XeeScore Update

Post viewed 3,577 times

XeeScore Update (as of Jan.1)

January 1, 2013 we released a new XeeScore algorithm and make it easier for users to understand its value and scale. Thanks to all the feedback, ideas and contribution in the process. The core principle of course will remain unchanged. The XeeScore remains to be a “Presence Value” Score indicating the value and attraction of a person’s or brand’s social presence.

The New XeeScore

The new XeeScore has a typical range from 0 to 100 and an extended range to 120 for extreme networker.

XeeScore 0-20

The social presence of a user is not very attractive or the person is diffusing the social traffic through various other uncoordinated channels. Indicating an opportunity for great improvement.

XeeScore 30-50

The XeeMe owner has an average engagement level and has a small to medium size network. The social presence is visited and appreciated. Indicating a good base with room for relatively easy growth.

XeeScore 50-80

The XeeMe is well used and visitors go visit the respective networks. The XeeMe owner is apparently a very active social media users attracting new connections to visit the various profiles. Indicating a very good engagement level and improvements are not very easy.

XeeScore 90-100

A true Power Networker who not only has a well set network but continues to grow the connections leading to more visitors and more social traffic. Usually a XeeScore above 90 is also accompanied with a Klout score above 70. The key task is to maintain the level by strengthening relationships.

XeeScore 100-120

Extreme networker. The score above 100 also indicating more than 1,000 visits per week to the respective networks, plus frequent visitors from other power networker and an extensive volume of new connections. On a thermometer this would be the red zone. Most likely it is not possible to maintain that level unless there is some massive automation involved that more often than not is counterproductive to the idea of relationship building.

What goes into the score

The new XeeScore is much more sophisticated than the first version. The old XeeScore basically measured the traffic from your XeeMe to your social sites. The new score algorithm takes the following information in consideration:

  • Number of incoming visitors
  • Number of outgoing visits to your social sites
  • Number of unique visitors
  • The ratio between incoming visitors and actual visits to your social sites
  • Faul up rate (pattern indicating abuse of the system)
The ratio between incoming visits and outbound traffic is called “Amplification Factor” and has quite a weight in the algorithm. In other words the higher the amplification factor the higher the score – even if the overall number is not outrages. Also the unique visitors are important to grow the score.
If the “Faul Up Rate” reaches a certain level, the system automatically switches to “faul up” it will emulate the behavior but no longer show the real values and after a week or two automatically remove all the reports from the user. The analytics part of XeeMe is quite demanding and we can’t afford to keep it up for users who game the system. Due to the way we compute the score, a faul may be recognized only after a few weeks.

XeeScore Update Cycles

The XeeScore is updated daily. However the update goes into a 30 day data pool so that 29 days of activity will easily “buffer” a few days of less traffic. The effect is that the score is no longer jumping up and down with a huge amplitude.

New “Score Bar” shows most relevant social scores

In addition to the changes of the XeeScore, we also added a  new “Score Bar” on top of the primary network list. The Score Bar gives visitors of a XeeMe an instant idea about network engagement and different types of scores. We selected the most relevant scores and update them daily. The current scores we use are:

XeeScore – indicating social presence value and number of social visitors
Klout – indicating influence and network activities
Kred – indicating credibility
PeerIndex – showing a comparison with peers
TrustCloud – a new score, exploring levels of trust

We use the respective APIs for all third party scores. The first visit maybe a bit slow but we buffer the score values during the day and reload it the next day. That process may cause a one day delay in the extreme case that we call the API a few minutes prior to their score updates.

Obviously the Xross Network Score (XNS) is based on the new XeeScore.