Def'n 'Addiction Marketing' - Pronunciation [uh-dik-shuhn mahr-ki-ting]:
A strategy used by companies [read Digg], whereby a product or drug [read front page appearances] with near instant addictive properties is given away free or very inexpensively on the first few ocassions until the victim is addicted. Prices [read level of difficulty] are then raised, demand having been secured.

I'll give credit for this terminology to Mike Myatt who claims to have coined this term a number of years ago.

Addiction Marketing Cartoon
Image courtesy: University of Sydney Australia

Now that the dust has settled somewhat ... lets do a quick post mortem and see if we can understand what changes Digg made, and why it made the changes it did. Understanding this, may help to anticipate such changes again in the future.

The Impetus for Change:
First ... it would appear that while Digg is still much larger than its next largest competitor (Reddit), its growth (monthly visitor counts) over the last 6 months has been flat, if not declining slightly. For any business, this is cause for grave concern ... and action. Especially someone putting themselves on the auction block. Anyone who understands product life cycles knows that if you've stopped growing, you're not too far from beginning to shrink.

Digg / Reddit 2007 Visitor Growth Comparison

Digg's visitor stagnation may be happening for a few reasons, but I'd venture to say that many people are leaving for 2 primary reasons:

    1) they're frustrated by the fact that their stories never hit the main page. I've heard this from more than one or two people.
    2) they're frustrated because they don't understand how some of the 'lesser quality' content does make the main page

Obviously, Digg needed to devise strategies to stop the bleeding, and resume visitor growth. So they started blueskying options, and assessing the impact of those options. Many month's later, the solution they ultimately selected actually spoke to both of the problems identified above.

The Digg Solution (Suspected):
Digg made a number of interesting changes, and this post touches only on a few of the more major changes we're aware of.

Change #1:
One of the biggest changes Digg appears to have made, was to discount votes from others who routinely vote for a person's submissions ... friend or not. We'll label these 'in-network' votes. Prior to the change, if someone had 200 friends that voted for everyone of their submissions, that was sufficient to promote their submissions to the front page relatively consistently.

Change #2:
I think its also likely that Digg implemented a minimum number of 'out-of-network' votes needed (diversity) benchmark too (not that this is a static number) ... otherwise power Diggers could build 1000 friends and once again dominate at will.

To be frank ... I'm not certain if both of the above proposed changes were implemented, or only the diversity requirement. Since the changes were implemented, I've had a submission go hot at only 45ish votes, though another not go hot at 125 (with very similar diversity scores no less), leading me more to believe there is yet more to the algorithm change (see Change #3).

Change #3:
Digg seems to also be evaluating the trustworthiness of the site the article is from. I've had submissions go hot at 45ish from well respected sites, yet requiring many many more votes from less well known sites (150+) even though they had equivalent diversity. Hmmmm deja vous ... where have I encountered this phenomenon before? Google maybe!

Many other smaller changes were implemented too, but those are beyond the scope of this post.

Impact for New Users:
In thinking about the changes, they make perfect sense. New users to Digg who haven't submitted anything previously, will not have votes for their submissions discounted. No one who votes for their submissions will be in-network (since they're a new submitter) so their votes are worth the full value (lets say 1 point). Accordingly, they should be able to make a story 'go hot' much quicker than experienced power Diggers if they submit from highly respected sites, often at 35-50 votes. Again, this is what I like to call 'Addiction Marketing' ... get new users hooked early. The earlier the better.

Impact on Power Diggers:

Essentially what Digg is saying is this; we understand that your friends have similar interests to you. We also understand that there are others in the world that should have similar interests, but that are not already friends. Since its highly improbable that any one person (power user or not) is that well connected to know absolutely everyone interested in a topic, we're not going to promote a story to 'hot' until a minimum number of out-of-network people have in fact validated the story by voting for it.

The result; experienced users (aka power diggers) now often need 150 or more votes to have a story go hot, unless they can:

    a) quickly expand their friend networks to get reasonable number of new friends voting for a story (this is difficult to do, and continue to do over time)
    b) their stories are of sufficient quality that they have natural momentum (ie. non-friends voting for the story because the like it).

It might just be however, that power users are conditioned to use brute force, when in fact the problem of first page status instead requires finessing ie. diversity vs # of votes.

What Does This Mean To Future Social Media Gurus:
As it happens, Stumbleupon (and likely a number of other social media sites too) is also looking at voting networks, and is thought to be discounting accordingly. Those in the social media space who want to excel far into the future must develop strategies and techniques to continually build out their followers (ie. friends and acquaintances). They can no longer rely on a largely static network. Ongoing New Friend acquisition strategies, while keeping existing friendships intact, will become crucial to social media success! This will be the subject of my next post ... Social Media - New Friend Acquisition Strategies.

In the end, Digg's new algorithm is merely employing a tried and true tactic to bolster its stagnating numbers.

Additional Reading:
Front Page Addiction: Destroying Families, Ruining Lives
10 Symptoms of Social Media Addiction
Social Media Addiction in the Workplace
Help Me ... Social Media Is Trying to Take My Life

* Please note that this post is in no way making fun or light of serious drug problems.