Importance of Business Intelligence, Pricing Optimization & Demand Analytics for Financial Services Firms – Interview of Samir Jayaswal, SVP & Head of Operations, Prism Cybersoft

Importance of Business Intelligence, Pricing Optimization & Demand Analytics for Financial Services Firms – Interview of Samir Jayaswal, SVP & Head of Operations, Prism Cybersoft

Importance of Business Intelligence, Pricing Optimization & Demand Analytics for Financial Services Firms – Interview of Samir Jayaswal, SVP & Head of Operations, Prism Cybersoft

Ca33333pture

Please throw some light on business intelligence.

Business intelligence (BI) is a broad term to describe tools or techniques used to transform raw business data into meaningful business insights. BI tools are capable of sifting and reading large amounts of data and producing meaningful information from it. Good business intelligence can generate several business opportunities and insights which can be used for better pricing, better customer engagement, better customer retention and eventually better profitability.

What is Price Optimization?

Price optimization is also one of the examples of business intelligence. It is the science of predicting the price sensitivity for any product or service in a way where customers buy at that price and firms meet their profitability or any other defined objectives. Differential pricing for different segments of customers is a very common strategy adopted by financial services firms in order to attract or retain them. However, a firm cannot just lower its price in an effort to sign up more customers. If it does so, lower prices will lower the profitability. Similarly, raising the prices to increase profit also doesn’t help because higher prices will mean lesser number of customers will sign up. Thus, there exists a price for every segment of customers which will be an optimal price where the firm gets enough number of customers as well as profitability. Trying to find out this right price is price optimization.

Please illustrate with an example.

Let us take an example of a home loan. If a bank prices its loan higher than competition, it will attract less clients impacting its business volumes and in turn profitability. When it prices the loan low, it becomes attractive for customers and while they may sign up in large numbers, the profitability for the bank may be low because the margin is lesser. The bank needs to strike a balance and arrive at the right price for optimum profitability. Another example where price elasticity of demand plays a crucial role is when two customers start negotiating with the bank for better rates. Assume the bank has offered both customers loan at 9.5% per annum. How does the bank know how much discount to offer to each of these customers so that they sign up with the bank? Most salespeople would offer the maximum allowable discount and signup with the customer. However, star salespeople manage to offer the lowest discounts and yet sign up customers. Imagine if the bank could scientifically predict this rate and prompt the sales person to offer this rate where the propensity for the customer to sign up with the bank was highest, every sales person would become a star sales person. A few basis points saved could also result in huge money for the bank.

Is it always price vs profits?

Price vs. profits is the usual organization goals. However, organizations work under several constraints. For example, competition, regulatory diktat, firm’s internal beliefs and policies and so on. The process of price optimization keeps these constraints in mind while arriving at the right prices. Obviously, more the number of constraints, the more difficult it is to find the most optimal price.

What are the benefits for the financial institution?

Without this kind of demand analytics at different price points, financial institutions are blind while pricing their products. Take the example of financial institutions running campaigns during festival times. How does an institution know for sure that by giving a 10 basis points discount, whether the institution will be more profitable because of increase in business or will be in a loss because of lesser margins? Demand analytics produces insights that could deliver increased profitability, increase volumes, increased number of customers and better customer retention. In fact such analytics could help management spell out their key goals and formulate strategy so that they can achieve these key goals.

We need to make sure that customers, who are price sensitive and hence elastic, get rates which make them sign-up with the institution and not go to competition and customers who are price inelastic pay in a way where bank receives the maximum value. Price inelastic customers normally subsidize price elastic customers while price elastic customers bring in additional volumes. Such additional business and margin could improve the profitability of the business to a great extent. This fine tuning of rates is called as ‘Price Optimization.’ How much additional profit can such a strategy bring in? Well, that depends on a variety of factors. Typically, it could be between 5 to 20 basis points. This means that if the institution is operating on a profit margin of 1%, it can improve the bottom line from 5% to 20% directly by just optimizing the prices. Imagine how much effort the institution would have to put in and how many more on the ground sales staff it would need to deploy in order to achieve this increased profitability without price optimization.

Is this a new science?

The science of price optimization is about a decade old but financial institutions are increasingly discovering newer and newer areas for its applications. The earliest adopters were non life insurance companies, and then came banks and now NBFCs and brokerages are also doing pilots to implement this concept. It can be applied in different areas like retail, channel optimization, brokerage and commissions optimization etc.

What is the future of demand analytics?

The future of demand analytics is very promising. As competition increases, financial institutions are under more and more pressure to come up with innovative mechanisms of increasing their profitability. Adding branches and sales people is something every institution does. However, this may improve volumes but not necessarily profitability. They need to do something additional which is innovative and scientific. Profitability is directly linked to margins which in turn are linked to pricing. Hence pricing optimization is important. Institutions need to study demand analytics for various segmentation of customers at different price points. Currently, institutions don’t record what happens when a customer receives a price. However, to understand demand analytics, institutions must record what happened at that price? Did the customer accept or reject it immediately or bargained? If they bargained, how many times etc. This data can be put to good downstream use to calculate price sensitivity of customers and price elasticity of demand.

What is price testing?

Before rolling out the new optimized prices, the institution would like to check how these prices are being received by clients. Typically small samples of customers are chosen or maybe all customers in a particular city are chosen and new prices are rolled out to them to check their reaction. If they respond as per expectation, the new prices are rolled to all clients, else business intelligence analysts go back to the labs to fine tune their models and prices.

What about segmentation?

Market segmentation is a marketing strategy where customers are grouped into subsets when they are perceived to have similar properties and buying behavior. Institutions are currently segmenting their customers intuitively. However, more scientific methods need to be applied. Institutions segment customers so that there can be ease of marketing. However, when detailed behavioral analysis techniques are applied, every customer can be treated as a segment and specifically catered to, dramatically increasing his stickiness with the institution. That is the power of demand analytics.

What if the market is very competitive?

Price Optimization as a science and demand analytics with price optimization is meant to be used in highly competitive marketplaces. In monopolistic & oligopolistic markets, the customers would anyways take any rate you offer to them. In hyper competitive markets, it is very important to understand who are your price elastic customers and retain them. Such business intelligence help you bring in additional volumes by attracting and retaining these price elastic customers and improving profitability from price inelastic customers. It will also help you retain profitable customers and let go of customers who make losses for you.

How does demand analytics and price optimization help in customer retention?

Once a client signs up and sometime down the line he is not happy with the service or prices, he will start taking quotes from competing institutions. Since he is a ready customer, competing institutions will be more than happy to oblige him by giving better rates. In such a scenario, the institution normally responds with a retention offer in order to retain the client. Since this is the only chance the institution has, it must revert with an intelligent rate where the propensity of client staying back is very high. Building intelligence into rates in such cases is a function of how well the institution understands and implements business intelligence tools such as demand analytics and price optimization. Typically, after such implementations, an institution can expect to retain about 10% more clients than what it was doing earlier. This is a good tool to retain clients who are leaving only for price.

What is the kind of technology that goes into such analytics?

Demand analytics is more of a business problem than a technical problem. From a technology standpoint, the analytics tool needs to be comfortable in working and mining large amounts of data. It should also be capable of statistical analysis of data. The trick lies in defining the problem correctly, building correct statistical models and interpreting the results from these models well.

Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *