Personalisation Approaches in SharePoint 2010


Digital Marketing Personalisation

There are three primary types of personalisation used across Digital Marketing today - these are:

  • Declared Personalisation - the user "ticks boxes" or includes words such as "[X]  Camera Enthusiast", "I like: Cameras, Walking".  The SharePoint 2010 Audience Targeting feature is a "declared" approach to personalisation
  • Learnt Personalisation - An approach whereby the user's interests are learnt from their behaviour - for example if a user "mostly" clicks on high-end "enthusiast" cameras, then the system will automatically 'learn' that preference and provide products that are relevant to the 'enthusiast'. 
  • Anonymous Personalisation -  Personalisation in which the interests of the user are 'learnt' in one location, but then applied as the user arrives at the SharePoint 2010 website.  A solution in this space is AdRock - a Microsoft solution that is able to learn personalisation from a user's behaviour on a site such as NineMSN, but then be able to use this behaviour on other sites such as Contoso.com.   This is termed 'anonymous' personalisation as the Contoso website has does not have any declared interests from the user, and neither does it have "learnt" personalisation from previous behaviour on the Contoso website.  Instead - behaviour from the Atlas advertising network is passed to the Contoso website (with the user's permission) to allow the Contoso website to provide a more personalised experience the moment that the user's Live ID is activated.

Audience Targeting - Declared Personalisation

Audiences are part of a User Profile service application that enables organizations to target content to users based on a flexible set of rules. Audiences can be defined by one or a combination of the following items:

  • By public properties in user profiles
  • Active Directory
    • Membership in a distribution list
    • Membership in a Windows security group
    • Location in organizational reporting structure

An audience is defined by a collection of one or more audience rules and by whether all or only one of the audience rules must be met when evaluating membership. An audience rule can be based on membership in a user profile
property (and other active directory related roles/groups). In a Digital Marketing scenario it is common for user's to nominate specific interests - and to store these in the user profile.  To define each audience rule, you must
select an operand, operator, and value.

Once an audience has been defined, it must be compiled on a regular basis because the underlying user profile properties and membership in directory services and groups can frequently change. An administrator schedules the timer job that controls when audiences are compiled.

See also: http://technet.microsoft.com/en-us/library/cc261958.aspx

 

MagiClick BehaviourPad- Learnt personalisation

http://www.magiclick.com/BehaviorPad/

BehaviorPad helps maximize conversions & revenues by enabling the
administrator to dynamically change the website's content based on visitors'
data and their tracked actions.    BehaviorPad provides a simple way to
implement behavioral targeting logic on your website with minimal or no input
from your IT department or development agency.

The process is as
follows:

  • Select the page on your website, on which you want to implement behavioral
    targeting logic
  • Select your targeting criteria
  • Select the action you want your website to take when the criteria are
    met

Available rules include:

  • Referrer - Which website did the visitor come from?
  • Search Keywords - If the visit has originated from a search engine, which
    keywords has the visitor been looking for?
  • User Agent - Which browser, operating system, platform etc. is the visitor
    using?
  • Geolocation - Which country or city is the visit originating from?
  • Visit Count - New or returning visitor? If returning, how many times did he
    or she previously visit our site?
  • Page/Directory Visits - Which pages or directories has the person previously
    visited on our website, and how many times?
  • Cookie Check - Does the visitor have a cookie we have placed previously by
    our site?

The solution also includes concepts such as "if the user visits type of page
X twice"

Please note that other solutions that cover not just personalisation are available - for example the IntLock SharePoint marketing Suite

* http://www.intlock.com/

Comments (0)

Skip to main content