According to the book, ‘Marketing Metrics’, it is easier to sell a product to an existing customer (a 60-70% conversion rate) than to sell a product to a new qualified prospect (a 5-20% conversion rate). With existing clients, businesses already know their clients’ pain points, and the clients may have already become loyal to the financial institution. In the banking industry, banks often have a variety of products, but a good fraction of current customers might only utilize one or two products.
It can be challenging for a banker to sell a full range of financial products to a single customer, so front-line employees may only master a few of the high-performing financial products. Fortunately, banks have a valuable asset: customer data. With the right approach, a financial institution can evaluate their data and generate insights on cross-selling opportunities. This strategic approach to cross-selling is where predictive analytics comes in.
Here is how predictive analytics can be used for cross selling in banking:
The Power Of Predictive Analytics
The commoditization of financial products makes cross selling in banking arduous. Since customers may feel that they can get a better deal somewhere else, they might pick the most affordable product from your financial institution and hunt for other products elsewhere. This commoditization has resulted in banks bundling multiple banking products in an effort to create higher perceived value for the customer.
However, push-based selling and “one-size-fits-all” campaigns might not suffice to lure the modern-day customer. They need access to valuable products, and they need it now. Any product you bundle with the rest of your financial offerings should add the most value to their lives.
Given that banks collect data through CRM software and online tools, they can use this data to identify what their customers need. The data provides insights into:
What customer to contact first
What to sell them
How to communicate with them
Predictive analysis allows banks to evaluate buyer behavior through recent account activities and sometimes even online activities such as reviews and complaints. Instead of offering a single generalized offer, financial institutions can personalize their products to a specific prospect group which can improve a campaign’s return on investment.
Predictive Analytics Steps
1. Start With A Question
Banks collect vast chunks of data, and they will be nothing more than data without analyzing them. To successfully identify opportunities for cross selling in banking, they must create a question and look for its answers through predictive analysis. Unlike conventional business intelligence (BI) tools that tend to be retrospective in nature, predictive analytics tools should provide insights into the future. You can get answers to questions like:
What customer demographics are the most likely to churn?
What is the estimated number of leads the institution will get from a marketing campaign?
What are the odds of a customer purchasing product y after purchasing product X?
How profitable might a specific product package be over the next two years?
2. Collect Data
The next step is to identify and collect the data that might bring the bank close enough to the answers. However, the level of confidence a bank can have in its predictive analytics software will significantly depend on the quality of the data it collects. As long as the data meets a quality threshold, it will provide trustable insights.
For the financial institutions storing outdated, inconsistent, or even incomplete client data in their CRM, data collection can become quite time-consuming. As a result, the onus is upon bank managers to spearhead data quality management that lays the foundation for a streamlined process.
Data stored within banking CRM might not be sufficient for some predictive models. Banks might need to get data from other sources. Some of these sources include:
Bill payment behavior
Personal financial management
Wire & check payment data
Credit cards and debit cards
3. Build A Predictive Model
Next, data analysts have to create a predictive model that will define and determine the probability of specific events happening. These analysts can leverage artificial intelligence and machine learning methods, such as deep learning or linear regressions, to predict this. Once the model is created, test data has to be used to assess the predictive power of the model. For models that do not meet the expectations of the bank, they can be fine-tuned to offer higher predictive accuracy.
Data normalization can help increase the accuracy of a data model. Normalization helps achieve greater overall database organization, reduction of redundant data, improved data consistency within the database, and makes database security more manageable.
Once an accurate model is created, it can be a game-changer. Managers only need to feed their normalized data into the models and get the output they need to make decisions for cross-selling in banking.
4. Pay Close Attention To Assumptions
The idea that the future will always mirror the past is a major assumption throughout predictive analytics models. While there is some truth to this, consumer behaviors do change with time. If changes occur to the behavioral assumptions you might have made when creating your predictive analysis models; the models can become invalid.
Also, the variables of the models might change with changing market trends or time. For instance, the financial crisis of 2008 was significantly driven to by the assumption that house prices would always go up, which was not the case. Banks should pay attention to these assumptions to ensure that their predictions are still viable.
Predictive analysis isn’t a silver bullet for achieving cross selling in banking. Not all variables can be predicted to come up with trustworthy insights. Everything from the weather to the country’s political landscape can change buyer behavior. However, predictive analytics offers a much better solution for insightfully allocating marketing dollars than running financial marketing campaigns on underdeveloped research and half-baked ideas. Predictive analytics can provide financial institutions with a much-needed competitive advantage.
Reach out to our team at Insight Financial Marketing today to learn how you can get started with predictive analytics and how to translate changes in customer behavior into opportunities for your business.
Rob Reale is an Associate Partner and National Sales Manager responsible for business development and sales at Insight Financial Marketing. Rob began working in the Mortgage Banking industry in 1990 and currently helps the financial service industry leverage unique and innovative solutions.