Hi guys,

Obviously one of the ways to attract new business is through a successful SEO campaign.

I've done a lot of reading and in my opinion, it helps to have a site that has clean markup.

I joined the eBay partner program which allows you to display eBay advertisements on your site. Unfortunately, when you paste their code in, my site no longer validates. I've posted on the eBay forums about this, and this was the response:

"What the does validating have to do with search engine friendly. Don't buy the hype."

My response:

"Every website I visit regarding SEO says code that validates helps in obtaining higher rankings in search engines."

His response:

"BS, IMO. I have over 30K pages indexed, get 5K referrers/day from Google alone, and a new topic on my forum is Google indexed in less than 3 minutes... Google hasn't left my site in 6 weeks. My site far from validates and I don't care."

I had to bite my tongue, because in my opinion, he's wrong.

What is everyone's thoughts?

And yes, I know clean code isn't the only technique ... alt tags, headers, keywords, metatags etc all contribute.

Cheers guys, sorry for the longwinded post.

I'd say the only reason the 30k site is considered by Google is that it's 30k worth of pageweight. That's the only driving factor as to why Google is even interested in the site in question.

Economies of scale present itself in designing a site.

To make a 1k or less worth of pages visible to Google when it neither has the collective history of 30k worth of pages, or the spread of those 30k, is where well formedness and design patterns come into play.

I say this as I've worked on (and remain working on) sites between 5 pages and 30k pages. Obviously a well formed site with 25k worth of history is going to outperform a poorly formed site with 30k.. Why ? Because whilst content is king, letting all forms of search engine (e.g RSS/XML/uF/casting) see that content is where the market can be verticalised.

Twitter writes...

"What the does validating have to do with search engine friendly. Don't buy the hype."

Isn't that correct, so long as invalid code does not prevent a text-based crawler from parsing the site? Markup that only causes visual problems doesn't matter - does it?

Ref. Google Webmaster Guide

Their suggestion to use Lynx to check your site for yourself pretty well underscores that point. But then that's only Google :)

No doubt you've checked out some of their other info already.

tbh i havent seen much of a problem.

I have some terms ranking 1 - 5 with a competition of 39 million.

The website code wise is far from compliant

My sites generally validate successfully until I add things like Amazon Associate code, so it isn't a problem unique to eBay in any way.

That said, I've not really seen it cause me any issues with search engine robots.

The only thing Google cares about is that they're able to spider it. Valid html can assist in that, but only if there was significant structural issues in the html before. Most of the internet is not valid html.

Google's ranking algorithms revolve around determining the quality of what's on the page, not the markup. I would be absolutely amazed if valid html played any part in the ranking given that it makes no difference to information on the page.

Thanks everyone - very helpful.

From what I've read, it basically comes down to:

*Inbound links
*Quality content (well written, obviously including the keywords)
*Titles, alt tags, headers
*Meta descriptions
*A good sitemap
*Keywords (to a *very* small extent these days)

And you can also look at using a robots.txt file.

Am I missing anything?

Twitter writes...

Am I missing anything?

* Appropriate internal linking structure.
* Outbound links.

Would anyone be willing to show me a site they have worked on where the SEO has been a success?

Or name some of the keywords you have had success with?

Thanks for your help guys!

It's debatable whether validation affects SEO on current sites or not. What isn't debatable is the fact that google's algorithms use large generalisations and trends in deciding their SERP's.. if a trend emerges between valid sites and their overall quality of content it's only natural that google will start rating valid sites higher. So even if there isn't any affect to this day, catering for logical progression will leave you well placed in the future.

Just my 2c

It's difficult because there's no certainty.

Noone knows for sure what works, and what doesn't.

For what it's worth, my take on valid/invalid markup in order of priority:
* valid
* nearly valid
* not even close

If you can generate a valid site, do it - you certainly aren't going to be penalised for it. In your case about the eBay code not being valid, don't spend another minute of energy on it - move onto something more important and I assure you there are more important things.

The reason I make a point about the invalid markup is that you can have completely valid (syntactically), however improper use of the HTML elements. As you're no doubt aware, using the appropriate tags (hX for headings, p for paragraphs and so on) makes a difference.

The other reason people may be going on about 'valid markup' is that they associate valid markup with semantically correct use (they are not directly associated), which in the same breath is often associated to XHTML and lean. You need to know the difference, so you can make the right choices for yourself.

Using HTML4.01 isn't a death stroke and neither is invalid markup. XHTML isn't a silver bullet, nor is compliant HTML.

If I were to give you a quick list of things, they'd be:
* Unique titles per page, which use that pages keywords
* If you want a particular SERP snippet, provide your own meta description element. This can be particularly important for CTR (click through rate) for certain sites, so pay attention to what is being listed.
* If you choose to use meta keywords, make sure they are relevant and visible within the page. Generic keywords are a waste of everyones time.
* Aim for lean HTML markup for your site design. While I know you could make the design by nesting 53 tables, opt to lay your site out using div elements and CSS to position everything.
* Semantically correct markup, ie hX for headings and p for paragraphs and so on; this can't be stressed enough.
* Search engine friendly URLs, which should include the keywords/phrases for a page if possible.
* Avoid overly complex URLs with many querystring parameters, 4 is generally considered the limit.
* Provide an XML sitemap, to help the major search engines discover your sites content.

