MNO transactional data, USSD session data
Transaction level segmentation, machine learning, uplift analysis
Inclusivity Solutions (Inclusivity) works with mobile network operators (MNOs), insurance companies and other distribution partners to deliver digital insurance solutions that meet the needs of emerging consumers. Their primary offering is free hospital cover awarded to mobile customers whose mobile phone transactions exceed a stipulated threshold. These include spend on airtime purchases as well as mobile banking transactions.
Inclusivity was awarded a technical assistance grant by Insight2Impact (i2i) to build their internal analytical capacity. The key objective entailed analysing their customer data that was, at the time, underutilized. Insight2Impact (i2i) appointed 71point4 to prepare a detailed analysis of Inclusivity’s customer data, while enabling the analytics team to independently analyse their data on an on-going basis.
The analysis used telephony and mobile banking data to segment Inclusivity’s customers into eight segments. These segments were based off each customer’s observed transactions behaviour. The segments ranged from customers who rarely transact to those that transact frequently and with high values.
The team also conducted an uplift analysis that assessed whether Inclusivity’s product offering drives revenue for MNOs. The methodology employed in this analysis is also used by companies like Google and Facebook, who regularly want to measure interventions such as a new product launch or the onset of an advertising campaign.
Throughout these analyses, 71point4 worked closely with the Inclusivity team (on-site and remote) to develop a separate analytics architecture. This new cloud based architecture, that uses Redshift as its backend, integrates with Inclusivity’s production databases to deliver on-demand analytics. The system also leverages technology such as docker to deliver machine learning predictions as a microservice.
71point4 used Inclusivity’s customer data to develop a spend-based segmentation model, allowing for targeted interventions based on an improved understanding of customer behaviour. The customer data was also used to calculate the revenue benefits that Inclusivity’s free loyalty-based product offering generates for MNOs. This analysis confirmed that Inclusivity’s customers generate additional revenue for MNOs after registering and, in combination with the segmentation exercise, identified the customer group driving this revenue growth.
Both these analyses were coded and automated so that they can be conducted on an ongoing basis and at regular intervals allowing Inclusivity to track results over time.
As yet, nothing in the public domain and unlikely to be in future because the underlying data is proprietary.