It’s The (Other) Algorithm, Stupid! Understanding DiggRank
Have you ever wondered what it really takes for a story submitted to Digg to get to the home page? Or why a certain story—even a really good, social media friendly story—never got to the home page? I’m frequently asked the question, “Hey, my story has [number] of Diggs but it still hasn’t been promoted […]
Have you ever wondered what it really takes for a story submitted to Digg to get to the home page? Or why a certain story—even a really good, social media friendly story—never got to the home page? I’m frequently asked the question, “Hey, my story has [number] of Diggs but it still hasn’t been promoted to the home page. Any idea what’s wrong?” And, relatively less frequently, I hear someone saying in amazement, “Wow, all it took was 29 Diggs and that story rocketed to the home page!” I’m always tempted to reply “It’s the algorithm, stupid!”
Digg has an algorithm?
Yes, just as PageRank and other algorithms are used to rank web pages by search engines, some social sites use algorithms to determine which stories become popular. Call it DiggRank, if you will. Let’s take a deeper look at what the Digg algorithm is and venture a few guesses about how it works.
What is the algorithm?
Unlike editorially driven sites like Fark or Slashdot, where news is handpicked by a tiny group of individuals, socially driven sites use the votes aggregated by the community to decide what content gets promoted to their home pages to be viewed by the masses. At the same time, content promotion isn’t as simple as just comparing absolute number of votes that each submission gets and then promoting the ones with the most votes. There are several other factors that come into play. Some of these factors we are absolutely sure about, whereas others we can only venture guesses about.
Before we look at what factors may be taken into account by the algorithm, it is important to know that the algorithm is used to ensure that community participation is fair and that no one can unduly influence content promotion. Therefore, the algorithm is kept secret from the community to prevent people from bypassing or manipulating it.
What does the algorithm take into account?
Having been a member of Digg for over 2 years and several other communities for over a year, I can confidently say that Digg’s algorithm is the most complex and therefore hardest to manipulate (though by no means is it completely fool-proof). Based on my experience, the elements Digg’s algorithm takes into account include:
Recent participation rank of user and followers. Depending on how successful you have recently been on Digg, subsequent successes may be more difficult. For example, I have had 7-8 stories submitted and promoted to the home page one day, only to see 3-4 stories get 100+ Diggs and not be promoted the next day. At the same time, if I decide to take a few days off, the “algorithm bias” seems to adjust itself and I have no problem getting good content promoted again. The rank and recent successes of a user are taken into account both when you are submitting a story and also while Digging (voting on stories). If you get a quick succession of Diggs from “high-value” users, you are likely to be promoted faster and at a lower number of Diggs than if even dozens of new users Digg you. This, of course, is to ensure that the Diggs are of good quality and the community is actually doing its work by voting for good content and burying bad content.
Voting activity. The number of Diggs your story will require to reach the home page is directly correlated to the number of votes (Diggs) generally being cast on Digg at any given time and how your story compares to the average.
Submission category and activity in the category. Competition in some categories (Technology, World and Business) is much fiercer than in other categories (Sports, Entertainment), and therefore it is much easier to submit and have something promoted in the Sports/Motorsport category than Technology/Tech/Industry News. Also, along with being compared to general voting activity on Digg, your content is compared more directly (and probably with more weight) to content in its category. For example, it is possible to have a story promoted at 50 Diggs even though it’s not high on the upcoming queue for all sections, as long as it is at the top in the queue for its category.
Speed of votes and diversity of voters. The faster a story gets votes, the lower the vote count has to be at which it is promoted. For example, a story may collect 120 Diggs over 24 hours and not be promoted. If the same story gets 90 Diggs in one hour, it will almost certainly hit the front page. At the same time, however, it is incredibly important to have diversity in votes. Diversity helps prevent people from banding together into “voting-rings” (i.e. circle-jerks) and unfairly pushing their stories to the top. This is one of the reasons why you see stories from top-ranked users sit at the top of the queue for hours waiting to fulfill the algorithm’s diversity requirement (i.e. they are penalized for having a following of users that Digg every one of their stories).
Buries received. This is quite straightforward. The more buries your story gets, the longer it will take for your content to be promoted. If the Bury to Digg ratio (which is not 1:1; buries are weighted more heavily than Diggs) is too high, your story will completely be removed from the queue. That said, it is possible for a story to acquire enough votes to outgrow the Buries it gets. Each bury, however, can be taken into account as a certain number of negative votes, which increases the total vote count the algorithm will require from your story before promoting it to the home page.
Comments and comment ratings received. Participation in the comments can help push a story over the edge. People think that inserting “great article—thanks!” will help further their cause, while in reality these fake comments have the exact opposite effect. There is nothing easier than spotting a spammy submission with fake comments and burying it to oblivion. Naturally acquired comments (and ones that are voted up by the community), on the other hand, help tip the content promotion algorithm in your favor.
Misconceptions about Digg’s algorithm
Now that we have covered the basics of the content promotion algorithm, let’s examine some common misconceptions about Digg’s algorithm.
Wrong: An absolute number of votes is required. There is no absolute number. The number varies daily and even hourly. As mentioned before, the number of Diggs you need varies based on submission category and the recent participation record of the submitter and subsequent Diggers, as well as the number of votes, the time in which they are aggregated, and the diversity of the voters.
Wrong: You’re doomed if your story isn’t submitted by a top user. There is no such thing as content being automatically promoted to the home page. Even the best content submitted by the most consistent user can get buried if enough people don’t like it and Bury it. The algorithm tries to ensure a level playing field for all users (though this doesn’t always work), and in fact is sometimes harsher on top users than on newer ones.
Wrong: Number of friends is important. The number of friends you add on Digg is completely irrelevant. What Digg looks for is diversity in the Diggs a story receives. The fact that a user is your friend and votes for your story is irrelevant. Even if someone is consistently Digging you without being your friend, those Diggs are weighed less.
Wrong: There is a 24-hour window for success. This is true of most content: If your submission doesn’t get promoted within 24 hours, it usually has no chance of being promoted later. However, a small number of submissions do get promoted—even 2-5 days after submission—if people continue to regularly show interest in them.
What have you learned about the Digg algorithm from your own experiences? Please share your thoughts on this discussion about DiggRank on Sphinn.
Muhammad Saleem is a social media consultant and a top-ranked community member on multiple social news sites.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.