IndiaFinance and insurance2023 – ongoingCheQ
Credit product discovery from customer emails and SMS
CheQ's credit product discovery module rose from 28.6% to 80.2% efficacy at one-third of the earlier cost
Problem
CheQ, an Indian credit management app, had to find each user's cards and loans in bureau records, emails and bank messages. The third-party provider it used at launch was inflexible, and its discovery module worked well in only 28.6% of cases. The startup had no data science team and only a small operations team.
Approach
CheQ moved its platform to Google Cloud and used Document AI to identify and extract credit product information from customer email and SMS accounts. The team improved the module through continuous feedback loops. The company said a few developers and one customer support team member now run the process.
Results
Takeaway
A small team without data scientists can run document extraction on a managed service. The company did not say how it defines efficacy.
Categories
- Size
- Small business
- Industry
- Finance and insurance
- Business function
- Production and quality
- Innovation type
- Product and service innovation
- AI technology
- Computer vision
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Google Cloud's customer story describes how the Indian fintech CheQ used Document AI to extract credit product data from emails and SMS. It reports efficacy rising from 28.6% to 80.2% and a module that cost one-third of the earlier third-party solutions. The page carries no date; the Internet Archive first saved it on December 22, 2023.
Full details are in the original sources.