Customer profitability analysis is used to identify the most valuable customers or customer segments to prioritise marketing, sales and service investments.
The notion of using KPIs is to identify those key drivers or metrics that drive profitability and they differ from organisation to organisation.
In some cases, there may be many KPIs as the profitability metric is quite complex.
This is particularly the case with financial institutions and/or banks, says Richard Boire, partner, Boire Filler Group.
In these cases, there are many drivers of profitability such as bank fee services, lending revenue, investment revenue, attrition, credit risk etc. In other cases, the profit metric can be quite simple such as purchase activity for a retail organisation.
“The complexity of profitability can also increase if we incorporate the expense side of the equation which essentially represents the company's cost of engagement with a given customer. Engagement can be measured by looking at the number as well as the different types of marketing activities which a given customer receives,” said Boire, who is scheduled to speak at the forthcoming Business Analytics Summit to be held in San Jose (November 12-13).
The key in establishing profitability, according to Boire, is related to complexity and its enhanced impact on decision-making versus the additional time which is required in integrating all this information.
“Analysis can be done to measure the impact of each KPI metric within the overall profit calculation. This way, we can measure how much a given profit value changes with and without a given KPI. This kind of analysis can provide a roadmap for those KPI metrics which contribute significantly to customer profitability,” Boire told Business Analytics News.
Boire also spoke about other critical issues in an interview. Excerpts:
BI provides statistical techniques and data mining algorithms to analyse any number of customer attributes to uncover patterns in behaviour. How is refining statistical analysis helping in identifying sales opportunities and also customer segments that share behavioural trends and demographics?
Richard Boire: In every project we undertake where rigorous statistical techniques are employed, the $ benefit of our solution is always demonstrated. The use of business reports such as gains charts represent the key business information in being able to demonstrate the $ value of a solution. This is of particular importance when attempting to provide the $ benefit of predictive models. Predictive models are used to identify customers that are most likely to exhibit a certain behaviour. The $ benefit of this solution is derived from the increased performance lift caused by using the model.
For customers that share behavioural trends and demographics, segmentation strategies are commonly using more robust type statistics such as cluster analysis to help identify these distinct segments with their common behaviour trends and demographics.
According to BFG, reports need to be designed not only to identify enterprise wide trends (Dashboard Reports), but also to measure the effectiveness of various marketing strategies and tactics. What do you recommend when it comes to setting up a reporting system?
Richard Boire: In being able to develop a more focused strategy on a building a marketing measurement reporting system, we always undertake basically a three-step process:
Needs/data discovery analysis: This stage involves two main areas of concentration. First, we need to better understand the business environment in order to gain more domain knowledge regarding how the business makes money, how has it performed in the past, and what are the objectives going forward. Secondly, we need to better understand the existing data environment in terms of what data is useful or not.
Marketing data mart development: It is the deliverables of the first stage that provide the blueprint or roadmap for creating the marketing data mart. With this necessary information,the same kind of discipline attached to any database development approach is then employed. The difference with marketing data marts is that we are typically working with much fewer tables and data elements than seen in enterprise-wide database projects. This lessens the complexity of what we are delivering and as a result, we are able to deliver solutions in a much more timely manner.
Development of Marketing Reporting System: This last stage represents the real deliverable that is of most interest to the marketer. But we need to do the first two stages in order to do the third stage. This is where the expertise in terms of so-called 'Cube' Technology comes into play as end-user functionality is empowered through the user's ability to generate adhoc reports. Solutions of this type are developed with the intention of both increasing the breadth and depth of analytical capability throughout the organisation.
How should companies go about deploying a measurement system that results in closed-loop visibility to customer behaviours and interactions throughout the entire customer lifecycle?
Richard Boire: This really requires a process, similar to a campaign management process, where key behaviours and interactions are recorded at certain periodic intervals. These intervals represent points in time where we can observe changes in customer behaviour or interactions. The purpose of being able to track this information overtime is to identify changing customer behaviour/interactions but more importantly to identify those changes which are significant. Those changes which become significant should be flagged and so-called trigger reports should be produced to identify these customers with significant changes.
Companies whose customers purchase products through many channels often deploy analysis and reporting capabilities to far more users than most other companies, and having a BI platform that supports external users in addition to the internal requirements is essential. Which is the best way to avoid this?
Richard Boire: The notion of a heavy multi-channel environment where there will be many different users requires a much greater degree of organisation and control over all this information. The information needs to be centralised in some manner in order to provide a more holistic view of the customer. Although analysts specific to the digital world will be interested in the online activities of a given customer in terms of open rates, click through rates,etc., this view of a customer is limited if we do not integrate this information with the other channels that impact this customer(sales force,call history,direct mail,etc.).
Business Analytics Summit
Business Analytics News is scheduled to conduct the two-day Business Analytics Summit at San Jose in November (12-13) this year. The conference will feature leading Business Analytics executives including ones from Monster Worldwide, JetBlue Airways, New York Times Company, Boire Filler Group and Data-Miners Inc.
For more information, click here: http://www.businessanalyticsnews.com/usa/agenda.shtml
Or contact: Ben Satchwell by email ben@businessanalyticsnews.com
Recent Comments
Start the Conversation!