Knowing your customer; Data Science in the service of Banking
Banks need to understand their customers in specific groups of similar needs in order to forecast bank operations and brand sales. This helps the bank get customer’s view on their brand, where, when and how the brand is consumed for better servicing of their customers in future. Banks also need to safe guard against fraud, how customers are serviced as far as their specific needs are concerned, customer profiling and getting informed of marketing strategies.
The churn challenge
For any bank, customer satisfaction and retention is important. Customer sign-up represents a significant cost for a bank; and customers who churn (meaning they stop using their bank products within a specific period) mean the bank doesn’t earn sufficient income to cover those upfront costs. Understanding patterns in churn behavior can allow the bank to identify areas of product and service dissatisfaction and can allow the bank to better forecast its operations and profits. Understanding this behavior is relevant for the activities of multiple departments.
Customer data
This consists firstly of static customer demographics data and secondly of dynamic transaction history data between Yr. 1 and Yr. 2 for instance. While a simple extraction of the different variables could provide a static table of the percentage of churned versus loyal customers per customer characteristic, the combination of these variables and data models allows banks to gain deeper insights. For example, key predictors of customer churn could be found in specific combinations of products, in the length of contracts and in customer age.
The sentiment challenge
Although the bank has internal data dating back to many years, tracking the activities of its customers and the various products that perform better or worse, this only represents the “what” of the bank’s business performance, not the “why”. For a deeper understanding of why their customers decide to join or leave the bank or why they appreciate certain products and disregard others, the bank has to rely on additional market research.
This can consist of engagements by its customer services department, customer surveys or focus group discussions — all of which can be relatively expensive and time-consuming. Moreover, the bank does not have easy access to information about the “why” of its performance compared to its competitors.
Banks can therefore identify the need to leverage publicly available social media data to perform a sentiment analysis. The purpose of a project like this could be a threefold investigation to understand sentiments of;
1) Current customers towards the bank and its products
2) Rwandan customers towards competing banks
3) the characteristics that set the bank’s products apart from those of its competitors
Twitter data on market sentiment
To start, the popularity of keywords related to financial services could be analyzed using Google Trends data. Working through a long list of keywords (e.g. ATM, mortgage…), then you compile the most popular keywords related to financial services in Rwanda for example. Subsequent analysis should be done on these keywords only. Extracting from Twitter, filtering to only include tweets from Rwanda in the last three years. You extract all tweets that mention the prioritized keywords and save them in a static CSV table for analysis.
The variables extracted in this way include all text, interactions (tags, likes…), user demographics, time and date, and location of the tweet. In my case, I was able to identify which bank was being spoken off by analyzing the tagged Twitter accounts and text in the tweets. By using a Lexicon Cloud service, segmented the tweets per keyword and per bank according to positive, neutral or negative sentiments.
Through analyzing the balance of sentiments across keywords and banks, I was able to rapidly answer the following questions on financial clients; 1) Feeling about the different products offered by the bank I work for; 2) Feeling about the different products offered by other banks in Rwanda; 3) Feeling negatively about, which competitor banks offer the same product that customers feel most positively.
Collaborating directly with different departments (as marketing department) allowed us to also answer specific strategic questions they had, based on the retrieved Twitter data.
Process recommendations
The Lexicon Cloud service utilized for the sentiment analysis only had English functionality, which means that tweets in French and Kinyarwanda were disregarded. I would also recommend that other organizations interested in similar projects investigate applying machine learning to the text of the tweets to allow for more rapid summarizing of themes.
Finally, organizations interested in using such a project to analyze their position in the market would need to ensure that their social media engagements are sufficiently high — to be certain that they could retrieve enough tweets that have tagged their bank. In the absence of a social media presence of the bank itself, it would still be possible to analyze the sentiment of Rwandans to financial products and banks in general.
By Samuel Philip Mutinda, AIMS Rwanda Alumni
Samuel Philip Mutinda is a Data Scientist experienced in machine learning and artificial Intelligence. He holds a Masters in Mathematical Sciences from AIMS, Rwanda and a Bachelor’s degree from South Eastern Kenya University, Kenya.
