AN ANALYSIS of cUSTOMER SATISFACTION AND LOYALTYDRIVERS

A fundamental aspect of a firm’s long-term market success relies on itscustomer’s satisfaction and loyalty as evidenced through repeated purchases and greater future share of wallet. Therefore, an understanding ofthe drivers of customer satisfaction, repurchase, and future share of wallet is critical.

Our first objective is to examine what are the key drivers of customer satisfaction and customer loyalty (repurchase and share of wallet). The literature often assumes satisfaction leads to repurchase but customers may think of different factors when they think of satisfaction and purchase decisions. Therefore our second objective is to understand if there are any differences in the drivers of satisfaction and loyalty. For most firms, customer satisfaction is a means to an end-- sales (measured in terms of repurchase intent and greater share of wallet); hence understanding any differences in the drivers of satisfaction, repurchase or future share of wallet can be particularly insightful. Finally, valuedrivers may vary across customersegments. For example, a service-oriented business might rate HP differently than a manufacturing business and may have different drivers of satisfaction and loyalty. Our third objective is therefore to identify any such systematic differences among groups of firms both in terms of their evaluation of HP as well as on the drivers that affect satisfaction and loyalty.

A Conceptual Model of Customer Satisfaction,Repurchase, and Share of Wallet

Conceptually, we distinguish between satisfaction versus purchase loyalty related variables such as repurchase and share of wallet. We seek to understand the factors that drive these constructs of interest. The following diagramexplains this conceptual framework:

The Econometric Model

Thesurvey data has measures on three constructs of interest: (1) Overall satisfaction, (2) Repurchase decision and (3) Share of Wallet. These will be the dependent variables that we seek to explain in the regression analysis (to be discussed below).We will explain these constructs through drivers, which are customer ratings of HP performance on different dimensions (e.g., Ease of doing business…). Each of the six metrics of HP performanceis a composite measure based on several related questions.

Dependent Variable / Drivers (HP Performance Metrics)
1. Overall Satisfaction / 1. Ease of Doing Business
2. Integrated Solution
2. Repurchase Decision / 3. Service and Support
4. Global Business Understanding and Partnership
3. Share of Wallet / 5. Overall Relationship
6. Product/Technology Quality and Reliability

1. Drivers ofCustomerSatisfaction

The relationship between customer satisfaction and drivers will be modeled using multivariate regression.Customeri’ssatisfactionfrom using HP service ()can be expressed as a function of HP performanceratings provided by the customeri(, j= 1, ..., k)on the driver jas follows:

(1)

Here, αreflects the impact of driver x on satisfaction Sand will be used to obtain insights about the factors that are important for driving customer satisfaction.

2.Drivers ofRepurchase Decision

The relationship between repurchase decision and drivers will be modeled using multivariate regression. Customeri’srepurchase likelihoodof HP service () can be expressed as a function of HP performance ratings provided by the customeri(, j= 1, ..., k) on the driver j as follows:

(2)

Here, β reflects the impact of driver x on repurchase likelihood RPand will be used to obtain insights about the factors that are important for driving repurchase.

3. Drivers of HP’s Share of Wallet

Since HP’s share of wallet is measured on a percentage basis, the appropriate model to explain the drivers of share of wallet is a logistic regression. Let Pi be the HP’s percentage share of wallet. HP’s share of Customeri’swallet (Pi) can be expressed as a function of HP performance ratings provided by the customeri(, j= 1, ..., k) on the driver j as follows:

(3)

wheres represents a metrics identification variable (s = 1, …, 6). Here, Υreflects the impact of driver x on share of wallet Pand will be used to obtain insights about the factors that are important for driving future share of wallet.

4. Market Segmentation and the Drivers of Customer Insight

It is quite possible that there are systematic differences among groups of firms in their views of HP’s performance along different dimension. These different groups may also have systematic differences in what drives their satisfaction, repurchase and share of wallet. We therefore segment the firmsinto ggroups using cluster analysisbased on their performance evaluation decision. Then we do the analysis outlined in sections 1-3, on each of these groups. This will enable us to obtain segment specific drivers of satisfaction, repurchase and migration decisions.

Managerial Insights

In summary, the proposed model can potentially provide the following insights on the following managerial relevant issues:

  1. What HP performance metrics are the key drivers of customers’ satisfaction?
  1. What HP performance metrics are the key drivers of repurchase?
  1. What HP performance metrics are the key drivers of share of wallet?(Subject to SOW data being provided)
  1. Identify differences in the drivers on the satisfaction, repurchase and share of wallet decision.
  1. How do HP performance ratings and the key drivers of satisfaction, repurchase and share of wallet vary across customer segments?

Faculty Involved in the Project

Ravi Dhar, Professor of Marketing, YaleSchool of Management & Director YCCI

K. Sudhir, James L. Frank '32 Associate Professor of Marketing and Private Enterprise and Management,

YaleSchool of Management

Wonjoon Kim, Adjunct Professor, Department of Economics, New YorkUniversityResearch Fellow, YCCI

Responsibilities

During the course of this engagement, we will be responsible for:

Analyzing all research data provide by HPfrom their eRAP surveys, proprietary reports, documents, etc.

Produce and present the final report as described in the deliverable section.

Interim reporting on engagement progress.

Resolving issues, as appropriate, that may have an effect upon the outcome of the engagement.

Deliverables

A final report that will include:

  • A model of customer satisfaction, repurchase and share of wallet drivers.
  • A segmentation model identifying differing groups of firms in their ratings of HP performance
  • A segment level model of customer satisfaction, repurchase and share of wallet drivers.

1

135 Prospect Street • Box 208200 • New Haven, CT06520-8200

t 203-432-6069 • f 203-432-3003 •