it is raining.
Don
Seopro

[Webnauts]

I might be old fashioned but this I believe.

Don you are not old fashioned. You are future fashioned buddy.

Great post by the way. ;)

[greeneagle]

Strange "Bedfellows"!

Sorry, I had to go there!

LOL - Go for "it", guys!

Someone, quick, call the paramedics... we have a "pulse" here.

What are we saying in synopsis here then?

Someone "wrap it", or I'll make a "stab" at it.

Ken

[Webnauts]

This is the appropriate code structure of the body of a page, which you can achieve with CSS, for enhanced Accessibility and SEO:

3. Main Header (h1)


Probably very true.
The OP stated experiencing less problems after removing H1, which could be due to H1 actually being positioned in the middle or end of the page, thus screaming "spam, spam...".
The <h1> headers should be unique for every page, and the first and only one on each page.

[kgun]

The <h1> headers should be unique for every page, and the first and only one on each page.

Good advice and of course a unique descriptive title tag.

[Seopro]

I don't know why I came into this forum in the first place. I gave up forums about a year ago because of the nut cases and if you didnt agree with the moderator you got booted. Now that I am here I found a group of designers that not only know what they are doing but are great people. Now I cant keep out of here because there is so much to learn from everyone.
Thank all of you very much
Don
Seopro

[Webnauts]

This was asked in Matt Cutt's blog and when he finally answered all he said is everyone should use valid code.

Can you tell the link?

[Seopro]

It was sometime last summer maybe August and I didn't think much of it at the time until my wife mentioned it to me while we were working.

[Peter (IMC)]

Here´s an interesting read: http://www.stonetemple.com/articles/interview-adam-lasnik.shtml

Just to get back to that code validation discussion. :)

A quote:

we do not penalize or treat differently those pages whose HTML won't validate very nicely.

Just to make sure,.. I don't say one should write bad code, but I am saying that validated code does not help you get higher rankings.

[Webnauts]

I don't say one should write bad code, but I am saying that validated code does not help you get higher rankings.
Just a side note here. I hope that it is clear here that some markup errors can hinder robots crawling a page though. Which errors I cannot tell, as I have no experience designing web pages with erroneous markup. :)

[kgun]

I don't say one should write bad code, but I am saying that validated code does not help you get higher rankings.
Just a side note here. I hope that it is clear here that some markup errors can hinder robots crawling a page though. Which errors I cannot tell, as I have no experience designing web pages with erroneous markup. :)

LOL I have. But for me 2 + 2 = 4 an not 5 and I would not advice any to say that 2 + 2 = 5.

CSS does not destroy your ranking if it is done correct. CSS is styling and should be separated from content. Period.

[Steven1976a]

SEOpro there is no link in your signature, any chance you could share it with us so that we can learn.

I dont think correct code is the essential point of SEO as i dont believe there is any single point to SEO. If there was, then spammers would abuse it but like all other factors im sure it pays a part and I agree that by not giving valid code it can only hinder the robots indexing files.

A couple of sites that you would expect to pay attention to valid code

www.amazon.com 1000+ errors
www.bbc.co.uk 30+ errors
www.cnn.com 40+ errors (Ranked 1 out of more than 2 billion for the word "News")

[Peter (IMC)]

Yes that is correct, Lasnik explains more about a robots ability to crawl in that interview.

[Jason Tor]

It looks like your SERP's have moved up a little since you posted this.

I would give Google a little more time. Any time you make a huge change to your site Google is going to get suspicious. Your rankings will come up. In the meanwhile you might want to consider anything new that you added to the site when you made the change. H1 tags? Did you use these on the old design? A lot of these little changes can add up and conflict with each other.

Jasontor

[bj]

Did you change the source order in the html? It's possible that may affect the importance the SEs put on certain pieces of your puzzle.

[bj]

A couple of sites that you would expect to pay attention to valid code

Amazon.com: Online Shopping for Electronics, Apparel, Computers, Books, DVDs & more (http://www.amazon.com) 1000+ errors
BBC - homepage - Home of the BBC on the Internet (http://www.bbc.co.uk) 30+ errors
CNN.com - Breaking News, U.S., World, Weather, Entertainment & Video News (http://www.cnn.com) 40+ errors (Ranked 1 out of more than 2 billion for the word "News")

I venture to say that now that IE7 has been released you may eventually see some movement there. All sites of this size and complexity running off huge database backends were waiting for that event before upgrading code. And upgrading code on sites that large does NOT happen overnight.

Which is why I think it's majorly stupid of anyone to point to the big sites and use them as an excuse not to validate their little site. And there will someday come a day when quality of code does make some difference.

[datetopia]

Basically, the most popular websites are not fully W3C HTML compliant. They just use a common, simplified format.
You should stick mainly to writing the main tags correctly as they do and then focus on content and back links.