Linkage Analysis

Why is linkage important, what are we linking, how do we do it, and what are the challenges in successful implementation?
 
 

Why is Linkage Important?

Linking key data is important because the goal is to integrate the voice of the customer into business practices and processes, resulting in a change in customer behavior. It builds on the premise that understanding customers’ needs is important and that there must be a structured process to identify, prioritize, and fulfill customers’ needs at the aggregate and segment levels. This provides a basis for improving performance within and across organizational boundaries and for management’s decisions regarding investments.

Driving these decisions is the assumption that leveraging the voice of the customer will deliver financial benefits, that customers are an investment, which should be managed so that financial return is maximized. This is important because senior management and stakeholders are increasingly demanding evidence that their investments in the company’s customer-centered initiatives do indeed have a favorable impact on the bottom-line. One final point. We cannot really understand the business impact of decisions unless we link the customer inputs with the financial value of the account.
 
 

What Are We Linking?

There are three levels of linkage.

1) One involves linking disparate sources of data for one facet of the enterprise, e.g., customer feedback. In this case, the intent is to integrate different sources of customer intelligence to truly give voice to customers. Whether it is feedback from different surveys, account manager inputs to a CRM system, or notes taken by agents at the call center; whether it is quantitative or qualitative (text-based) data, all serve to deliver a comprehensive view of where customers are coming from.

2) The second is to bring non-financial (customer, employee, operational) and financial measures into the picture, literally combining them in a meaningful way on dashboards, so that the relationship is immediately apparent. As the time to solve a problem increases, does customer satisfaction decrease? As overall satisfaction increases, does sales revenue increase? As employees’ ratings about the availability and quality of tools they are given to solve problems for customers drop, does customer satisfaction with agent expertise also drop?

Looking at the employee/customer connection, for example, we ask employees about the tools, processes, management support, training, etc., that they have access to so that they can deliver high service quality to customers. We identify employee key drivers and areas of employee dissatisfaction. We also analyze customer feedback, based on their key drivers and areas of their dissatisfaction. Decisions to improve the level of employee satisfaction can then be made. For example, if customers rated agent expertise poorly, and agents indicated that lack of training was hindering them in their ability to resolve issues, then attention should be given to the relevance and clarity of training courses.

3) Lastly, we seek to quantify relationships, so that we can predict business outcomes if certain steps are taken, i.e., investments for change are prioritized and implemented. First, we establish whether that linkage exists. Secondly, we set up a statistical model that identifies the drivers of customer satisfaction. The hypothesis that we seek to prove is that through specific improvements in any of those drivers, such as satisfaction with professionalism of the call center agent, overall satisfaction with Support will increase, which will lead to an increase of overall satisfaction, customer retention, and ultimately increased revenue. These insights tell Sales, Support, Engineering etc., which actions will have the highest impact on the part of the customer experience they are responsible for.
 
 

Challenges

Successfully implementing these linkage exercises has many challenges, including:

• Identification of the right data

• Ability to locate, access, and assemble the data

• Owner’s permission to use the data

• Organization’s willingness to share the data with the third party who will do the modeling and analysis

• Resources to assemble the data and populate the models

• Resources to perform the modeling

Integrated Quantitative Analysis Methodology

We offer expertise in the use of advanced analytical tools for more comprehensive analysis, which can uncover interrelationships that might not be evident from tracking individual metrics.  Customer satisfaction, employee satisfaction, operational metrics, and other indicators are often dynamically interrelated in ways that benefit from a more holistic approach to assessment and solution development.  Some of the advanced quantitative techniques we are experienced in applying to integrated solutions include:

• Mutiple regression
• Factor analysis
• Cluster Analysis
• Discriminant Analysis

Back to top