Here’s an excerpt from a candid conversation between Devashish Fuloria, Co-founder and CEO at GeoIQ, and Markandey Upadhyay, Head-BIU at Piramal Finance.
The Challenge: Reducing drop-offs and unserved user instances with alternate data sources
Being in the lending business, we get all sorts of different loan applications. The traditional way to get more information about these applications is through bank statements, bureau data, and the likes of it. Although these are powerful data sources, it is possible that for some applications there is no data with the bureau. Also, for small-ticket loans, some users might not be comfortable sharing their bank statements.
Now, these instances contribute to the shooting drop-off rates and a higher number of unserved users. So we wanted to explore alternative data sources that could help us make smarter decisions.
How did you come to know about GeoIQ?
In our strive to explore alternate data solutions that we could use to improve our models, we came across GeoIQ as a prime provider of location data and solutions for Indian boundaries. We have been drawing data from you and another alternative data provider in the location data space for more than six months now and the data is extremely valuable. I also believe the extent of usefulness of this data will evolve over time as we consume more and more data. This has been a promising experiment and we would continue to be invested in this process.
How do you witness growth with Location data?
I believe we need to give the model and our association some more time before we see any transformational impact as it has been just six months since we started using location data in our models for cutting population. But as of now, there are some metrics where we can observe movement in terms of impact. Rank ordering is better than any other tool which we have been using for cutting population. Therefore, we intend to use this for the long term when it comes to bucketing and cutting the population for loan applications.
Although, I am just adding a disclaimer that our model is just not about location data, rather it’s a combination of multiple other datasets and variables. Having said that, location data plays an important role in the sanctity of our models.
How has your experience been working with us? Has anything exceeded your expectations?
When it comes to the data, the experience has been good and the data is promising. But as I mentioned earlier, I would have to give it some more time before we see a transformational change that goes beyond expectations. Although, your APIs, the turnaround time, and the means to exchange data are really good and can seldom be seen and experienced while working with legacy companies, which I do not find to be this technologically abled as you folks are.
How likely are you to recommend our services to others?
I think your product is not as simple as serving a pre-maid delicacy like biryani or something. I think you guys provide great ingredients for businesses to make unique dishes of their own. Now, what they make out of these data ingredients and how they use them depends on their intelligence and abilities. I truly believe that if a business is ready to invest time and resources with patience and the executives are smart enough to enable and appreciate desired outcomes, you provide some really good ingredients for that. Again, what businesses make out of your data variables and ML capabilities will vary from organization to organization. On a scale of 1-10, I think I will give you a 7 on the recommendation score.
Describe your experience with us in one word.
Promising
Is there something that would not have been possible without our solutions?
When we talk about loans, we do not stop anyone from applying, and then when someone buys we make a decision. Now we wanted these decisions to be smarter. What we look forward to is the quality of these decisions. The early signs that I am seeing,
This is what I am seeking to do with this data and once I have ample confidence, I will start using this data for other applications.
Although, I must say that in the lending business, we do not make hasty decisions. We strive to do incremental things in the right direction, continuously, and over a long period of time. We avoid doing dramatically different things in a shorter time span. Even though the data is promising and a critical part of our analytics model, we will have to wait and see the bigger impact that it leads to.
Conclusion
GeoIQ has been able to provide valuable location data solutions to Piramal Finance, helping them in their pursuit of smarter decision-making and cutting population for loan applications. Although it has only been six months since the collaboration began, Piramal Finance has already seen some positive movement in their metrics. While Piramal Finance will have to wait and see the bigger impact that the collaboration leads to, they believe that GeoIQ’s data variables and ML capabilities will play an important role in the sanctity of their models in the long term.
Access a detailed case study on “Serving NTC Customers with Location Data and Intelligence”
To learn more about our solutions, visit: https://geoiq.io/