W3C Valid Code and SEO

[Cricket]

A topic that has been discussed extensively by some people in the SEO community is the notion that having a W3C compliant web site is either critical to SEO or it’s not. Read more: http://www.searchenginejournal.com/w...reality/18566/

Matt Cutts - Yeah, there's so many people that write invalid HTML with syntax errors, that still is good content, that we need to be able to rank that good content even if somebody doesn't, you know, have something that is completely lint free in terms of validation. http://www.webpronews.com/topnews/20...-and-much-more

W3C Compliance - Not an SEO Factor?
http://www.seoconsultants.com/google/validation

W3C Validation for SEO – Myth and Reality
http://www.searchenginejournal.com/w...reality/18566/


After reading the above articles I would like to hear your take on this highly debated topic and why you feel that way.


This thread will be for serious discussions only and will remain respectful at all time.

[Hades]

I am by no means a SEO expert but I do think both compliance and code semantics are important from a Marketing POV.

While W3C compliance may not be a relevant factor to Google it does have a marketing impact on your site, that is: if your site cannot be viewed on some particular browsers, or platforms due to poor standard compliance you will be in disadvantage compared to your competitors who have a site that can be viewed by anyone.

[guy]

It honestly should not be as highly debated as it seems to be. It's a little bit annoying to hear people go on and on about code validity. It's almost like those annoying people that point out you have a grammatical or spelling error in your page long response, when they don't find anything else to critique. (end of off topic rant, woo)

But the thing is, while I believe it's awesome for ones site to validate according to the W3C, I'm more concerned about things like, browser computability. People browsing my site are not going to care if I've ignored that lonely double quote or didn't close that crummy tag, they care more about how the site looks and functions in whatever browser they're using.

While it's awesome to make valid code, which is in truth easier to maintain if you know all the nesting is in order, Google and other search engines crawl and index and return results on the web even if your pages don't validate. There's no special rank treatment to pages that do validate...

So I personally spend my time making sure my sites are compatible across as many browsers as possible and that they're as user friendly as possible, rather than whether the pages validate according to the W3C. That's not to say I don't 'try' and keep my code valid, but it's more to do with sleeping better at night knowing my code is all sexyfied than anything else because of the type of coder I am (and also so I can stick those cute pointless 'this site is css/html valid' in my footers

[mjtaylor]

Great topic, Cricket, and a wonderful opportunity to tickle the purists. Of course, from a usability standpoint, and from a professional standpoint, it makes sense to make sure a site has valid code.

Is it a ranking factor? I see Edward Lewis' point: Cutts didn't really say it isn't. It is logical that it could be a minor factor.

The real issue is: can the content of the site be fully spidered and indexed? If the content is accessible to the bot, I believe the SEO factor should be, and probably is, negligible.

If there are errors that are so bad -- such as broken links -- that Google tells me about them in the Webmaster Tools, I had better clean those up, but the closing paragraph tags? Nah!

I see the Alan Bleiweiss' point here. In most cases - we are talking inherited sites here -- there are more important, more influential factors that should get our attention first.

I'm going to get some popcorn now, and sit back for the show.

[guy]

Originally Posted by Hades

if your site cannot be viewed on some particular browsers, or platforms due to poor standard compliance you will be in disadvantage compared to your competitors who have a site that can be viewed by anyone.

Hey Hades, first, I agree with you that code semantics are important, totally. But with respect, your above quoted statement is not completely accurate. Sometimes, even a page that Validates according to the W3C ends up being none browser compliant. Sometimes you have to edit a page in a way that it no longer validates according to the W3C for it to truly become Browser Compliant.

Same thing goes for those that say, "if you validate your page then it'll be faster" - not entirely true either. Sometimes you need to cut some corners that don't effect browser compatibility in order to reduce the size of the page and load time etc. When you then take this page and validate it, it no longer becomes browser compliant. Tricky business, so, seeing as how it has no factor on SEO, why should I worry about it? Do you get what I'm trying to get at?

[Hades]

Originally Posted by guy

Hey Hades, first, I agree with you that code semantics are important, totally. But with respect, your above quoted statement is not completely accurate. Sometimes, even a page that Validates according to the W3C ends up being none browser compliant. Sometimes you have to edit a page in a way that it no longer validates according to the W3C for it to truly become Browser Compliant.

