Thursday Mar 08, 2012

The Era of the Decision Graph

Gone are the days when “electronic billboards” for targeted merchandizing programs were leading edge.

Over the course of the last few years we observed a dramatic qualitative shift in how companies have applied Analytical Decision Management techniques to drive Customer Experience Optimization programs. It used to be the case that marketers were happy when they were allocated a dedicated piece of real-estate on their company’s web site (or contact center or any interaction channel for that matter) that they could use at their discretion for the purpose of one-off targeting programs. What companies now want is granular control over the whole user experience so that the various elements composing this cross-channel dialog can be targeted and relevant. Such a shift requires a new approach to analytics, based on understanding how the various elements of the user interaction relate one to the other.

Let me introduce the concept of Decision Graph in support of this idea.

To move from electronic billboard / product spotlight optimization to customer experience optimization, analytics must shift away from focusing on “the right offer for the right customer”. The focus of the Decision Graph is to identify “the right user experience for the right customer”. This change of focus has a critical impact on your analytics requirements as one dimensional targeting approach for matching customers with offers won’t address the need to optimize multiple dimensions at once.

This is where the Decision Graph comes in. Let’s consider the following graph eliciting the relationships between the various facets of the user experience to be optimized in the context of a Marketing Optimization use case.

Now imagine that for every offer presentation on any interaction channels, your analytical engine can record and identify the characteristics of the customer interactions that are associated with success (say click or offer acceptance) across all those dimensions.

Let’s take an example.

  • You see a nice picture of bear cubs on a forest background with a punchy banner stating “please give us back your share of the 20,000 tons of annual account statements” call to action to sign-up for electronic bill payment on the “recommended for you” section of the login page of your financial service web site and … you decide to click on the “one click wildlife donation” link.
  • Our Decision Graph, can then record the fact your customer profile is positively associated with “positive responses” to marketing messages in the following context: Channel (Web), Offer / Product (Electronic Bill Payment), Creative (The Bear Cub image), Tags (Environmental, Wildlife, Donation, Provocative), Slot Type (Image), Slot (Recommended for you), Placement (Login Page). As predictive models are attached to the Decision Graph, this means that such a business event updates 10 predictive models that marketers can now use for reporting and decision management purposes.

You can now generalize the idea and imagine that this graph collects information about all marketing events across all channels and you end-up with an analytical system that let all the actors of customer experience optimization discover the relationship between the different facets of user interactions

With the Decision Graph

  • Marketing stakeholders will learn about customer segments that are receptive to eco-centric marketing messages and which customers in the right context will step out of their standard routine (why they came to web site in the first place) to subscribe to specific causes.
  • Web user experiences stakeholders will learn about which type of marketing messages are appropriate and for whom at the start / at the end or throughout a logged-in web session
  • Content owners can focus their digital agencies on the most effective creative themes as they will be able to correlate response rates based on associated tags
  • And the company as a whole will have learned who is receptive to eco-centric marketing messages when displayed in a given context of a secured dialog from which it will be in a position to dynamically tailor user experiences across channels based on such empirical evidence

Now contrast this with a system that would only record the fact you’ve subscribed to the Electronic Bill Payment option as part of the “Go Green” Campaign and you will get a sense for the power of the Decision Graph. The bottom line is that companies need analytical systems that operate at multiple levels of the Decision Graph if they want to delight their customers with relevant customer experiences.

My next post will be on how the Oracle RTD Decision Manager product enables you to create and configure such graphs and to automatically identify the predictive drivers of response across the whole spectrum of the user experience.

About

Issues related to Oracle Real-Time Decisions (RTD). Entries include implementation tips, technology descriptions and items of general interest to the RTD community.

Search

Categories
Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today