Outdated SEO Coding Theory Debate

[PTC]

I think this was some of the first website related education I ever received (link originally provided by a PAL/PAP member Im sure). Some good stuff here that is not worth discussing. Some stuff related to CSS and crawling efficiency that interests me.

Chuck the SEO Rapper

Maybe this is too much extrapolation on my part, but from this and some other sources I learned that if search engines can crawl your site quickly and efficiently, this is better than having jumbled coding and nonsense coding created by junky CMS. There was some talk along the way from folks about code to text ratios (this is nothing big to worry about; is the advice I am taking these days, although this doesnt mean it has no effect, it is just insignificant and not a priority.) <--- if look and feel increase conversions by 50%, increasing search traffic by 15% via sacrificing look and feel is not worth it.

I know I recall many folks (Will Reynolds- Seer SEO I think) saying alt tags are worthless and they don't use them (edit; Im leaving this sentence in because 321 referred to it; but this was not a quote and I guess maybe I wrote this poorly). For the record I still use them and my google image incoming searches are doing great. (almost certainly no harm here, only possible benefit)

Some of the stuff in this video is related to deisgn/conversions etc, I think we can leave this out of the thread because im certain most of this advice is still valid.

I'd love to hear thoughts on this because Im wondering if some of these concepts are outdated or just wrong to start with. I'm speaking mostly about coding efficiency (using css), using strong instead of bold, avoiding tables, labeling links etc. Do you still care about this stuff or believe in it?

I believe in a lot of this, as stated above, no harm; there is only possible benefit in believing.... but I certainly have sites that use tables etc... it becomes a cost benefit thing I suppose. My issue is that now I have to decide whether or not I tell someone to go through complete template redesign because search engines crawl your site more efficiently if you use css. Just my opinion, but this is totally not worth it unless the scope of your site is an anomaly in some way. (not quite a literal example although close - point being, a lot of money/effort for something very small that has debatable ROI... so lets debate)

Edit; I have invited chuck to participate in this thread via youtube.

[321]

because search engines crawl your site more efficiently if you use css.

Clean code is the main point. CSS usually has cleaner code, but CSS can be used stupidly too.


As for alt tags being useless, I think you probably misread that as only someone totally clueless would say alt tags are useless. The title attribute on images is useless seo-wise, but those are different things.

[PTC]

Thanks for the response 321. I think a great question that would sort of suck the information out of you would be this; would you recommend that an SEO client go through a large website and move everything to CSS because it might imporove things related to SEO? Thousands of dollars in labor to move all pages to css on some occasions would not be out of the question.

[Ridge]

I think it still matters and I have spent a lot of money and hours cleaning up code and non-sense. The big thing is just having clean code and removing things that do not need to be there. to me its common sense and just makes sense. I don't think css vs html vs CMS makes a difference. Although we use a custom CMS and have 100% control over what it does. We still have some older sites we made before we knew about clean code and such that need cleaning, bit overall they are not horrible.

Table - Ugh I just hate tables unless you have an actual table lol - Again its just bloat IMO -

So I guess to me its all about clean and quick loading sites, or it should be. Not so much from an SEO perspective, but a user experience perspective

[TheGooner]

I don't buy into ANY of the clean code works better for SEO debate.

It jest does not make any sense at all from Google's point of view to be "judging" and reward site positioning and ranking based on the technical merits (or otherwise) of the site construction methods.

I can't see how ANY search engine algo would benefit from it being a factor - as it would reward techicians concentrated on structure over content ...

The loud squeal of coding nerds the world over that want W3 complaint code, CSS and any of the latest coding fads to mean something valuable does not make it so.

Google wants to index good content - and wants to know what it's about - it does not care about the coding structure.

As long as your HTML bloated tables are not broken beyond repair - if they can be read and consumed - then they're fine.

[pokerprop]

The website I design now you can read the source code about as easy as you could read the text on the website. I'm not sure if it would be worth it to go to great lengths to convert a thousand pages of a site, but when starting a fresh site once you've done a couple coding in this way is very simple even with wordpress and is not any more time consuming than coding with boat, so in that sense even if the benefit is minimal why bloat code when it is not required.

Of course today I think more and more people are going with purchase someone elses theme and then hack it up (plug-ins or w/e) to make it do what you want it to do. This has benefit in saving maybe a few hours per site and a month learning curve or a hundred or so to a coder but over the long haul has many other factors making it less than optimal move for most for a variety of reasons but perhaps that is a different conversation.

I'm slightly interested in the topic here, but really won't be changing anything because I always work to get cleanest code possible already.