Same thing goes for those that say, "if you validate your page then it'll be faster" - not entirely true either. Sometimes you need to cut some corners that don't effect browser compatibility in order to reduce the size of the page and load time etc. When you then take this page and validate it, it no longer becomes browser compliant. Tricky business, so, seeing as how it has no factor on SEO, why should I worry about it? Do you get what I'm trying to get at?

I get what you say and I agree on the fact that standard compliance is no guarantee of compatibility.

But it's the closest you can get. By making your pages be both standard compliant (by that I mean choosing the right schema and writing the right code for it) and built with good semantics (by that I mean using the proper code for the each job) you will have the approximate certainty that if a page does not display correctly in a browser it won't be the HTML's fault but:

• the browser's fault,
• or bad CSS,
• or bad JS

... and then all you will have to do is look for small workarounds for those issues.


Again it's no guarantee but it helps your chances

[mjtaylor]

When I saw the article By Ed Lewis, I sent a link to Mike McDonald the Web Pro News Interviewer referred to in Ed's comments. I have his permission to post his response here:

Have to say, I was a bit perplexed when Ed decided to take shots first at me for 'leading' Cutts in a discussion about W3C then have a go at WebProNews as being poorly optimized and impugning our reputation a little.

I never intended to 'lead' Matt Cutts in any reply re: W3C validation. I am altogether unconcerned with the fact that WebProNews has some validation errors. That's because most sites that rank well for competitive terms also have W3C errors.

W3C Validation is NOT an SEO factor. I really don't know how much more plainly Cutts could have said so in the video… it's right there to watch. Of course he isn't going to come out and say 'don't validate W3C' -- neither would I. If that is something you like/want/feel the need to do, then by all means, go for it.

I'm a big fan of real life results and taking long hard looks at the facts. So, as much as I hate to put 'ol Ed on blast… he took the first shot, so I figure it's like my grandfather always told me; "Don't stick it out there if you're afraid to get it cut off".

Let's look at the aforementioned Results and facts. Shall we?

We'll start with Edward's site: pageoneresults.com. I didn't check to see that it's validated W3C, but I figure that's a pretty safe assumption at this point.

So how's Ed doing for his key terms (oh look… he has META keywords… how cute). Here are Ed's Meta Keywords (Meta keywords, Ed? Really?)
Search Engine Marketing,
SEM,
Search Engine Optimization,
SEO,
Edward Lewis,
pageoneresults

So, now, let's check and see how many results Ed has for his chosen terms in Google's top 100 results.

Here they are:

pageoneresults

Not a typo. I listed all of his top 100 results right there. That's right. Ed ranks third for Pageoneresults - his domain basically and only if you type it in as all one word. I hate to poke fun (ok, maybe that's not true) but Ed isn't even pulling a top 100 rank for Edward Lewis. How is that even possible? I mean, WebProNews is number 8 for Mike McDonald and I don't even own the site.

So I'm writing this and I'm thinking to myself; 'self, you're being too hard on Ed. You should try looking for some other terms his site might rank well for. Those, after all, are awfully competitive.' So I did. I scoured the Google top 100 for:

w3c validation
SEO tutorials
SEO News
W3C SEO
SEO Marketing
SEO Marketing News
SEO tips
SEM Search Engine Marketing
SEM info

Gosh…he isn't top 100 for ANY of those either (WebProNews, incidentally is top 100 for about half of those - including 'W3C validation' in the top 20) --- Oh the irony! Not to mention the fact that the VAST majority of the sites ranking on the first page for these terms - you guessed it - fail W3C validation.

Unless, I'm missing something here, that W3C just really isn't bringing home the bacon for you there, Ed. I mean, I'm getting a top 20 result for W3C Validation with my jacked up, non-validating, sloppy ol' WebProNews page and you are nowhere to be found (psst… maybe try adding it to your META Keywords). I'm sorry. But yes, I'm LOLing a little here.

But now hang on a minute… before I get too carried away… Fair is fair and all. Maybe I should look at WebProNews and subject it to similar scrutiny. Now, we don't have a meta keywords tag (mostly because they were pronounced dead, like, 5 years ago…) but we do have some competitive terms of our own that we would like to result for.

