Analyzing the voice of the customer is an important way to measure and improve satisfaction. In this blog, we'll demonstrate how to turn a large number of text comments into actionable information to better serve your customers.
Let's discuss a use case for a satisfaction survey conducted by the French Sports Association, which ranks in the top six in terms of the number of license holders. This license enables members to practice Xi in affiliated clubs and participate in official competitions. The association asks its members to rate their satisfaction with the licensing process.
The driving force behind this licensing process improvement project is to reach potential members more effectively and encourage members to renew their licenses. Sports federations are funded by institutional subsidies and sponsorships, proportional to the number of members. Maintaining or even increasing the number of licenses is critical and is directly affected by satisfaction with the licensing process.
Three main problems were identified that gave the impression that the licensing process was unsatisfactory:
The number of 50,000 undocumented players is growing.
Other sports associations seem to have provided licensees with more modern tools.
Affiliated clubs find the processing of license applications cumbersome.
The sports association decided to launch an improvement project to address these challenges.
Improvement project begins
This Gantt chart in Minitab Workspace details the steps and timeline for your project.
The sports association consulted with regional and local associations and focused on the following elements:
Licensing Procedures for Other Sports Associations.
Legal Obligation. Price.
Segmentation (player characteristics, such as age, gender, geographic location, type of exercise Xi, etc.).
* Licensed platform.
Licensee's Opinion.
In order to gather opinions from licensees, the association conducted a survey to obtain feedback on compliance with the following criteria:
The survey should not take more than 6 minutes to complete.
The analysis of the answers should provide metrics such as ratings, ratings, and suggestions for improvement.
The sample responses should be representative of the different classes of licensees.
A questionnaire to collect opinions and ideas
200,000 questionnaires were sent,**19,921.
The survey platform provides some basic descriptive statistics for an initial assessment of the results. A breakdown of responses to demographic questions proves that the sample is fairly representative of all categories of licensees. The association found that the majority of members did not experience difficulties in the license renewal process. One-third of potential licensees seek help from their clubs, and another third consult** for help.
The basic charts and statistics provided at this stage do not take into account differences in reaction between classes of licensees. Does geography, age, gender, seniority, or practice influence opinions?
Moreover, this interpretation of the findings ignores textual comments. The next question is "What valuable insights can be drawn from these reviews?".”
For further research, Minitab statistical software was introduced to provide data analysis.
The output of principal component analysis (PCA) was explored to identify discriminatory criteria. The team realized that the respondent's geographic area had no effect, so this parameter was excluded from the later analysis. Contribution maps help teams visualize clusters, report similar complaints or requests that can justify program changes or customizations.
The load plots in the Multivariate Analysis menu in Minitab's statistical software reveal several trends.
This analysis shows that young players and beginners are interested in training during the school holidays. Adult gamers want mobile apps, tournaments, tournament tickets, and merchandise from the store. Regardless of age, athletes want to have an account with a sports association and benefit from additional services.
Text mining to better detect suggestions
19,921 out of 800 participants answered open-ended questions, so the team thought it would be useful to analyze these comments using text mining.
Semantic analysis and word clustering, discovery of phrases and topics led to some interesting insights.
The most popular word (using wordstat).
Similar comments about the steps of the permit application were repeated many times, such as the documentation required, the complexity of the procedure, the necessary simplification, and the breakdown of the cost.
The most popular phrases.
Phrases that are repeated multiple times refer to the requested certificate, fee, payment, and signature.
Seven meaningful themes were identified.
Topics that appear include breakdowns, electronic signatures, simplification, license models for managers, and required certificates.
The treemap above reinforces the conclusions drawn from the previous **.
Recommendations based on an in-depth analysis of the opinions expressed
Through further analysis, the team was able to make recommendations for improvement. It is advisable to sign the application form electronically and to simplify the **platform, eliminate redundant paperwork to validate volunteer licenses, and better communicate around the free services available.
Before the next round of license subscriptions, the process and ** have been optimized.
Opinion surveys help measure customer satisfaction. Valuable text feedback is often overlooked. Without semantic research, it's tedious to consider every line of text. The challenge is to distinguish between the opinions expressed in certain customer profiles. These can be the key to decisive decisions to improve customer satisfaction.