Test Update: Hidden Content
If a piece of content really important don’t hide it in any way on your page. Display it clearly visible for both users and search engines.
- Don’t hide important content using mechanisms such as tabs and accordions
- If it must be hidden then use a common framework to hide content
- Avoid custom-made, unusual ways of hiding and revealing content
- Click-to-expand element must not be invisible or tiny
- Avoid large ratio of hidden to visible content on your pages
The above sounds like common-sense advice, yet we still see important information being tucked away behind tabs, accordions and various visual content expanders including my own hypotext. Visual information compression and structuring makes sense from the usability point of view. You don’t want to overwhelm users with too much information at once. Google, however thinks otherwise.
Back in August I wrote a post on Moz that discusses user behaviour as a useful ranking signal. A few months later I copied the same article but made one key change to it. All the content was visible by default.
I tested the performance of the Moz article in Google search for terms which are hidden by default but revealed on click, one such phrase is:
“The site caught my attention, as it outranks the official page for a fairly competitive term.”
Google search for this phrase returns everything but Moz, despite its authority and reputation. It’s not just that my copy of the post outranks Moz, it’s pretty much every lowly scraper out there that outperforms it too.
Long-form content may discourage reading with many of your visitors. Some authors keep their articles short by linking out to all the extra information instead of including it into their content. The problem is that by sending visitors out to external sites interrupts their reading experience. Some may not come back to your page. The Whiteboard Friday session I recorded at Moz in July this year talks about benefits of presenting a concise text version to your readers and letting them expand interesting parts in-line with the content.
After testing its performance in search we learned that the method didn’t sit well with Google. All hidden parts were devalued in search.
When I first published “Here’s Why Nobody Reads Your Content” it created quite a splash in the community, but Google ultimately disagreed with my innovation and devalued the content in the same way it did with the Moz post. I went back to the original article and made one key change. All content was visible by default, but I gave my readers an option to click on the “5 minute” version link and crunch this 5000 word article into 500 words instead. Go check it out to see how it works.
Matt Cutts has a nice video which explains under which circumstances Google may trigger their hidden content filter:
Oh the irony!