In the past I have had dealings with a SEO company that has done a process that they call “reverse build”. They move what they believe is less important content such as the header and navigation to the bottom of the HTML document. And more important content such as the page’s name and it’s individual unique content to the top of the HTML document.
They then use absolute positioning to make the page appear in it’s correct order. They state that this helps with search engine rankings, do search engines use the order that elements appear in HTML as a factor for how and where a site ranks?
Is this common practice amongst SEO firms?
I have found this article which suggests Google does give a higher priority to text that appears higher in a document.
Yes, Google gives a bit more weight to content near the top of the page than right down at the end, but it’s also smart enough to figure out what’s content and what’s navigation, and to allow for that when reading and indexing a page.
There was a school of thought that said that “content first, navigation last” was better for accessibility, but I don’t think that is still the preferred method. “Navigation first, content second” is a much easier design prospect.
The original reason for this kind of technique rests in the history of Googlebot. Googlebot used to only look at the first 101KB of each webpage. Thus, if your main content did not appear within this limited you were missing out.
Today, Googlebot has the power to crawl to the bottom of all but the largest of webpages. So, the necessity of having your content before the usual navigation code etc. is removed.
Having said that, IMHO it is still not good practice to have, for example, massive javascript files at the top of your source code.
In addition, Matt Cutts has said that Googlebot places slightly more value on content that appears towards the top of the source code.
That was a really long time ago. Google has been reading at least 500KB for at least 5 years … and how many HTML files are more than that?
(This isn’t a criticism of what you’ve said, just pointing out that the mythology around SEO is often still rooted in tricks that stopped being relevant ten years back)
Having said that, IMHO it is still not good practice to have, for example, massive javascript files at the top of your source code.
Very true – because once a Javascript file has started downloading, you’re unlikely to get anything more done until it’s finished, so it’s better to leave it as the last thing to load. I get very frustrated by sites that load huge Javascript libraries before they’ve got the whole page, causing the whole site to hang for ages.
As more and more sites adopt HTML5 coding and standards this should become less of an issue because you will be able to markup the different sections of a page so the search engines will be able to clearly tell which section is the header, navigation, footer, body content, etc. This should make crawling easier for the search engines if sites implement it correctly.
Of course not all old sites will switch to html5, so in my experience I have had some success by moving the most important content up higher in the code.