Ensuring data-driven progress in invoice finance

Basel 3.1 and the data-driven business case 

Let's start with a challenge familiar to many European factoring companies: Basel 3.1 (widely known as CRR3 across Europe, and often also referred to as Basel 4).  

The regulation imposes a 100% risk weighting on factoring which ties up significant capital and limits lending capacity. The EUF is actively working to challenge it, arguing that factoring is a safe and effective lending practice with minimal losses that deserves a more favourable risk weighting.  

But good intentions aren't enough. To convince regulators, the EUF needs data. They need to demonstrate low loss rates, effective risk management, and consistent performance. The problem is, however, that the data needed to prove it just isn’t there.  

Without this data-driven business case, the industry remains at a disadvantage. 

While the UK market offers a more flexible approach, the need for consistent and reliable reporting at a market level remains a challenge without data. Centralised registers and mandatory data submission seem like a straightforward solution, but they too face significant hurdles. The creation of a central register will remain a forlorn hope while there continues to be considerable reluctance to share data.  

This is where technology providers can step in, offering innovative solutions to collect, analyse, and share data in a secure and efficient manner. 

Underreporting on fraud 

Another critical aspect of the data landscape is the issue of underreporting on fraud. While fraudulent transactions represent a small proportion of overall activity, they can have a significant impact on individual companies and the whole industry.  

As Kevin Day of Lendscape pointed out in last year’s BCR yearbook, "businesses have historically omitted to flag fraud accurately in their datasets". This reluctance to acknowledge and report fraud stems from a variety of factors, including reputational concerns and a desire to avoid revealing vulnerabilities in internal systems. 

However, by failing to openly share information about fraudulent activities, we limit our ability to learn and develop more effective prevention strategies. As an industry, we need to foster a culture of transparency and collaboration, where companies feel safe sharing their experiences and insights. 

Technology plays a vital role in detecting and preventing fraud, but it's not a silver bullet. Advancing technologies in AI, and in particular machine learning, are only as good as the data they’re trained on, or their rules are based upon. If that data is incomplete or inaccurate, the results will also be unreliable.  

Hype vs. reality for AI 

Yes, I did mention AI – and as with most industries, AI-powered solutions are promising to revolutionise the challenges faced in our industry – including our data issue. But in the invoice finance industry, it's important to approach these claims with a healthy dose of scepticism.  

The reality is that many "AI" solutions are, at their core, advanced statistical analysis and algorithms. That’s not to say these tools aren’t valuable – in fact, what Lenvi Riskfactor is already delivering for our users daily is what many AI developments are promising to deliver in the future. It’s more to say that the promises of these AI systems are not true AI, such as systems that can learn and adapt without human intervention. The distinction is crucial, as it sets expectations at a realistic level.  

It's not about dismissing the potential of AI altogether. Far from it, it's about recognising that AI is a tool that can enhance human capabilities, not replace them. It’s also important for us not to demand development for development’s sake and focuses on the real and practical added value AI capabilities can bring to the solutions already out there.  

Surely, the real opportunity lies in finding ways to combine the power of AI with the expertise and judgment of experienced professionals. 

The value of human expertise 

Ultimately, the invoice finance industry is a business built on relationships and trust.  

It's about understanding the unique needs of each client, assessing risk with a nuanced perspective, and making informed decisions based on a variety of factors. These are things that technology can assist with, but it cannot replicate. 

AI, in this field, serves as a tool to empower and augment human capabilities, not as a replacement for them because the profiles of a fraudulent case can often be so similar to what is a very good, strong performing client that it relies on human judgement and the ability to connect seemingly unconnected dots to spot the bad seeds.  

Instead, by embracing a collaborative approach that combines the best of both worlds, we can create a future where the invoice finance industry is more efficient, and more resilient to fraudulent activity and market shifts.  

But success of the human/ tech collaboration also demands a transparent approach - for example, as raised earlier in this article, discussing cases of fraud. By educating and informing one another on the nuanced threats and risks, businesses can prepare themselves and ultimately provide a better service to their clients. 

By fostering partnerships across the industry, we can unlock valuable insights from existing data and translate them into actionable strategies that drive growth and innovation for businesses and the wider industry.