In the last few months, spurred by the growth of Fintechs / digital lenders/ smart finance solutions, the conversation around “alternate data” has grown. For simplicity’s sake, let us define alternate data as “all such data which companies have not used traditionally”, and explore new avenues, in addition to the one's offered by the Mastercard compendium.
Alternate data usage started in the world of alternate lending, where businesses were looking at data similar to bank data, such as payment history of utility bills (postpaid mobile, electricity). Additionally, some markets also looked at alternate behavioural data such as your social media presence (what you like and don’t on Facebook), calling patterns (“are you a late night caller”, “do you recharge often”), location stability (“are you on the road too often”). The primary purpose of alternate data was to enable lending, and drive financial inclusion (credit for the underbanked/ under-served.
In India, the Fintech/ NBFC space is full of companies exploring alternate data – LendingKart, CapitalFloat, Indifi, Faircent, RedCarpetUp, Rubique, Fullerton, Aditya Birla Finance, and many more. Additionally, Companies like Walnut and Moneyview have built interesting expense managers by leveraging the hidden information in your SMS inbox, for instance.
At its core, alternate data helps businesses understand their customers/prospects better, especially in the absence of traditional data. And in many cases, in an instant! And such understanding can be really useful, not just for risk management/ lending start-ups, but also other consumer start-ups.
Let’s take a few examples –
· Food-tech – If you are a Swiggy/ Zomato/ Freshmenu/ Innerchef/ etc… instead of sending mass market “notifications” at around Noon for lunch orders, you could be leveraging the calendar information (presence of a meeting between 12 and 1), competitor information (order history with other food-techs), account balance information (across bank accounts) or spending history (across all purchases) to drive context specific recommendations. A 1:02 PM notification - “In a hurry? Order Combo Meal from Ming’s World. At your desk in 20 minutes.Eat Now. Pay Later.”
· E-commerce – E.g. Koovs/ Voonik/ Jabong/ Myntra- If you knew how the true worth of your customers (total income/ apparel spend/ e-com spend, etc.), your share of her wallet (sometimes, including offline competition) , and her engagement with mobile apps, and that she’s just walked into a mall - can you craft a more powerful recommendation? Maybe incorporating your browsing history or the upcoming key events? “Neighbour’s envy, Owner’s Pride. A dress for that birthday just around the corner”.
· Offline Retail - A Starbucks could leverage the calendar and the balances information along with your location information to deliver a custom proposition aimed at a nearby store, and only for a specific duration – “Here’s a little something for your meetings today. Free Wifi-Great Coffee- Great Conversations. And a free beverage customization.”
· Wallets - You’re about to order food on Swiggy using your PayTM wallet. At this last minute, you realize that you’re low on balance. You either load your wallet, or change the transaction method (to another card/wallet or a cash on delivery). If you’re PayTM/Mobikwik/…, you lose out more than 33% of such occasions. If you’re PayTM, with access to this customer’s alternate data, and a bit of “appetite”, you lose – guess what – <10% of times. You could have, like a short-term credit card, offered to let the transaction go through.
While this is just the tip of how alternate data provides interesting and powerful contextual insights, the real reason why alternate data is important is – an aha moment for a customer is worth more than the investment that goes into creating it. As most practitioners would tell you, a customized/ segmented offer has >6x chances of response compared to a mass market offer. AND, an aha-fied customer returns 3-4x more often.
And won’t your investors love that metric!