Analysis of customer
|Analysis of customer|
- identification of customers
- gathering information necessary to understand customer needs
- dividing customers in sub-groups with similar characteristics
- gathering data about customer needs and behaviours
- asking customers about their satisfaction from past interactions and purchases
- analysing profitability of customers or groups of customers
Gathering basic information about customer leads to creating successful and profitable CRM strategy (customer relationship management) and deploying more personalized marketing activities. Developed CRM helps with understanding customer behavior, acquiring potential clients and maximize customer value. To gain the needed information for analysis there are various methods:
- periodic survey – directly way to determine customer's approach to the company and satisfaction from the purchases
- number of lost customers – enable to see the size of rate
- analysis of consumer complaints – identify the issues with offered products.
Methods of analysis
The concept is related to the income and it is based on Pareto's rule: 20% of customers generate 80% of total turnover achieved by company. Contribution to sales split customer into three categories:
- A – the most valuable for the company (the highest return)
- B – average contribution in amount of the turnover
- C – the lowest money spent
ABC method is based on the past data what due to the seasonality or other fluctuation can lead to mistakes in proper customer segmentation.
Customer Lifetime Value (CLV)
To analyze large database related to personal information and customer transaction, there are few tools which enable to assess the customer equity. One of the most popular is the concept presents value of all profits engendered by customers in the future – Customer Lifetime Value (CLV). The method has a lot of applications:
In literature there is six approaches distinguished in CLV model:
- RFM – based on Recency (the number of periods since the last purchase), Frequency (number of orders in the certain periods), and Monetary (the average money spent in the periods considered)
- Probability model – based on Pareto/NBD and Markov chains
- Econometric model – based on Pareto/NBD with including customer acquisition and retention
- Persistence model – based on behavior
- Computer Science model – based on theory (like utility theory)
- Diffusion/Growth model – based on customer equity
Data mining is the automatic (through mathematical and statistical techniques) process of searching useful information in large database and creating novel and beneficial pattern. The knowledge gained through this advanced analysis is predominately used for forecasting and decision making due to possibility of identifying valuable customers and predicting their steps. The model is divided into two categories: descriptive (classification) and predictive (clustering). The descriptive approach means looking for a model that enables to describe data classes, while the clustering analysis involves data object within unknown class label.
The implementation of customer analysis in practice is sometimes limited by important barriers. The value of taken activities depends on the availability and quality of data which are becoming critical aspect of model correctness. Deploying some technique requires data comes from different business area as well. Data ownership and responsibility are define by the organization what usually create a need to expand the accesses and shift the data privacy. Integration analytical model into company processes is successful when it meets with deep engagement of all employees. Complexibility and development of models are one of the most common reason for failure in analysis customer data. Understanding and deploying them are too difficult and time-consuming for company's managers. In addition, the result of analytics do not match often with the decision made by the management team what leads to the next investigation needed.
- Bahari F., Sudheep E.M., (2015). An Efficient CRM-Data Mining Framework for the Prediction of Customer Behaviour. Procedia Computer Science 46, 725 – 731.
- Bijmolt, T. H. A., Leeflang, P. S. H., Block, F., Eisenbeiss, M., Hardie, B. G. S., Lemmens, A., & Saffert, P. (2010). Analytics for Customer Engagement. Journal of Service Research, 13(3), 341-356.
- Khajvand, M., Zolfaghar, K., Ashoori, S., & Alizadeh, S. (2011). Estimating customer lifetime value based on RFM analysis of customer purchase behavior: Case study. Procedia Computer Science, 3, 57-63.
- Radulescu V., Cetina I., (2012).Customer analysis, defining component of marketing audit. Procedia - Social and Behavioral Sciences 62, 308 – 312.
- Khajvand, M., 2011, pp. 57-63
- Bijmolt, T. H. A., 2010, pp. 341-356
- Bahari F., 2015, pp. 725-731
- Radulescu V., 2012, pp. 308-312
Author: Justyna Kurnik