The list just keeps on going and they get progressively more complex as you go along but getting your own house in order is a good place to start, work out from there.

Hi Big Al,

That was very helpful - thank you.

I think I understand what you mean - you could have a site that validates, however the use of the HTML elements could be incorrect.

I spent the weekend combing through my site, and so far, I've:

*Combed through the code to make sure I used HTML as effectively as possible
*Rewritten the page titles (each page has a unique title)
*Rewritten the headers and alt tags
*Rewritten some content to include necessary keywords
*Rewritten the meta descriptions and keywords
*Checked the site in a text browser

I coded the site in HTML/CSS originally and avoided any use of tables.

It's only a small site, so there are no overly complex URLs.

I'm also looking at link exchanging.

Although I've already got a sitemap, it's for the visitors, not the crawlers so I'll look at creating one soon.

I don't think it's necessary for me to use a robots.txt file at this stage.

Thanks for your help everyone, much appreciated.

yeah i'm also started to read about SEO. Any ebooks or books anyone recommend. From what i have gathererd combining PPC + SEO = $$$$$

Could you elaborate on this point Big Al?

"If you want a particular SERP snippet, provide your own meta description element. This can be particularly important for CTR (click through rate) for certain sites, so pay attention to what is being listed."

Cheers!

It's the text search engines and directories can use to describe your page.

Some like Google will provide the text around the relevant portion that matched the search result, others will get their description from the <meta name="description" content="" /> element.

en.wikipedia.org/wiki/Me...iption_attribute

Meta descriptions are ignored most of the time. As long as your markup is STRUCTURAL (rather than decorative) and you use your keywords correctly, your site will index well.

I managed to get my page rank from 0 to 4 in less than three weeks, and number one for most of the search terms I'm targetting (and within the top 5 for those that are not number one) without using any meta data, with a page that has no errors, but doesn't strictly validate.

And this is because my markup is purely structural and weighted correctly.

i agree it dosnt matter if your site validates or not, look at whirlpool its not a valid site (last time i checked)

pixelbunny writes...

Meta descriptions are ignored most of the time

i doubt this... it does play a part , to what extent thats something only google knows.

MigA writes...

i doubt this... it does play a part , to what extent thats something only google knows

I read somewhere google ignores them, but there are other search engines which still consider them

pixelbunny writes...

And this is because my markup is purely structural and weighted correctly.

Any books you recommend? You sound like a SEO guru

SUPERPAK writes...

read somewhere google ignores them,

i would say that the author that wrote it had nfi what he was on about. A client of ours got some SEO done just on titles, keywords and descriptions and it improved their rating and traffic substantially from goggle. Nothing else changed, if goggle did ignore these then that wouldn’t of happened. I would agree that much more weighting is placed on the actual content, but to come out and say that it ignores them with no proof to support the claim and in the face of contradicting evidence means its just nonsense.

Thanks at&me, I know what a meta description is ... I wasn't sure what Big Al was trying to explain.

I'm ashamed to admit I hadn't heard about Google Webmaster Tools - it's very useful.

I've found a lot of the content on the Internet regarding SEO is outdated - from what I know, the importance placed on keywords and meta descriptions by spiders and crawlers is small.

Also, I think the way I was thinking about SEO was flawed. You could have different web pages on the same site targeting different keywords.

Another quick question guys:

Do you use a robots.txt file?

I've used one to block access to some directories and certain web pages I don't want indexed.

I know you can also add code to the web page ... is the robots.txt file enough?

Just a clarification - I did not say "google ignores meta data".

I said "meta descriptions are ignored (by search engines) most of the time".

Too very different statements with very different meanings.

If you want to put in meta descriptions for every page on your site, go right ahead - it can only harm you if the description doesn't match the content.

SUPERPAK writes...

Any books you recommend? You sound like a SEO guru

It's really difficult to tell if you are being sarcastic, so I'll give you the benefit of the doubt. I'm not a guru, simply someone who has been employed in the field of SEO (formally known as "can you get my page to show up when people search for these words?") for the past ten years.

There are no books I have read, so none to recommend. The SEO world is full of people who use shonky practices to gain page rank or higher listing visibility. I do tend to follow the advice of the very good people at ALA and 37 Signals, but other than that it's purely from experience in the field.

Twitter writes...

I know you can also add code to the web page ... is the robots.txt file enough?

Some engines and spiders don't obey the robots.txt file, so it isn't enough to simply put that command into your robots.txt.

If you truely don't want something indexed, use .htaccess to protect it.

Thanks pixel. Besides allowing/denying access to files, is there any point using a robots.txt file?

How did you get into SEO?

SUPERPAK writes...

Any books you recommend?

There is always www.seobook.com.

edit.

Paul v8.1.0240.006 writes...

There is always www.seobook.com.

Recommended. Aaron Wall is far from the best writer but he is a very smart guy...

Twitter writes...

