Using Data for Corporate Lending

June 15, 2018 / Probe Research

The retail lending space has changed rapidly over the past two decades, thanks to the use of data and technology. We are starting to see a similar transformation in the business lending space, through use of public data and automation. Based on the trends we have seen over the past six years, we believe that keeping the following in mind will help you get maximum mileage out of using data.

1) Measure the benefits

In most cases, there is an existing lending process. This process is done using an ‘old way’ (for want of a better term) of doing things. With the availability of data and with technology, the same process can be done more efficiently. The right way to approach the problem is to:

  • Measure the overall cost of the existing process (end-to-end cost)
  • See how the process can be changed using public data that was not available earlier
  • Gauge what the material change will be, in the overall cost of the process
  • Run a controlled prototype of the new process
  • Roll out the new process, if it makes sense

One of the challenges we see is that the approach to using ‘data’ is done tactically, without taking an integrated view of the whole process and the more significant implications.

2) SaaS model

In the new disruptive world (which, one has to admit, has made a significant change to the way we live our lives, all in just the past few years), you can use SaaS-based solutions to solve most pain points. The advantages of this model include:

  • Quicker trial and implementation: See benefits rapidly or move on if it doesn’t work out
  • Easy implementation: Does not require any significant upfront technological or financial commitments
  • Pay-as-you-go model

In a world where managers are used to conceptualizing large projects that consume a lot of time and require significant upfront investments, this approach is both critical and effective.

3) Structured vs unstructured data

There is much talk these days about AI, NLP, Machine Learning, etc. With social media and various other sources of information, the potential of unstructured data is enormous. By mining and using unstructured data mining to its potential, your process can go a long way in terms of efficiency and intuitiveness. In the process however, don’t lose sight of ‘good old structured’ information. After all, only a strong foundation of structured information can form the basis of unstructured information and analytics. Note that you can’t analyze without having good clean data in hand. Analysis without data is just an opinion.

4) Good data costs, but pays back quickly

In today’s world, led by the Internet and Google, one gets a lot of information freely. As a result, we do see a strong underlying mindset that all data is of ‘low value’. In reality, good quality data takes a lot of time to develop and money to build up. More importantly, having a foundation of useful quality data saves significant costs down the road. One would be surprised at the number of processes that exist out there, where the sole purpose is to capture and cleanse data. Instead, if good clean data had been available at the source, much of the downstream costs could have been saved. Build on a foundation of high-quality data, and the downstream benefits can be massive.

5) Keep an eye open for strategic advantages

In an industry that has been around and regulated for several decades, there are several outdated processes that have been adopted by too many people. Fear of questioning the status quo is high. Technology and data are changing things rapidly. What is even more interesting is that the regulatory system is also learning to question some of these outdated ‘requirements’, and is willing to adapt to things that benefit the end borrower (in this case, a business).

In 1995, who would have thought an individual borrower could get a loan in 30 seconds? This would not have been possible until the day someone believed ‘it can be done’.

Similarly, data and technology have started to throw up several options which question the status quo. If you understand the tools available (read data and technology), and approach it with the intent of benefiting end borrower — while bringing down overall costs, improving credit quality, and reducing cycle time – bold thinking can take you places.

The Uber, Swiggy, Paytm kind of disruptive ideas are not limited to just the B2C world. Lots can be done in the B2B world too.