Improving customer loyalty programs in B2B context with a data-driven approach
Lotsari, Otto (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202502193113
https://urn.fi/URN:NBN:fi:amk-202502193113
Tiivistelmä
This thesis explores loyalty programs in the Finnish technical wholesale sector and how such programs could be improved utilizing an approach based on data analysis and machine learning. The primary objective was to measure the performance of a loyalty program utilizing a variety of metrics as well as analytical techniques. The study focused on a B2B technical wholesaler operating in the Finnish market.
The analytical findings were used to create recommendations for improving the performance of the target organization’s loyalty program. The thesis also contributes to the academic discussion regarding the efficacy of loyalty programs as a whole while providing insight into the best practices for utilizing data in improving loyalty programs.
The theoretical framework consisted of a variety of topics relating to customer relationship management as well as data analysis. The framework was particularly focused on the concept of customer loyalty and how it is built and measured as well as how machine learning and data analysis can be utilized in customer relationship management.
The data was collected using various tools from the company’s database, a third-party webstore and a customer survey. The time period chosen for the data collection was January 2023 to August 2024. The data was collected in September 2024. The data was utilized in KPI and machine learning analysis. The KPI analysis relied on Excel-based calculations, while machine learning analysis utilized coding to provide predictions based on the collected data. This data was further analyzed and used for recommending improvements in the loyalty program.
The key findings of the research show that the target loyalty program functions adequately based on the fact that program members outperformed non-members across all major metrics. Machine learning results further highlighted factors driving sales growth among members. The recommendations for improving the program consisted of findings from data analysis and machine learning combined with insights from academic literature. The recommendations aim to improve outcomes by focusing mainly on the loyalty program structure and features.
The research did not seek to give answers regarding the technical and financial viability of the proposed suggestions. The scope of research is limited to the technical whole sector in the Finnish market. The research was also not able to find a causal link between increased loyalty and sales, though strong correlation between loyalty program membership and improve sales performance was discovered. Lack of resources for coding and data quality also proved to be challenges for the analysis.
The analytical findings were used to create recommendations for improving the performance of the target organization’s loyalty program. The thesis also contributes to the academic discussion regarding the efficacy of loyalty programs as a whole while providing insight into the best practices for utilizing data in improving loyalty programs.
The theoretical framework consisted of a variety of topics relating to customer relationship management as well as data analysis. The framework was particularly focused on the concept of customer loyalty and how it is built and measured as well as how machine learning and data analysis can be utilized in customer relationship management.
The data was collected using various tools from the company’s database, a third-party webstore and a customer survey. The time period chosen for the data collection was January 2023 to August 2024. The data was collected in September 2024. The data was utilized in KPI and machine learning analysis. The KPI analysis relied on Excel-based calculations, while machine learning analysis utilized coding to provide predictions based on the collected data. This data was further analyzed and used for recommending improvements in the loyalty program.
The key findings of the research show that the target loyalty program functions adequately based on the fact that program members outperformed non-members across all major metrics. Machine learning results further highlighted factors driving sales growth among members. The recommendations for improving the program consisted of findings from data analysis and machine learning combined with insights from academic literature. The recommendations aim to improve outcomes by focusing mainly on the loyalty program structure and features.
The research did not seek to give answers regarding the technical and financial viability of the proposed suggestions. The scope of research is limited to the technical whole sector in the Finnish market. The research was also not able to find a causal link between increased loyalty and sales, though strong correlation between loyalty program membership and improve sales performance was discovered. Lack of resources for coding and data quality also proved to be challenges for the analysis.