Back on November 7th Beanstalk’s Ryan Morben decided to run a test on iFrames to see how they get crawled (and to answer the question … do they?) I had to support the test as we’d been receiving mixed signals. A good clean test running an iFrame of our own domain containing only text. The notion was … if a search for that unique string of text produced our page as a result then we know Google crawled it.
The result was interesting … the content got crawled but the ranking page wasn’t our blog post but rather the URL of the frame source itself. It appears that Google treated the iFrame call as a link more than content on the page.
When we published the results I got an interesting email from Stefano: “I noticed the results of your iframe test showed that google did indeed index the unique phrase. Can you do another test where you load the phrase from a different domain ? Thanks!”
My assumption to this question is “yes” based on the initial results however it’s definitely worth a test. To that end we’re running two separate tests on this, the first we will be running here on the Beanstalk site with the following frame:
The second test location isn’t being released in this post just to make sure the process isn’t gamed. I’d rather have slow or even no results than false positives.
Stay tuned – as soon as we have conclusive answers as to how it turns out, we’ll let you know.
And until then … enjoy the weekend !
EDIT: We will still be doing a follow-up post with more code examples, but we have results of the first test of iframe text crawling:
Essentially this result is what we should have expected.
A search engine needs a crawler to understand the iframe syntax, and since a lot of iframe data is secure or private, there’s little motivation to go that extra mile, and it’s no surprise that the ones that do crawl the frame don’t publicly index the result since that’s just asking for privacy and other issues.
SEO news blog post by Dave Davies, CEO @ 4:02 pm on November 25, 2011