Digg Algorithm Musings

Posted by Miles Evans

On ProfitPapers digg replaced Google as my top traffic referrer by a huge margin a couple of months ago. I see more and more publishers reporting the same thing. There has been endless discussion over what ranking factors Google uses to prioritize our content, but what about Digg?

Don’t read this article attempting to glean any digg spam techniques! That is not my intention and I certainly do not submit all my stories to digg. Digg users are very aggressive with burying spam – me included. So again – don’t use digg as a spam tool or you will do more damage than good. Think rabid dogs.

That should please the diggers. Alright moving right along…

Riding the Queue

With the release of the latest version of digg there has been some discussion as to the relative ease with which a story is landing on the front page. Well it’s certainly true and the reason is obvious. If you noticed, there is now many more categories which means that stories are riding the queue for a significantly longer time frame than before. More time in the queue = more votes. Makes sense right?

Having said that, nobody aside from the digg team knows for certain what factors are required to get a story on the front page. I have seen it take anywhere between 40 diggs and 100, from a few hours to days. Granted from my own experience it would seem that with digg v3.0 it is taking less digs to make it to the front page, but this is open to debate. Everything is. The point is that there is no magic number, as it depends on several factors.

Raw Speculation

• If a story gets around 50 votes in the first hour there is a good chance it will be promoted to the home container right away. Alternately submissions often take a couple of days and then magically appear on the home page once they are released from the queue.
• The pace with which a story is being dug is important. This is an obvious ranking factor. Anyone who browses the queue regularly will know what I mean.
• One would guess that all diggers can not be equal. When I get a digg from a user who has 10 posts on the home page shouldn’t his vote count more than a user with 0? What about the length of time they have been a digg user. Or maybe the amount of comments they have.
• The raw impressions on your submission. This seems obvious.
• The number of comments a submission receives. Also the number of these comments that have positive vs negative ratings. Or, as mentioned above, the reputation of those commenting on your submissions may be a factor as well.
• The algorithm may take time of day into account but I personally don’t think it does as digg has a large overseas following. There is certainly more activity during North American peak surfing hours so it's quite possible the devlopers took this into account.
• Buried stories are obviously going to be at a large disadvantage of reaching the front page. Buried comments as well. Our spam fighting votes at work.

Ok so most of this will likely seem pretty obvious but I am going somewhere with it. In a follow up article I will go over how using what we know and suspect about digg might go a long way towards effecting what we don’t.

Posted Jul 04, 2006 at 03:30 PM | Permalink | Trackback URL | Del.icio.us | DIGG!