How much Javascript can SEs read?

Something came up on another thread which made me think that we should test it.

I think the general consensus is that search engines can read some level of javascript.

I propose that we test exactly how much and what javascript functions can be read.

The obvious one is document.write("<h1>Hey can you read this?</h1>");

I think we should use a common non sensical keyword. I have a domain that I can set up a page on.

Please suggest any thoughts or ideas on the details of the test eg;

Using Option function. Functions that produce text on call etc.

Ideas welcome.

Sounds great... let me know if you need any help.

If you could link to http://www.theseofiles.co.uk/seo-test-page-1.html using the following anchor text SEO Test Page 1

That would be great, thanks. The idea is not to include the keyword in any link. The only place that the keyword will come from is javascript.

I plan to run a series of tests on different aspects of the page and javascript functions etc.

I have an area of this test I am uncertain about.

Obviously, Google's cache is what they have indexed and use to rank a site but... Is the cache just the code that is rendered in your browser (I believe so) as it would normally or is it the page as Google sees it?
For this reason I am keeping the keyword in the javascript only.

Thanks, any help is appreciated.

Another thing you could possibly test on that page (or another page) is to see if Google crawls javascript code for urls (which is what someone mentioned Google Analytics bot does).

For example, create a new page that is only linked to through JavaScript and see if it gets indexed.

as far as i know scripts are crawlable... its just so happen that its hard for spiders to crawl site with embedded scripts, thats why its recommended to save them as external file...
do you think this can help rankings than natural way of putting some keywords or content..? thanks.... ;-)

justsearchltd

Interesting suggestion


If I can be of help with the whole experiment, I am all ears.

I think this study can really be helpful for all of us... I'll come back in this thread once in a while...

Yeah justsearchltd thats a good idea.

I will put up another test page to test that out.

What type of javascript URL are talking about though?

An OnClick function that changes the window.location property? I would guess that that can't be followed. Although if it was an inline onclick function rather than an external function it might be read.

This is definately interesting to test.

I will put up a second page. seo-test-page-2.html at the same URL. Please link to that using a javascript only link and lets see if it gets indexed. I will use a nonsensical keyword for the page. Please link to it with the text SEO Test Page 2.

Thanks.

I'll be writing a short article about the subject in a few hours on our blog, with a link done in javascript (the blog gets indexed every few hours so we'll hopefully get a quick answer!)

I'll put the link in through any way that has the whole url in one piece, as my theory is that search engines won't execute any JavaScript but just crawl through it looking for relevant content or links.

In response to leftybogs, if we find any of the content to be crawled, I don't believe it'll be a good technique for ranking better (and certainly not for spamming a keyword!). You'll never beat normal content (headings, titles, link text, etc.)

Thanks very much for your help.

I agree.

The rule was / is that only href and rel links were followed but I think that SEs have got a lot more advanced. They can read flash so a bit of basic javascript.... who knows!

SEOs often consider that an SE spider is like an old HTML 2.0 Version 2 browser - ie bumb! But browser technology advances quite quickly, we never discuss or investigate the bot or crawler version being used. I am sure that they are updated and have additional capability.

This test will hopefully find out to what extent they have been upgraded!

Nothing that will help the test but which goes along with your last comment Trevor. I have heard rumors and keep in mind they are just rumors although there is a good bit of information to insinuate it is true. I have heard that Google has worked with the Mozilla team and now uses that technology for their robot, if this is true you can see where alot more language support could be implemented as Mozilla IMO is #1.

But again that doesn't mean anything as far as the experiment goes....

I agree and it wouldn't suprise me at all.

I thought I would check this URL through the SEO Chat Spider Simulator

http://www.seochat.com/seo-tools/spider-simulator/

It gave me an Error - no links found message.

It wouldn't suprise me if Google can read document.write functions. Hopefully with this test we will soon know for sure.

Update - this page has been indexed but neither of the two test pages have been indexed yet.

Any links are appreciated.