Interview With Bruno Farinelli - Senior Director of Risk and Customer Success at ClearSale

Published on: May 27, 2024
Shauli Zacks Shauli Zacks
Published on: May 27, 2024

SafetyDetectives recently had the opportunity to interview Bruno Farinelli, the Senior Director of Risk and Customer Success at ClearSale. With a career spanning over a decade at ClearSale, Bruno has played a pivotal role in shaping the company’s analytics and fraud prevention strategies. His journey with ClearSale began in 2012, and since then, he has spearheaded various initiatives, leading to his current leadership position. In this interview, Bruno shares insights into ClearSale’s unique approach to e-commerce fraud prevention, the transformative impact of AI in risk management, and emerging technologies that are set to revolutionize the industry.

Can you introduce yourself and talk about your role at ClearSale?

I was invited to join the Analytics and Data Science team at ClearSale back in 2012. My time with the company has been very rewarding and one of the greatest adventures of my life! In 2015, I headed the Analytics team of the International Business Unit, where the rapid success of this initiative and the excellent performance achieved ended up merging additional responsibilities with my role. This led me to the Director of Operations and Analytics role at ClearSale where I guided an excellent team of data scientists and fraud experts responsible for the main KPIs of the company. Now, I have been promoted to Senior Director of Risk and Customer Success, responsible for Risk and Account Management of the entire International Portfolio of ClearSale.

How does ClearSale differentiate itself from other e-commerce fraud prevention companies?

What truly sets ClearSale apart is their comprehensive hybrid approach that blends advanced technology with human intelligence. Their solutions utilize ML models, AI and automated rules to analyze a vast array of data points from orders, customer profiles, device fingerprinting and more. However, ClearSale doesn’t solely rely on automated systems – orders identified as high-risk go through a proprietary advanced review process by expertly trained fraud analysts. While most ecommerce fraud prevention companies decline the majority of flagged transactions, our trained in-house analysts take a proactive approach by handling all flagged orders to decrease chargebacks and improve authentic customer relationships. With customized solutions tailored to each merchant’s risk profile, ClearSale provides a holistic, end-to-end service covering everything from fraud prevention to customer loyalty preservation.

How has AI transformed risk management in e-commerce, and what are the biggest challenges you face when implementing AI in fraud detection systems?

AI has transformed seemingly every industry, and ecommerce fraud detection is no exception. Although ClearSale has been using AI models for years, large scale implementation of these systems comes with significant challenges. Data quality is paramount, as poor training data can lead to biased and inaccurate decisions. While AI excels at handling routine cases, there will always be edge cases requiring human expertise and contextual understanding. Additionally, fraud is an ever-evolving landscape, and fraudsters can adapt their tactics faster than AI models can learn, potentially leading to losses if the models fail to keep up. Ironically, fraudsters are also starting to leverage AI to create fake identities, forge documents, and launch sophisticated phishing scams, further complicating the battle against fraud. To navigate these challenges, ecommerce companies must ensure high-quality training data, maintain human oversight for complex cases, continuously update their AI models, and proactively monitor for potential AI misuse by fraudsters.

How do you balance the need for data collection with privacy concerns in your risk management strategies?

Data is the lifeblood of any AI system, but we can’t be cavalier about how we collect and use it, especially when it comes to people’s personal information. On the one hand, the more data we can feed our fraud detection models, the smarter and more accurate they become at sniffing out fraud. But on the flip side, we have to be respectful of privacy and not go overboard playing data hoarders.

It’s a delicate balancing act. We try to be surgical about only collecting the data points that are absolutely essential for assessing risk. Anything extra is just asking for a privacy headache down the road. And wherever possible, we anonymize that data to protect identities. Access is also on a strict need-to-know basis with heavy encryption.

That said, we also try to be upfront with customers about how their data may be used, and we make sure to get proper consent. The latest privacy-preserving tech like differential privacy and federated learning is helping too, allowing us to train models on distributed data without ever centralizing the raw stuff.

What emerging technologies do you believe will have the biggest impact on e-commerce fraud prevention in the next few years?

Artificial Intelligence (AI) and Machine Learning (ML) as these technologies are revolutionizing fraud prevention by analyzing transactions in real-time. AI and ML algorithms can detect patterns, anomalies, and suspicious behavior, helping merchants identify fraudulent transactions more effectively,

Biometric Verification, with methods like fingerprint and facial recognition enhancing security. Biometric data is unique to each individual, making it difficult for fraudsters to impersonate legitimate users. Integrating biometric verification into e-commerce platforms can significantly reduce fraud risk.

Blockchain technology ensures transaction transparency and immutability. By recording transactions in a decentralized and tamper-proof ledger, it becomes harder for fraudsters to manipulate or alter data. Implementing blockchain in payment processing can enhance security and build trust among users.

As none of those should be viewed as a silver bullet, it’s also important to highlight that collaborative efforts between merchants, payment processors, and financial institutions aim to strike a balance by minimizing fraud risk while ensuring a seamless shopping experience for customers.

What advice would you give to e-commerce businesses looking to enhance their fraud prevention measures?

Ecommerce businesses looking to make strides this year would be wise to make AI the strategic centerpiece of their practices. While traditional rules-based systems still have their place, ML models deployed for real-time fraud detection and continuously updated to adapt to evolving tactics offer a potent new line of defense. However, AI shouldn’t be treated as an omniscient black box – human experts must still be looped in to scrutinize complex edge cases that require contextual nuance. Beyond current AI applications, ecommerce fraud teams should also explore emerging technologies like deepfake detection to thwart synthetic media and decentralized identity solutions built on blockchain to remediate identity theft vulnerabilities.

About the Author
Shauli Zacks
Published on: May 27, 2024

About the Author

Shauli Zacks is a tech enthusiast who has reviewed and compared hundreds of programs in multiple niches, including cybersecurity, office and productivity tools, and parental control apps. He enjoys researching and understanding what features are important to the people using these tools.