Case Study
AI|ML Product Engineering
Predictive Lead Scoring by ML Automation
NutaNXT team built a ML model to assign a lead score to all potential sales leads. A higher score would mean that the lead is hot, i.e. is most likely to convert into a new customer, whereas a lower score would mean that the lead is cold and has a low probability of conversion.
Client Introduction
Client is a leading BFSI FinTech enterprise which offers unsecured loans to MSME, SME enterprises, retailers and merchants across verticals.
Technologies Used By NutaNXT
- AI & ML Solutions
- Predictive Analytics
- AWS
- Scorecard
Use Case Applicable To Verticals
- E-commerce
- Retail
- Loans
- Insurance
- Financial Products
Customer Conversion Rate
40%
<=
9000
Customers
Per Quarter
Client
Challenge
The primary client challenge was to improve the Sales Conversion Rate for new customers coming through their digital channels (Web Forms, Social Platforms etc.). The current process was unable to identify and categorize leads based on probability – Hot, Warm, Cold, etc.
Key challenges with the existing model were:
- Poor lead prioritization process resulted in missed sales goals for our client.
- The client was evaluating data to validate buying interest from the prospects and assign a score to indicate the purchase ability of the prospect.
Team that we built
- 2
- Data Scientists
- 1
- Cloud Architect
- 1
- Data Engineer
- 1
- DevOps Engineer
Our Solution
NutaNXT built a ML model with accuracy greater than 70% as well as mined and engineered data to fulfill 85% of null data in their prospects database, and this process enabled the sales team to concentrate on the convertible leads.
- Our ML model could predict prospects with higher propensity of conversion, by mining specific information about customer attributes.
- Each lead was evaluated, scored, ranked and tagged with key factors and shared as a partner of a well qualified lead list with sales personnel.
Technology Stack Used
Customer Conversion Rate
Increased to
200%
>=
No. of Customers
Increased Per Quarter to
22000
Business Impact and Results
The predictive lead scoring model allowed the client to allocate their operational bandwidth according to the lead score, that led to sales target achievement, less turnover and overall lower sales costs.
- 200% increase in customer conversion rate
- 22,000 prospects converted per quarter
- $700 million loans disbursed so far
More Works
From MVP to end-to-end digital products, our culture of intelligent innovation and engineering excellence have solved complex Digital Product Engineering challenges and delivered maximum business ROI for our clients.