Small businesses are turning to loans that require zero human oversight

Shopify is one of several companies using AI systems to automatically offer small business owners loans.
Shopify is one of several companies using AI systems to automatically offer small business owners loans.
Image: REUTERS/Chris Wattie
We may earn a commission from links on this page.

When Xiomara Rosa-Tedla needed a small loan in February 2020 to fund her e-commerce startup Unoeth, she didn’t reach out to a venture capitalist or a bank officer. She asked an algorithm.

Rosa-Tedla founded Unoeth with her dad in 2015. The company sells leather handbags and other accessories handmade in Ethiopia, and Rosa-Tedla was sitting on a backlog of unsold inventory. She needed a few thousand dollars to buy ads on Facebook and Instagram so she could get her products in front of the right customers and sell out her supply.

Rosa-Tedla didn’t want to sell a piece of her family business to a venture capital firm or jump through the hoops required to get a loan from a bank where, she jokes, you have to “give up your house, your car, and your firstborn” as collateral. Instead, she downloaded an app from a lending company called Clearco and gave it access to data about Unoeth’s sales, revenue, and website traffic on the e-commerce platform Shopify.

Within minutes, an AI algorithm analyzed Rosa-Tedla’s business and presented her three funding offers: one for $5,000, one for $10,000, and one for $18,000. She chose the $5,000 offer and had the cash in less than two days, without ever negotiating with a human.

Clearco is part of a growing industry that offers loans to small businesses, particularly e-commerce startups, with virtually no human input. In 2020, Clearco doled out $2.5 billion in funding to 5,500 businesses, relying entirely on algorithms to decide which businesses to give money to, how much to offer, and what the terms of the deal should be.

Payment and e-commerce firms join the debt party

E-commerce platforms like Shopify and payment companies like PayPal and Square are now offering their own versions of the service, too. These companies have mountains of granular data about their clients’ businesses, allowing them to train algorithms to predict which companies are safe bets to lend money to and which are riskier investments. Algorithms can then adjust the terms of the deal accordingly, raising the cost of capital to account for greater risks. (Clearco says humans never review its algorithms’ decisions, but Shopify, PayPal, and Square employ human reviewers to check some deals above a certain size.)

AI-powered lenders argue that they’re filling a funding niche that isn’t served by venture capitalists or traditional banks: They offer quick infusions of capital under a revenue sharing model that doesn’t require founders to give up equity, cultivate personal connections with members of the Silicon Valley elite, or jump through the hoops of banks’ due diligence processes. Lenders like Clearco also claim that their funding method can distribute investments to a more diverse set of founders overlooked by traditional funding methods. Some founders say AI-approved loans give them a faster, easier way to cover routine operating expenses and grow.

AI lending expands its footprint

AI lenders are attracting a growing number of customers. Shopify Capital has lent $2 billion since its launch in 2016—and half of that total has come in the past year. Square Capital says it has lent $9 billion since it launched in 2014. Clearco’s lending totals boomed in the years between 2017 and 2020, and the company now expects to lend more than $1 billion in 2021.

While those numbers pale in comparison to US venture capital ($130 billion invested in 2020) or small business loans ($23 billion issued in 2019), AI lending is growing much faster than either of these two more traditional models.

Zavain Dar, a partner at the venture capital firm Lux Capital, says the rapid growth of AI lending platforms is a promising sign. “If you look at the traction these [AI lending] companies have gotten, it’s showing there was a need for a form of funding where it wasn’t a loan against your home to set up a bodega, and it wasn’t ‘I’m going to go build a $100-billion high-risk tech startup,’” he said. “A lot of businesses needed something in the middle, and the market had overlooked that large middle.”

AI decides differently from humans

Algorithms deciding which businesses get funded are designed to have tunnel vision. While human bank officers or venture capitalists might make lending decisions based on who founded a company, where they went to school, or what kind of products it sells, companies deploying these algorithms say they only consider a narrow set of sales data. Square Capital says its model looks at just a handful of data points including “processing volume, payment frequency, and customer mix.” Clearco’s model looks mainly at revenue figures, but it also takes in data about a company’s margin profile (a measure of how much profit vendors make from each sale), sales growth, and the number of customers who browse the company’s online store each month.