Like:
ebusiness news
search engine news
search industry news
seo videos
google seo
seo tactics
search strategy
paid search

Now let's look at how many of those we manage a top 100 result for:

ebusiness news
search engine news
search industry news
seo videos
google seo
seo tactics
search strategy
paid search

Well, gosh… That's all of 'em. If you take those and add to them the 4 or 5 terms out of Edward's list that he doesn't rank for (but we do). Well, I'd call that a pretty good showing from sloppy old WebProNews with it's non-validating nastiness.

To be honest, I don't see how anybody could watch the video, listen to the question and answer and come away with anything other than what was actually said. It isn't going to hurt you but it isn't going to help you. Matt, nor I would suggest anybody abandon W3C if that's something they want to have on their website but I don't see how much more clearly it could have been stated that W3C validation is NOT an SEO factor.

Do some queries for competitive terms. You tell me how many you find in the top 10 that do and don't validate.

In case you want to watch the video and see what Matt Cutts actually says, it's here: http://www.webpronews.com/topnews/20...-and-much-more

[evetorres]

A W3C validation ensures that all your webpages are free from coding errors, deadlinks & broken links. A W3C logo on your website indicates that you are W3C certified & your website is completely harmless. This helps improve your online reputation. Technically since all errors are removed the download time of your pages are decreased, there is smooth flow of data & navigation,

[HTMLBasicTutor]

If good coding has no affect then why does Google recommend it here:

•Check for broken links and correct HTML.

http://www.google.com/support/webmas...answer=35769#1

and here:

Write good, clean HTML

While your site may appear correctly in some browsers even if your HTML is not valid, there's no guarantee that it will appear correctly in all browsers - or in all future browsers. The best way to make sure that your page looks the same in all browsers is to write your page using valid HTML and CSS, and then test it in as many browsers as possible. Clean, valid HTML is a good insurance policy, and using CSS separates presentation from content, and can help pages render and load faster. Validation tools, such as the free online HTML and CSS validators provided by the W3 Consortium, are useful for checking your site, and tools such as HTML Tidy can help you quickly and easily clean up your code. (Although we do recommend using valid HTML, it's not likely to be a factor in how Google crawls and indexes your site.)

http://www.google.com/support/webmas...?answer=100782

BTW: The editor of the HTML 5 spec is Ian Hickson, Google, Inc. http://dev.w3.org/html5/spec/Overview.html

While standards compliant, valid coding may not be the end all for SEO and cross browser compatibility, think about it, are the search bots going to spend a whole bunch of time (plus wear and tear on their machines) trying to decipher your bloated, out dated, non compliant coding when they can zip through a cleaner coded site? Not likely.

No matter how you want to twist Matt's words,
having a site that loads fast because you have less coding to process (besides other things), is cross browser compatible has a built in SEO benefit. People link to your site without being asked to because they like it which in turn equates to a natural incoming link. Thus you have a "vote" to improve your standings in the results.

Just because Mr. McDonald wants to follow what Google evaluates doesn't make meta tags invaluable. Other search bots do take these into consideration so there's no harm including them.

[HTMLBasicTutor]

Originally Posted by evetorres

A W3C logo on your website indicates that you are W3C certified & your website is completely harmless

This is not true.

The logo on a site only shows the visitors that you have passed the W3C validation tests. They do not check that the site is harmless.

The W3C "valid" icons may be used on documents that successfully passed validation for a specific technology, using the W3C validation services. The image should be used as a link to re-validate the document.

How to use the Validation Icons

[HTMLBasicTutor]

Originally Posted by guy

Sometimes you have to edit a page in a way that it no longer validates according to the W3C for it to truly become Browser Compliant.

No you don't.

You can make a valid standards compliant site that is cross browser compatible. Ya just need to know how to get it to work without hacking it.

[guy]

Originally Posted by HTMLBasicTutor

While standards compliant, valid coding may not be the end all for SEO and cross browser compatibility, think about it, are the search bots going to spend a whole bunch of time (plus wear and tear on their machines) trying to decipher your bloated, out dated, non compliant coding when they can zip through a cleaner coded site? Not likely.

