Understanding Flat Site Architecture

Along the last months I've heard quite a lot about the concept of Flat Site Architecture and when discussing it I see some misunderstandings in jeopardy to cause more harm than god, so I would like to help you get some concepts really crystal clear.

I found this interesting video from @randfish, let's go through it:

At the beginning of the video he uses two site architecture schemes, one in the antipodes of the other what could help to blur some ideas.

Second, Rand is associating more the idea of 'flat site' with the internal link structure as he continues talking about 'linkujice' between pages.

Third, from an usability perspective, he talks about 'one click distance' from home to all the pages you may potentially need access. In sum, he defines flatness as 'few layers of click as possible' to reach the content 'keeping the depth to a reasonable number' Bastian Grimm adds wisely in his article.

I guess any SEO agrees with those ideas and sitemaps have done the trick since the beginning of the Html era, specially for sites where navigation is based more in an internal search box as classifieds sites with tons of pages.

What else helps users and search engines to understand the structure of content? Bread crumbs are a good resource but tags, tag clouds, calendars and similar secondary alternatives to a categorized structure, are usually adding confusion to organized information unless you make a clever use of canonical tag or even better the right multi categories website structure.

@richardbaxter's article explains better the idea of horizontal cross linking to solve somehow the lack of relevancy at pages down the hierarchy.

Page organization, click distance, relevancy distribution... great but where the mentioned Information Architecture thing fits here? Not clear at none of the two articles and video mentioned.

Taxonomy and URLs are best friends

No matter how simple or huge a site can be, for sure the content must be organized under different concepts. Here it comes, taxonomy, the practise and science of classification.

Sooner or later this ethereal organization of content will have to be materialized in URLs and let me remind you that URLs are the authentic ID of pages inside the vast internet so not only they identify a page but also can tell in advance what the content is going to be about.

Question, which one defines better the kind of copywriting 'company.com' offers?

http://company.com/copywriting-services.html
http://company.com/internet-marketing-seo-copywriting-services.html
http://company.com/internet-marketing/seo/copywriting-services.html

No meed to discuss that first is the worst one but what about the second and the third if both have the same words?

A flat site architecture does not mean at all that you have to get rid of directories in URLs and these are my two cents to those articles. On the contrary, it is something to add to the click distance + relevancy distribution equation.

Directories in ULR structures, please

First of all, worth mention, directories constitute a standard being there since the first days of Html, for technical issues first, and since the invention of directory-level configuration files (.htaccess and similar) for IA reasons too.

The directory thing presents advantages, it helps to:

• structure content in silos, related at top levels but independent at the lower ones, achieving the right granularity where needed
• disambiguate concepts creating better semantic relations between them
• create more room to allocate content and keywords going from generic and competitive to long tail ones
• have a more scalable content strategy, important if a wide diversity of subjects appears in the horizont
• extra hint: helps a lot with the analytics tasks, pathing, content drilldown reports and so, reducing drastically the sometimes required regex (and extra headaches)

What makes non sense at all, and seen a lot lately, is having directories in URLs but remove them from URLs when the page level is reached, for example:

http://professionalseo.ca/vancouver-consultants/
directory of SEOs by city
http://professionalseo.ca/ani-lopez.html
consultant page

that is totally incongruent, this is the right one:

http://professionalseo.ca/vancouver-consultants/
directory of SEOs by city
http://professionalseo.ca/vancouver-consultants/ani-lopez.html
consultant page placed right under directory

Worried about the length of the URL including directories and subdirectories? As long as they look pretty natural and not ridiculous long I would not care.

Even @maileohye, Google Developer Programs Tech Lead, preached recently in favour of trailing slashes at the end of directories to make them look like what they are although from a pure Google perspective they handle / and no-/ the same.

Hard to isolate the impact in traffic of changing to well structured URLs as those changes don't usually come alone, they are part of a bigger plan of optimization. Content structure is a pretty invisible task in any SEO project but absolutely fundamental. I have always seen positive consequences anyway.

In short, directories in URL make search engines understand better all the concepts under the umbrella of your site the way you want, not leaving this task for them to guess. As an obvious consequence, if done properly your site gets better organic traffic.

You may not agree with me but don't take the wrong decisions just because you didn't dig deep enough this particular and important aspect of SEO.

Is it better to have directories in the URL path?

(Update Aug. 25th) Today @mattcutts talked about that in one of his videos at GoogleWebmasterHelp Youtube channel.

Seems quite clear that directory structured URLs or just flat ones are not going to make a big difference in your rankings per se but the advantages, in top of the ones assumed by Matt, the can add to your SEO strategy are going to add more (indirect) benefit to it for sure, I should say.