“We wanted to start from the first principles of what we thought made an e-commerce business work,” said Clearco president Michele Romanow.

In the beginning, the Clearco team took a guess at what a successful e-commerce business looked like: They coded their algorithm to only lend money to companies that met a certain threshold for sales, profit margins, and web traffic. Their initial guesses turned out to be pretty lousy. “In our early cohorts, we were losing like 20% of our money,” Romanow said. But over time, Clearco used data from past deals to train a machine learning model to come up with its own rules, and it’s been gradually fine-tuning the algorithm ever since. The AI examines the outcomes of its past lending decisions to learn to spot patterns and more reliably predict which companies will be able to repay their debts.

After the first year, the AI improved enough to make a steady profit on its loans.

Paying off loans by sharing revenue

A key difference between algorithmic lenders and other financial backers is how loans are repaid. Banks and credit card companies typically charge monthly interest payments, while venture capitalists take an ownership stake in a company. But AI lenders focus on revenue sharing: Companies pay back a piece of their debt every time they make a sale.

Terms vary but almost all work the same way. Lenders give a company a lump sum of money up-front—say, $10,000. Then, the company gradually pays that amount back by giving the lender a small cut of every sale they make, which can range from 1% to 20% of each sale, depending on the terms of the deal. If a business makes a $50 sale, it would have to send the lender anywhere from 50 cents to $10.

The startup keeps paying the lender back bit by bit until they’ve returned the original amount, plus a flat fee—which is usually something like 6-12% of the amount the company borrowed, depending on the deal. In this example, the startup would ultimately wind up paying anywhere from $10,600 to $11,200.

The ideal recipient has a low sales volume. The faster a company sells its products, the faster it pays back its debt. A company doing very brisk business might pay off its loan—plus a 6% flat fee—in a month. But paying a 6% fee in one month is the equivalent of an annualized interest rate of 72%. In this scenario, taking out a small business loan (typical interest rate: 3-7%) or even carrying credit card debt (typical interest rate: 15-18%) might be cheaper.

But the AI approach has other advantages. None of the AI lenders report transactions to credit bureaus, meaning that if a startup fails to pay off its full loan amount, it won’t affect the owner’s credit. They also don’t demand any form of collateral, which means if a business goes belly-up, its owners can walk away from their debt without penalty.

Can AI funding reach diverse entrepreneurs?

Self-reported data from Clearco seems to suggest AI lending can help reduce some of the biases that lock women and people of color out of traditional forms of funding. Venture capital and small business lending are marked by clear race and gender disparities: It’s harder for women and people of color to get investments. When US lawmakers approved $659 billion in emergency small business loans to help companies survive the pandemic, 83% of the loans went to white-owned businesses, compared to just 2% for Black-owned businesses. Just 1% of US venture capital goes to Black-owned startups, according to data from the venture-tracking firm Crunchbase.

By contrast, Clearco announced in April that 13% of its funding went to Black or Latino founders, well above those receiving funding from banks or venture capitalists. Clearco also claimed that it funded “eight times as many companies headed by female founders as traditional VC firms,” and that the majority of its funding went outside the traditional tech hubs in California, New York, Texas, and Massachusetts that typically absorb the lion’s share of venture funding.

Those outcomes earned cautious plaudits from Jeanna Matthews, a computer science professor at Clarkson University who studies the ethics of AI systems. “If you’re watching the impact of a deployed system and seeing that it’s helping them avoid bias, that’s a good sign,” she said. But Matthews warns that the data doesn’t guarantee AI lenders’ algorithms are free from bias. Even if they’ve excluded data about founders’ identities or the nature of their products, innocuous data points like sales and revenue figures could become stand-ins for founders’ identities. “Oftentimes the bias is in the data even with those columns removed,” Mathews said, “and if you’re not careful, you can end up rediscovering those same data points through proxy variables.”

Ultimately, Matthews says, the best thing about the AI lenders’ algorithms may simply be that they’re able to process more funding applications from more business owners more quickly than any human could. As a result, they’re able to accept funding applications from virtually anyone, anywhere.

“Maybe what we’re saying is it’s better to say yes to a lot of people,” she said. “Maybe, there are great ideas from women, people of color, people from lots of states, that no one was picking up before, so when you say yes to people in those areas you’re getting value others didn’t.”