The Experiment So Far

This site was set up as 1) a tool I could use to find good curated content, and 2) an experiment in automatic curation. So far, the experiment has been a resounding success. I will note a few of the more interesting results in this post.

First, the site is not old at all. The very first “automatic curated post” was made on 3/25/2012 (less than a full month ago from this date – 4/22/2012). Today, so far today there have been 165 unique visitors, 250 page views, and 4 repeat visitors. In less than a month…

…and there are NO links to this site other than trackbacks. SEO Quake tells me there are ZERO backlinks to the home page, and ZERO backlinks to the domain. Put that in your pipe and smoke it.

And, the site is made up of 100% curated (duplicated) content. Three of the 700+ posts made so far are unique (including this one). Every post is “kicked off” according to where we sit in the Google Trends cycle for the day (I’m not going to try to explain the code involved for that), and consists of a few Yahoo news feed summaries plus a few Amazon and eBay products. Comments are automatically generated from Yahoo answers.

Post Panda, this is not supposed to be working.  No way, no how.

The site is getting dozens of top 20 rankings. Dozens. For trending news. Take this one, for example:


As I write this post, it is sitting at the #4 position for the search phrase “secret service prostitution sting” on Google – two days after it was posted!

#1 is Huffington Post.
#2 is (whatever that is)
#3 is
#4 is

Heck, I’m beating Washington Post, WikiNews, and Chicago Tribune for this story. Man, I’m absolutely blown away. If this was the only one, I could chalk it up to coincidence – but it’s not the only one. Not even close to the only one. (For – Ireland – I actually own positions #2 and #3 for the phrase).


The first thing that is noticeable is that most of the top rankings are tag pages. My auto-posting script reads through the Google Trends feed, pulls the content based on a given trend “phrase”, creates the post and assigns a new category and tag using the phrase. In the case of the result mentioned above, the exact phrase was “secret service prostitution sting”.

Tag pages are apparently pretty powerful in the eyes of Google.

Not all of the top rankings are tag pages, but the ones that are not are for more obscure phrases – like “humber”. This found itself in the #6 position for ” white sox humbert rebroadcast”.

Some pages just “luck out”, especially if other sites determine the phrase as being unworthy of attention.

All of the posts are a relatively random mixture from different sources. News feed from Yahoo, products from Amazon and eBay, comments (questions and answers) from Yahoo Answers. The feed sources I chose may or may not return content based on whatever keywords are used, so each post is slightly different from the last.

Mixing content in a random fashion works, even if it is mostly duplicated.

Since the content is primarily based on Google Trends, it is always going to be what Google thinks should be important at the time. I’m really not sure if this technique will work with less trendy keyword phrases (although I’m going to experiment with that too).

Giving Google what it wants is a pretty good strategy.

So, I’m going to sign off for now. I will keep reporting what results I’m getting as they happen. If you like this experiment, and wish to see it continue, consider donating a few bucks to keep it going. Thanks!

Click here for more information about 'The Experiment So Far'.

Leave a Reply

Don’t Miss Anything!
Subscribe now!

Enter your email address:

Delivered by FeedBurner


The breaking news and utilities available here are free for now. If I get enough donations, I will keep it that way.