The majority of websites online, that Google crawls and indexes, are in fact, not at all standard compliant, and some, if not most, are in fact outdated. Even most of those that are not outdated are not Standard Compliant so while having broken links and missing alt tags might hurt your chances in ranking, it has nothing to do with W3C compliance. There are simply a couple exceptions that both help SEO and are variables in Stands Compliancy. That doesn't make having Valid coding a must in any way. Google, (in my opinion) simply barks about it to make life simpler for itself (i could explain what I mean about that but I'll try and start a new topic later next week to touch up on some of those points.)

[guy]

Originally Posted by HTMLBasicTutor

No you don't.

You can make a valid standards compliant site that is cross browser compatible. Ya just need to know how to get it to work without hacking it.

While I totally respect your opinion HTMLBasicTutor, can you do me a little favor, and go validate www.google.com for me

Make it standard compliant, and have it cross browser compliant, make sure it works on mobile devices too, etc etc. Keep the page size as it is, too. Oh, and I'll give you a month to do it.

Some times, no matter how you look at it, things come down to one thing, user experience. If I'm to tailor a site to validate rather than tailor it to function well and provide the most awesome user experience ever, then why the heck am I doing it in the first place?

[Cricket]

First, I want to take a moment to thank everyone for their involvement in this discussion.

These are the types of passionate and informative discussions I would love to see more of in this community. If we all agreed on every topic, it would be boring and we would never push the boundaries of what we believe. Heck, we wouldn't even be a community in the true sense of the word, because communities are built on the efforts of an entire team and their thoughts.

I was going to stay out of this discussion all together, and just make sure that the topic remained friendly, but I decided to jump in here for a moment and share my thoughts.

I really believe that we are all saying much of the same thing, that what we are actually debating is semantics.

• Can clean code help your site load faster? Yes. Is this possibly part of SEO? Yes. Can you do well in the SERPs without it? Yes.
  
• Is the proper use of the alt attribute part of valid code? Yes. Can it help with SEO? Yes. Can you do well in the SERPs without it? Yes.
  
• It the proper use of heading tags part of valid code? Yes. Can it help with SEO? Yes. Can you do well in the SERPs without it? Yes.
  
• Does your site have to be valid to do well in the SERPs? Heck no, but many parts of a valid site can help your SEO efforts.

In the end, SEO is just many pieces of a puzzle. If you figure out enough of those pieces, you are going to do well. If you leave out too many of the pieces, you may do poorly.

Just my 2 cents, and I will even provide change.

[HTMLBasicTutor]

This is how happy people are with the new "user experience" at Google: http://www.v7n.com/forums/google-for...yout-back.html

Not going to debate with you here on the merits of standards compliant coding vs user experience.

The topic of this thread it standards compliant coding (a.k.a. valid coding) and SEO.

IMHO - it helps. It's not the be all end all solution of doing well in any of the search engines but as with any search engine optimization techniques, every little bit helps and they all work together, not individually.

[MikeMcD]

Originally Posted by HTMLBasicTutor

Just because Mr. McDonald wants to follow what Google evaluates doesn't make meta tags invaluable. Other search bots do take these into consideration so there's no harm including them.

Hello folks. Had to throw in a comment on this one. Couldn't lay off. W3C is one thing but META Keywords? C'Mooon.

Check this out
And this
Because Meta Keywords...
Are COMPLETELY dead. Dead dead.
Used by NOOOOObody and nothing
- human or otherwise.

[HTMLBasicTutor]

Welcome to the forum Mr. MacDonald.

Not to hijack this thread and deviate from the original topic of the thread, I respectfully present the following to back up my statement:

•Use a "keyword" metatag to list keywords for each page of your website. Use distinct lists of keywords that relate to each specific page on the site. Do not use one broad set of keywords for the entire site.

How to Improve the Position of Your Website in Yahoo! Search Results Last Updated: February 24, 2010

So in a nutshell, written in the form of naturally spoken language, use your keywords in the following areas for the greatest impact:
...
4. Meta tags (specifically when used with the name="description" and the name="keyword" parameters)

Put your keywords where the emphasis is (SEM 101) - Bing Community Webmaster Blog May 20, 2009

If meta keywords have no value, why are the following still using them:
Search Engine Watch

http://blog.searchenginewatch.com/100519-142531

Article published today.