Could you elaborate on this point Big Al?

"If you want a particular SERP snippet, provide your own meta description element. This can be particularly important for CTR (click through rate) for certain sites, so pay attention to what is being listed."

I am not Big Al but I'll give it a try.

Basically what pixelbunny said is correct that meta description tag has no influence on your ranking on SERP (Search Engine Result Pages), i.e. having a keyword-stuffed meta description tag will not improve your ranking from page 3 to page 1, for example.

HOWEVER, Google does show your meta description, and show it as the match-snippet (between page title and URL) on the SERP. For example when you search for "Whirlpool" on Google, the snippet says

A fully independent community site keeping you informed about Australian broadband issues.

which is exactly what's in the meta description tag, and you won't find that text snippet on the Whirlpool front page.

As most people actually read the match-snippet before clicking on the link, what has been presented is actually important as it affects the Click Through Rate (CTR). Big Al's point was that you can actually alter the way Google displays those snippets, by adjusting the meta description appropriately. This will hopefully get more searchers to click on your link.

Thank you for the explanation - that made sense. I thought it did have an impact on ranking.

You're right, I just printed out some of Google's information and I just flicked through it ... they have no impact on ranking.

pixelbunny writes...

Meta descriptions are ignored most of the time. As long as your markup is STRUCTURAL (rather than decorative) and you use your keywords correctly, your site will index well.

They are to an extent but a lot of people seem to forget that Google use a pages meta description for the description of the SE result. So if you have a crappy description, people are less likly to click the result.

Just because you rank highly you will still need to worry about converting the clicks on the results page. Some things you can do to help is make sure your descriptions are keyword rich (NOT SPAMMED)as Google highlight the search term if it shows in the results somewhere.

NotSoMad writes...

They are to an extent but a lot of people seem to forget that Google use a pages meta description for the description of the SE result. So if you have a crappy description, people are less likly to click the result.

Just because you rank highly you will still need to worry about converting the clicks on the results page. Some things you can do to help is make sure your descriptions are keyword rich (NOT SPAMMED)as Google highlight the search term if it shows in the results somewhere.

Agreed. As someone who works in the industry, I take the time to arrange unique description tags for key pages on a site as it's no use ranking #1 if you have a crappy description which people skim and then go to next option as it caught their eye instead of your crap one.

How did you guys get into SEO? Any particular degree?

Twitter writes...

Any particular degree?

Are you talking about uni degrees?

Twitter writes...

Any particular degree?

lmao, a degree in seo, yeh i dont think so, anything you learn in year 1 will be out dated by the time you graduate...

Twitter writes...

From what I've read, it basically comes down to:

*Inbound links
*Quality content (well written, obviously including the keywords)
*Titles, alt tags, headers
*Meta descriptions
*A good sitemap
*Keywords (to a *very* small extent these days)

right on. another significant factor - content must stay fresh. The google cache will match previously indexed content and if it hasn't changed, the bot will visit less.

I was alluding to what degree they completed before they worked in the field of search engine optimisation.

Obviously there are no SEO degrees.

Thanks billfredtom, good to know I'm on the right track. Fresh content ... another good point.

I've printed off all of Google's information ... I didn't realise they had written so much about SEO. Wish I had gone to the source straight away!

Besides blocking access to files and directories, is it worth using a robots.txt file?

Twitter writes...

I was alluding to what degree they completed before they worked in the field of search engine optimisation.

None

Obviously there are no SEO degrees.

There never will be. It's nothing but a bunch of educated guesses.

All the SEO I have learned has been from trail and error on 100's sites over a long time.

Twitter writes...

How did you guys get into SEO? Any particular degree?

I have a double in Internet Systems/Commerce. Majored in webprogramming(i sucked at Java..) & Marketing (wtf are the 4p's again?)

Further education from that involved getting an Advanced SEO certification from www.sempo.org

It covered the basics pretty well and a definately puts you ahead of the competition when trying to get in the industry.

Does anyone know if using server side includes for eg:

<!--#include FILE="blah.html" -->

affects page rankings? Do search engines index the included file at all?

TGM writes...

Does anyone know if using server side includes for eg:

no effect... in the end HTML is generated so thats what it indexes it has no notion of the include script.. the problem is though that you could be repeating the same text in which case you would be penelised as the same text shoudlnt be on every site. but for a footer or hearder theres no issue.

Thanks Miga.

Will only be repeating the same portion of content on two pages out of about 50, and the content itself is only a small portion of both pages. Just want to make it easy for my brother to update without him having to modify both pages when he adds a new 'featured act' :)

edit: come to think of it i might do the same for the header and footer thats on every page as well...

I recommend SEOmoz.org, great SEO downloadable guides for free! You can also get www.seobook.com - either buy the book or use all the tools!

I recommend SEOmoz.org, great SEO downloadable guides for free! You can also get www.seobook.com - either buy the book or use all the tools!

SUPERPAK writes...

yeah i'm also started to read about SEO. Any ebooks or books anyone recommend. From what i have gathererd combining PPC + SEO = $$$$$

for the latest trends in the industry have a look over @ forums.digitalpoint.com