The challenge
A ride-hailing app signs riders and drivers up from a phone. In its markets, most have no credit history, few use email, and documents are hard to trust. Two decisions ride on that phone number: is this a genuine user, and, for credit features like a ride credit limit or deferred payment, how much can they be trusted with? The question was whether one number could carry both.
The solution
Monnai scored every user for fraud and for credit in one call, from the phone number alone: no legacy credit score, ID document or email. Each model returns approve, review or reject, and together they place each user on a simple grid: clear to onboard, safe identity but limited credit, check identity first, or decline. How complete the number’s messaging profile is was the strongest fraud signal; steady use of the network over the past year drove the credit score.
The impact
92% of riders and drivers cleared the fraud check from a phone number alone, ready for automated onboarding, and up to 96% were cleared for credit with no credit history. Users the fraud model rejected had far fewer social profiles than those it approved. And in one market, some users passed the fraud check but not the credit one: safe identities, better served by a credit limit or an extra check than by a decline.