Michael Gray

http://www.wolf-howl.com/tools/how-to-ftp-upload-files-from-an-ipad/

There are evidently 2 camps on this issue still and we evidently are not in the same camp.

[pageoneresults]

I'm going to get some popcorn now, and sit back for the show.

Good day everyone! You knew I was coming, didn't you? I woke up this morning and my ears were singed!

But the thing is, while I believe it's awesome for ones site to validate according to the W3C, I'm more concerned about things like, browser computability.

My question would be, how do you ensure browser compatibility without validating the markup being used? The two work hand in hand. Oh wait, you mean what you see in the browser, that kind of compatibility?

Google and other search engines crawl and index and return results on the web even if your pages don't validate. There's no special rank treatment to pages that do validate.

Are you 100% certain of that? Have you performed the research and testing to substantiate that statement?

So I personally spend my time making sure my sites are compatible across as many browsers as possible and that they're as user friendly as possible, rather than whether the pages validate according to the W3C.

You really can't do one without the other. Not unless you are some sort of code guru who fully understands markup and can write valid code out of the box. For example, how do you catch 35 missing alt attributes in a streamlined fashion? Your Editor may flag them for you or a validation routine will catch them. So, to say that you spend your time making sure your sites are compatible across browsers and are user friendly yet dis validation, confuses the heck out of me.

The real issue is: can the content of the site be fully spidered and indexed? If the content is accessible to the bot, I believe the SEO factor should be, and probably is, negligible.

Yes, that would be the primary concern when reviewing crap code. Ever wonder what a UA has to go through when processing invalid markup? Do you know which errors the UAs can recover from easily? Do you know which errors they don't recover from so easily?

I see the Alan Bleiweiss' point here. In most cases - we are talking inherited sites here - there are more important, more influential factors that should get our attention first.

I find it difficult to perform any type of quality work knowing that the platform I'm working with is broken. There is no reason for markup to be broken. This is not something that a client should be addressing after the fact. Also, it isn't difficult to cleanse a site and bring it up to standards.

We'll start with Edward's site: pageoneresults.com. I didn't check to see that it's validated W3C, but I figure that's a pretty safe assumption at this point.

This is the part that made my day, thank you Mike! Apparently you didn't check a few things. When I went into semi-retirement a couple of years ago, I removed all the content from that site and disallowed it via robots.txt. It should only rank for the term pageoneresults as you so eloquently found out.

Well formed valid markup is probably the single most important thing you can do for a site.

Now I'll wait for all the replies where people will argue their points for writing broken crap code, it happens every single time, history proves it. I'll let you debate the finer points of broken code and then I'll come back and drop some bombshells to keep the topic fresh. I love these types of discussions.

[mjtaylor]

Originally Posted by pageoneresults

Well formed valid markup is probably the single most important thing you can do for a site.

I was almost believing you until I got to this line.

Hyperbole always stops me dead.

If we're talking the most valuable thing you can do for a site, I will vote for creating unique content.

The problem with these sorts of discussions is that the points cannot be determined with any certainty. We don't know whether valid code is a ranking factor. Only Google knows. And while it's easy to skew Matt Cutts' comments to Mike McD to either perspective, nothing Cutts said can be construed to suggest that it is a significant factor.

Does it make sense to have valid code? Yes. Single most important thing? You're welcome to your opinion, but I'm finding that statement pretty ridiculous.

I'll go with it's possible that it's a minor factor, but I won't bet on it.

[Webnauts]

Valid code is not a ranking factor for Google. But correct and clean code is a requirement of the Google Webmaster Guidelines.

There are HTML elements that are included in the Google rankings factors, like titles, headings, bold and italic tags, alt attributes, etc.

I still recommend to check your code with a validator to make sure that you do not have fatal errors that can choke the crawlers.

For example, one fatal error is BOM that could choke a crawler.

For the Web Standards Advocates here, I would like to add here that I am a Fanatic Web Standards Advocate, but still I must live with the fact that Google claims to not give a damn about valid code.

But, if you are using RDFa for Rich Snippets as I already do on my web site, if that code does not validate, google will not be able to parse it. Strange, or?

They even have a tool so we can test our work. For an example one of mine:
http://www.google.com/webmasters/too...torials/&view=

Nice thread by the way.

John