One other thing though not mentioned: in response to Gooners comment. One factor it for sure makes at least a small difference is load time. There has been some talk this will be (or is now) a small ranking factor. The more code and less efficient means to deliver it the longer a page takes to load perhaps making it a ranking factor beyond the reasons you mentioned.

[321]

Thanks for the response 321. I think a great question that would sort of suck the information out of you would be this; would you recommend that an SEO client go through a large website and move everything to CSS because it might imporove things related to SEO? Thousands of dollars in labor to move all pages to css on some occasions would not be out of the question.

It's never to late to do things right, but it does break down two ways: if you start a site, it is critically important to have clean code for your basic template. A lot of sites get put up now with inexcusably lazy, bloated code. It's the equivalent of loading a boulder in the trunk of a new car. It's unneeded and will impact performance for eternity.

The second thing of cleaning up an existing site is clearly less important than anew site, but still, those things that can be easily fixed should be fixed. And, those things marginally time-wise to fix should be seriously considered. On that latter point, one thing that can be done is to experiment. Anally clean up the code on some test pages, see how long it takes, see how many bytes you save.

For TheGooner's sake, the point of clean code is mostly threefold:
- load pages faster because Google does care about that;
- make pages leaner because you absolutely will get more pages crawled... it is significantly to a website's advantage to get 5 40k pages crawled in a day than 4 50k pages. It doesn't work exactly like that, but it does work something like that.
- if you can't read your code, you are going to have more problems diagnosing problems... both sitewide problems and on an individual page level

Gogle doesn't think "oh this is pretty code we'll rank it better", but any contraption that has unneccessary boulders in its trunk is just not going t perform at optimum efficiancy.

[GregW]

Making good use of css and xhtml is all part of good website construction. Personally, I don't try and create websites with good code just because of the possible SEO benefits, but more because it makes your website easier to work with. If it helps with SEO then that's just a handy secondary benefit.

I don't buy into ANY of the clean code works better for SEO debate.

I can't see how ANY search engine algo would benefit from it being a factor - as it would reward techicians concentrated on structure over content ...

The loud squeal of coding nerds the world over that want W3 complaint code, CSS and any of the latest coding fads to mean something valuable does not make it so.

Google wants to index good content - and wants to know what it's about - it does not care about the coding structure.

I don't think it's fair to brand all those that strive for clean code as "nerds". Sure, there are elitist coders that are incredibly anal about coding, but having good, semantic code can only be a good thing and not something you should ignore.

I believe that having semantic code helps Google index your content. For example:

If you have a paragraph of text, put it in a p tag.
If you have tabular data, put it in a table.

You wouldn't put your paragraphs in to <blockquote> tags and your headings in to <acronym> tags. Just as you shouldn't put your whole page in to a <table> tag. Surely that would only work to hinder Google's ability to "understand" your page.

[Mike]

It jest does not make any sense at all from Google's point of view to be "judging" and reward site positioning and ranking based on the technical merits (or otherwise) of the site construction methods.

I'm one of those nerds, for sure. It's actually a bit of slight ocd or something I have with it. Even when I'm developing a backend or something like that, I still make sure it all validates even though no one but me will ever see it. It's kind of ridiculous and I think I need professional help.

With that said, I think with all things being equal, the cleaner / faster website should get rewarded. I think it shows a stronger website, if the webmaster took the time to make sure it was standards compliant. IMO, it's probably a small (very small) piece of the algo but still a piece.

If you're building new, definitely concern yourself with clean code. If you're doing fine without it, don't bother.

As for the b / strong or i / em tags, each one has a purpose, I just don't remember what the difference is. Neither one is better than the other, you just need to know how to use them. Obv, I don't, lol

[dirkjan]

I don't buy into ANY of the clean code works better for SEO debate.

It jest does not make any sense at all from Google's point of view to be "judging" and reward site positioning and ranking based on the technical merits (or otherwise) of the site construction methods.

I can't see how ANY search engine algo would benefit from it being a factor - as it would reward techicians concentrated on structure over content ...

The loud squeal of coding nerds the world over that want W3 complaint code, CSS and any of the latest coding fads to mean something valuable does not make it so.

Google wants to index good content - and wants to know what it's about - it does not care about the coding structure.

As long as your HTML bloated tables are not broken beyond repair - if they can be read and consumed - then they're fine.

I don't agree. Clean code and the correct use of symantics and css ís valuable for the user experience. Disabled people can use your site, people can more easy enlarge the font size, the site looks decent in all browsers. So as using the right html codes in a clean way is usefull for visitors, Google should send visitors to the most accessible site first if the content is similar imo.