I’m working on some new tools for Caplin Trader that will help banks better understand their customers.
Detailed knowledge of client behaviour is vital for any bank operating in a competitive environment. From a client’s perspective, the bank should understand what the client does, why they do it, and most important how to help them stay in business. Reports I have seen indicate that banks have some work to do to address client expectations in this area.
As part of the research, I thought I would investigate the market leader in understanding client behaviour, i.e., Tesco. I would recommend anyone interested in this topic read the interview with Clive Humby here.
Themes to be drawn from Tesco’s experience
While a bank cannot hand out loyalty cards to its SME, corparate or institutional customers, there are some themes that can be drawn from Tesco’s experience.
- This is not a magic bullet. Just knowing more about customer behaviour does not equate to improved flow, profitability or a reduction in costs. The data is merely the first, albeit most important, step in the chain.
- It is non-trivial to (1) capture tha data in a contiguous set (2) normalise the data so that it can be used by other systems & (3) make the data available to any – current and future – system.
- The bank must understand that the information can add significant value to the client relationship. This attitude must be engendered from senior management.
- The data is no substitute for the inter-personal relationship between a relationship manager and his/her clients. An experienced RM/dealer can add enormous tangible and intangible value to the relationship.
Where can the technology help?
From a bank perspective, highly experienced people are a key differentiator but they are thin on the ground and are expensive. This point has been made by my colleague Paul Blank here. Where technology can help is to efficiently provide the raw inputs to the business so that all client behaviour is captured for analysis.
I have been working with Mike Poston to provide a solution to this issue. This is still a work in progress but initial indications are encouraging. The other takeaway (in addition to those above) is “capture everything at the earliest point” and don’t make assumptions about how the data will be used.
We have more work to do before making the package available but I am encouraged by the initial feedback.