Today, the artificial intelligence (AI) scene is glowing red hot as millions of users delight in the performance of advanced chatbots such as OpenAI’s ChatGPT. And you don’t have to look far to find headlines predicting how AI algorithms will shake up all kinds of sectors, from fashion modeling to healthcare. But amidst all this excitement about the future, it’s important not to overlook examples of how the combination of big data and AI already benefits firms today. And one of the strongest use cases, with a years-long track record in helping businesses, is the application of AI in commercial finance.

In 2019, the Bank of England surveyed hundreds of financial services providers, including banks, credit brokers, e-money institutions, financial market infrastructure firms, investment managers, insurers, non-bank lenders, and principal trading firms. The UK’s central bank was keen to know how machine learning (ML) was being deployed, the business areas that were benefiting, and how mature applications were. Two-thirds of respondents reported that they were already actively using ML models. And the number of deployments was expected to more than double by the end of 2022.

The use of ML and AI in commercial finance helps firms in a number of ways, and to understand the benefits it’s useful to step through the business lending process. Companies need to raise funds for a variety of reasons – to use the money as working capital for day-to-day operations, to increase inventory, hire more people, or even purchase a new business. And, historically, firms would approach different lenders or apply via a broker. But a common bottleneck was the length of time taken by lenders to determine whether or not a company’s application for a business loan should be approved.

Having waited weeks for a decision, firms faced a double setback if their applications were rejected. Not only would their original funding issue still be unresolved, but they now had even less time to arrange financing, which is a stressful scenario for business owners. Also, for asset-based lending (ABL) – where organizations borrow against plant and machinery, or property that they own – the collateral needs to be valued by the lender, which adds another step to the business loan process. But, as the results of the Bank of England survey hint at, the finance sector has been busy innovating, and the fruits of that work can now be seen in the evolution of unsecured financial products.

Alternative lending options

Today, thanks to the use of AI in commercial finance, companies have alternative lending options, which allow them to receive an offer in minutes and even get access to their business loan on the same working day. “It all boils down to data,” Chirag Shah – founder and CEO of Nucleus Commercial Finance – told TechHQ. “The ML and AI is already there.”

Nucleus Commercial Finance began as an ABL lender and has expanded into new products as the firm demonstrated how effective algorithms could be in predicting the capacity of companies to pay back loans. Back in 2016, its ABL products had just £5.8k of bad debts on more than £400m worth of loans, which shows just how accurate models can be if you give them the right signals. And its unsecured business loans, which, although they carry more risk, still have 30% lower defaults than the market average, based on benchmarking carried out by the firm.

Shah explains that the apps, which fire up in the background once applicants complete and submit a simple online form, consider somewhere in the range of 40 to 50 key performance indicators (KPIs) accessed via APIs. And for around 90% of cases, the system can make a loan decision in a matter of minutes by comparing signals to historical data as well as new information that the company continues to add to its technology platform.

What’s more, when a company’s application for a business loan is turned down, it doesn’t mean that the firm is permanently rejected. “We approve or decline the business at a moment in time,” said Shah. “And our tools are able to contribute back to the applicants.” Companies that have given Nucleus access to their data can receive monthly snapshots that show owners – based on its KPI analysis – how their business is going.

Supporting growing companies

The use of AI in commercial finance can identify red flags that might not yet be apparent to management teams and warn of potential cash flow issues further down the road. And there are other benefits too, particularly for younger companies, which can find it harder to access business loans. Rather than trawling through years and years of trading histories, which early-stage firms won’t have, alternative funding providers instead seek other data that can be used as a substitute for predicting long-term performance.

Models are continuously evolving to support growing companies, increasing the number of firms that qualify and helping to make business loans more accessible. And it’s tempting to think the more data, the better that AI in commercial finance will become. But it’s important to double-check the contribution that new information brings. “You need to establish that the extra data adds value and is better than what you already have,” explains Shah. “At the end of the day, it’s about better decision-making.”

LeackStat 2023