Important: Still Testing
An embedded YouTube video and a few original lines of text will pass the thin content filter just fine. So what’s all the fuss about with thin content? Why did so many sites with “thin content” get nailed while others are totally fine?
We have a theory, and it is held up in the data we have seen so far. We have 50% confidence in it.
Our Theory
Thin content itself isn’t bad at all. Matt Cutts (former Spam Czar) said franchises with many locations should not have the same “welcome paragraph” on each location, as that would be duplicate content. When pressed for what each location page should have, Matt said something as simple as an address, phone number, and the managers name would be all that is necessary.
Considering those few lines of text would be okay, why do so many SEOs want at least 300 words for content? What about people who want quick answers without reading a book? Isn’t that a bad user experience?
The reason for the “at least 300 words of content”? It was a statement made in passing by Cutts at PubCon about ten years ago. Yes, an “informal comment” treated as absolute gospel by the SEO community. That is how rumors not only start, but how misinformation is just widely accepted because “Matt Cutts said it.” Don’t be foolish. This is one of the main reasons we test EVERYTHING before implementing it on our money sites, even stuff Cutts says. We trust no one. We only trust data.
Back to the thin content issue…
What if thin content is just more prone to get nailed by Panda, and is actually only a Panda issue? For example, a site with famous quotes is very likely to get nailed by Panda. Especially when the only thing on the page is a quote that 50 other websites have. How are you supposed to remain unique?
For a site about famous quotes where it seems impossible to make unique content, there is a solution. Use blockquotes for the quote itself, then add extraneous data to each quote. So instead of just the quote, give context of the quote, when and where it was said, and other people who referenced that quote. If you want to spend a lot of time on the quote, write about how it affected you or feelings it evokes.
Why We Don’t Consider It Panda
Thin content is not a blanket rule to be penalize pages under a certain KB or word count. Going back to the famous quote example, if those quotes were original to you (and passed the Panda Copy/Paste test) you would face absolutely no penalty.
Further, thin content in some cases can actually help to avoid penalties. Consider which is easier to test for content issues, a site with 50 sentences or a site with 50 pages of 5 paragraphs of content each?
There is an example of a site ranking #1 for a 14,000 exact searches per month keyword with “thin content” that most SEOs would advise against. How thin? The entire site has less than 60 words. That’s counting the title, tagline, menus, comment box, footers, everything. The actual “content” of the site has 14 words. Yet it ranks #1 without an issue. No cloaking whatsoever either.
Examples of True Thin Content
- Contact pages with nothing more than a contact form
- Blank Indexed Pages
Best Practice
This area is difficult for us to test and prove because it really goes back to being a substantially similar content issue. At the end of the day, we are advising to treat thin content as a Panda issue. Just be aware that really thin pages are likely to get nailed with Panda penalties.
Finding Thin Pages
Scrutiny / Screaming Frog
An option for large sites is to grab a tool like Scrutiny (Mac OS X) and let it run a crawl on a site. Click on the SEO tab, scroll to the right and observe the KB size of each URL. Take a look at the smallest ones and manually check if those URLs are okay.
Small Sites
We advise building small sites and making each page awesome. Typically thin content is only an issue on massive sites. If you don’t have a massive site, chances are you can ignore the thin content issue and you can check your entire site by hand faster than loading up a fancy tool or plugin.
Feeders
For feeders, we are actively testing very “thin” content. So far very thin content appears to pass juice without any issue. If you have a large feeder network and are spending a lot of money on content, you may want to adjust your model to include “thin” content and experiment to see what happens.
In short, “thin content” is not an issue like it was in the past, but when you incorporate Hummingbird into your campaigns and page creation, “thin content” is basically impossible to achieve due to the sheer size the page becomes with the different issues to address.