This Bengaluru-based AI-enabled startup helps identify and remove fake users

Digital transactions may have made our lives easier, but how many of us are tech-savvy enough to detect identity frauds ? 

Hackers and cyber thieves have found several ways to fool internet users. Some fraudulent practices include identity theft, phishing, cloning, smishing and vishing. With mobile phone numbers commonly used to sign up to several websites, cyber fraudsters find it even simpler to wreck damage. 

Synthetic identity fraud is the fastest growing cybercrime in the world. It results when criminals fabricate fraudulent identities to establish new accounts or lines of credit. Synthetics are difficult to detect, costly and often misclassified. 

Bengaluru-based TrustCheckr, launched by colleagues Adhip Ramesh and Shivraj Harsha in 2017, aims to solve this problem. An AI-enabled tool that helps identify and remove fake users, the startup has a proprietary algorithm tested on 1 million+ data points and used in AI and ML in image comparisons.

“During the onboarding stage in any B2C/fintech companies, lots of fake/fraud profiles are created with synthetic phone numbers and email addresses with an intention to scam/fraud the systems. This is across industries including fintech, banks, crypto, dating, marketplaces, classifieds etc,” says Adhip Ramesh, Co-founder, TrustCheckr in a chat with YourStory.

ALSO READ

[Funding roundup] Pi Beam Electric, Seekho.ai, UPCRED, TrustCheckr raise early-stage capital

“This problem was leading to - user follow-up losses  via call centers, SMS, emails etc; unnecessary underwriting expenses; inefficient way of utilisation of marketing spends; and challenges in building a trusted community,” he adds. To beat these issues, Adhip’s startup assigns a trust score to the user based on phone number, Facebook id or email id. The company has already completed more than 10 million “TrustChecks” till date across seven clients.

Although TrustCheckr’s solution can serve businesses across industries, including fintech, dating and ecommerce, it provides a cross-industry fraud insights platform for BFSI (banking, financial services and insurance sector) in particular.

“We also do AI-based social scanning across LinkedIn, Facebook, Twitter, Instagram, websites, deep web, news articles, alternate data sources etc., to understand the sentiments behind posts in real-time, extract key fraudsters information like name, email, phone, IP address and update real-time fraud data sets and risk scoring algorithm,” Adhip adds.

 Since TrustCheckr does social scanning for more than 1000 keywords and handles, it processes 50 million tweets to fetch the data and extract the fraud sentiments on social media.

The startup claims to provide simple REST APIs as Phone Risk API, Email Risk API, UPI Risk API, IP Address Risk API and Social Media Negative List. Representational state transfer (REST) is a software architectural style that was created to guide the design and development of the architecture for the World Wide Web. REST defines a set of constraints for how the architecture of an Internet-scale distributed hypermedia system, such as the Web, should behave.

Along with this, the startup is also due to launch the “TrustCheckr Alerts” platform to help fintechs, banks, credit bureaus, NBFCs get early warning signals about fraudsters online.

The USP

According to Adhip, “We are pioneers in social scanning technology. We are different from our competitors based on - cross-industry fraud social scanning, custom built fraud analyser, custom built fraud data extraction modules, closed user group support and technology built for scale.” 

“Since the nature of the data we work with is complementary, we suggest all our clients to use our product first (pre-KYC checks) and then optimise the entire funnel of the customer onboarding,” he says.

 The team

Adhip and Shivraj were colleagues at strategy consulting firm called RedSeer Consulting.

An alumnus of IIM Lucknow, Adhip has previously worked with companies like Simplilearn, KPMG, ATK, and NetApp. Shivraj is an alumnus of University of Liverpool, UK, and has previously worked at I-Tiffin besides RedSeer.

TrustChkr currently has a team of 10 full time employees and a couple of interns.

Funding and monetisation

The startup was initially bootstrapped with Rs 10 lakh, before raising Rs 45 lakh in an angel round from Rangsons LLP (Investment arm of NR Group) in 2018.

In March 2022, the firm secured Rs 6 crore from early-stage fintech fund IIFL Fintech Fund, with participation from Rangsons LLP as well. 

“Our monetisation model is pay per API call. We offer two payment models. The first is the prepaid model, which is more suited for small companies where they buy 25,000 API calls upfront and can consume it depending upon their growth. The prepaid package starts with a minimum billing of Rs 1 lakh. In the second model, the startup enters an annual contract of around $25,000 with a well-established NBFC,” explains Adhip.

The startup has delivered more than 10 million API calls till date.

“Our MRR is around Rs 13.5 lakh to Rs 15 lakh. We expect to close FY 21-22 with a revenue of Rs 1.5 crore. We are able to breakeven with the current revenue flow. In the current fiscal, we expect to double the revenue,” states Adhip.

The challenges

The startup’s journey has not been without challenges. For validation and underwriting of customers, the startup had only existing normative Aadhar combined bureau centric framework to resort to, which had its own set of cons. 

“Though this frame ensured frictionless validation of the KYC process, it had its limitation, as it does not consider the digital id of a user. Which essentially meant that companies were not sure if both the digital id and physical id (Aadhar details) belonged to the same person. This left major chinks in its fraud prevention armour,” says Adhip. 

“Our product was essentially a paradigm shift, hence, we had to educate our clients in the beginning. However, with the onset of the COVID-19 pandemic, there was a sudden spike in the number of fraud cases. Companies are now forthcoming about investing in products that augment their existing systems,” he comments. 

The way ahead

As per industry analyst firm Aite Group, the cost of synthetic identity fraud is expected to reach $2.4 billion by 2023. 

Although TrustCheckr’s product is sector and geography agnostic, Adhip says, “our present geographic focus is in India. We started with the fintech sector, now we are looking to expand into the rental economy. Most of our established NBFC clients are either located in Mumbai or Delhi. Our new age fintech clients are in Bengaluru,” he adds.

The startup competes with the likes of Socure, Seon.io, Fraud.net and Threatmetrix among others.

With an active pipeline of 50+ clients including IIFL, Piramal, Snapmint, the startup will continue its thrust on the fintech sector.

“What we have explored now is just the tip of the iceberg. With the anticipated explosion of credit growth in coming years, a product like ours will ensure that companies can focus on their core activity of lending, leaving us the task of safeguarding them from fraudsters,” says Adhip.

“In terms of expansion, we will continue to invest heavily in tech talent. In the immediate weeks, we are looking to hire 15 tech resources with a mix of full stack developers and data scientists. Our recent fundraise will give a runway for 15-18 months. We want to do 100 million TrustChecks per year in the next two years,” Adhip signs off.