It has been a while since I have posted…I have been busy merging SpinShark with PMDigital, and now I head up all of the search engine optimization efforts at PMD. It is heady and humbling. And, on top of that, there has been an incredible shift in the world of search.
The tried and true methods of determining the authority of a search engine result have changed dramatically. With the recent changes in indexing speed from Google Caffeine and the introduction of social activity into the search engines (see my thoughts, Here Comes The Flood at MediaPost) the search world is turned on its head. Age and linking were the strongest measures of authority for any page on the internet. But with the torrent of real time information flooding the search engines, there is simply no time for authority or age to accumulate. So what is a search engine to do?
Here is my idea. Instead of Google’s page rank as a determinative factor of authority, the metric needs to change. In a world ruled by URL shortners, 140 character bursts and a live Facebook feed, there is no replacement for page rank. The model simply falls apart. So I am going to suggest a new measure of authority that is based on engagement and impact:
THE REPUTATION INDEX (should I tm this?)
Let’s imagine that you are a prolific tweeter. To a search engine, your stream of tweets looks a lot like the tweets of a spammer. But you aren’t. Let’s imagine that you are Robert Scoble or Gary Vaynerchuk who are both prolific and informative. A constant stream of tweets is a great thing, but how is a simple search algorithm supposed to be able to differentiate you from a stream of get-rich-quick-internet-marketing-gurus? Here is my idea. A measurement of how much impact your social stream has on your follower base can replace the tried and true interlinking authority of page rank. Just a pure number of clicks or followers to your social stream is ripe for spammers, so I suggest that we look at importance of a stream as a relative factor compared to the total content of your followers total stream and that compounded with the actual level of interactivity with said stream.
I am sure that was as clear as mud. I will try to explain further.
Let’s imagine that between your Facebook friends, MySpace minions, Bebo buddies, Orkut posse, Twitter tweeps etc, etc, etc, you have presence on 1,000 social media streams. The reputation index would measure the total number of social media impressions your network has. In this example, it would be the aggregate number of events in Twitter feeds, Facebook feeds, MySpace feeds, etc that each member has times 1,000 (the number of members in your social sphere). So, let’s say that is 500 events times 1000 people in your network. You are now publishing into an ocean that is 500,000 events deep. Now, to measure YOUR influence inside that network, we would examine how many interactions( retweets, shares, etc) your postings receive relative to the entire number of interactions your entire network has. And in this ocean of events, your network each engages with 5 events. That means that the total ocean is (1000 x 500) or 500,000 events and each person in your network engages with 5 events (5 x 1000) or 5,000 engagements. That means that the entire network has an engagement rate of 5,000/500,000 or 1% This is the Network Engagement Index (NEI).
So, let’s imagine that into this ocean of 500,000 events, you contribute 10 events. So you represent 10/500000 (.002%) of the total ocean. We could express your expected influence (if you were just an average member of the network) as the percentage of events added times the Network Engagement Index (NEI) times the number of people in the network. This would be .002% x 1% x 1000, or .002 interactions per day. Because I like bigger numbers, we would multiply that by 100 to get the expected REPUTATION INDEX (.002×100= 2). The expected reputation index of someone in your network is 2. But imagine that you are the the person that your network wants to hear from…and your 10 events are interacted with by 500 people in your network. Your engagement index would be 10×500/500,000 or 1%. To calculate your reputation index, we would multiply your engagement index multiplied by the network engagement index by the number of people in the network (times 100 because I like bigger numbers) 1%x1%x1000x100 or 10. Your REPUTATION INDEX is 5x that what is expected in your network.
If we apply a concept like reputation index as a measure of authority, it can help sort the visibility of social activity to folks outside your network. The higher your index, the more visible you would be for relevant topics….
Your thoughts? Is this a silly equation(I was an English major in college, so maybe you math folks can make this easier or more clear?) Is this a way to help engines and watchers determine the influence and authority of social media streams? How would you do it. (Oh yeah, if you want to jump into my social stream, check me out: Twitter/Tim Kilroy)