AI fintech start-up looking to tackle small business late payments gets seed funding

Previse has bagged £2m in a funding round led by Hambro Perks, which also featured Founders Factory and a range of high profile angel investors

A fintech start-up that’s looking to tackle the late payments epidemic, and ensure small businesses get paid on time, has raised seed funding.

London-based Previse has bagged £2m in the round, which was led by Hambro Perks, but also featured Founders Factory and a range of high profile angel investors.

Founded by Paul Christensen, Philipp Schoenbucher, David Brown, Giulio Rossi and Andre de Cavaignac, Christensen was previously global co-head of Goldman Sachs’ principal strategic investments team.

Once a small business issues an invoice, Previse uses advanced AI to scan hundreds of millions of data points and calculate the likelihood that buyer will actually pay the outstanding bill.

Previse then provides this score to banks and asset managers, who pay the small firm instantly on the buyer’s behalf – with the suppliers offering a small discount on their invoice in exchange.

It’s hoped this model will then result in reduced transaction costs for buyers, improved working capital for suppliers and small businesses, and a safe asset class for funders.

Previse will use the funds to further develop their proprietary AI technology.


Action point:
See if you can get a Start Up Loan to help you start a business idea
(external site, opens in new tab)

Christensen said:

“Small businesses are the backbone of the world economy, generating the majority of growth, employment and innovation. Yet, most of them are consistently paid late by corporate buyers. It is an unsustainable position which damages the entire economy.”

Schoenbucher said:

“Payment decisions are a perfect candidate to apply machine learning. Our advanced proprietary machine learning algorithms were developed using data sets of multiple billion dollars of corporate spending, building upon state of the art binary classifiers and highly innovative domain-specific feature engineering methods.”

Comments

(will not be published)