Companies in sectors such as financial services and insurance live and die by data. In particular, it's about how well we can use data to understand what people and businesses will do next, a process that will increasingly be dominated by AI. Now a startup called Finbourne, founded in London's financial centre, has built a platform to help financial firms organize and use more data in AI and other models. The company announced £55 million ($70 million) in funding, which it will use to expand its reach outside the Square Mile.
Highland Europe and strategic backer AVP (the venture arm of insurance giant AXA) are co-leading the Series B, which will give the company a post-money valuation of just over 280 million pounds ($356 million). It is assessed by value.
CEO Thomas McHugh, who co-founded Finbourne, told TechCrunch that he came up with the idea for the startup after working as a senior quant in Dorsey for several years. Most of them worked for the Royal Bank of Scotland. One of these was in 2008, when RBS, then the world's largest bank, was overexposed to the subprime lending epidemic and came close to a dramatic collapse.
The major changes came internally in the form of a major reorganization.
Previously, entire banks were made up of a series of business silos, which changed not only how people operated, but also how the data within them operated. All of this is cost-prohibitive to implement and costs that need to be urgently cut. “We had to eliminate hundreds of millions of dollars in costs from the business in a very short period of time,” he recalled.
They decided to take a page from the nascent but fast-growing world of cloud services. Founded in 2006, AWS is now only two years old, but the data team has found that AWS presents a powerful and comparable model for how banks can store and use data. So we have taken an integrated and united approach to this issue as well.
“We’ve built an incredible amount of technology that works across basically every asset class. Until then people were saying this was actually impossible. But we had an incredible reason to change, and we knew it would allow us to build better technology, much more scalable technology,” McHugh said. He said the equity system, bonds and credit, which all previously operated as separate systems, are now on one platform.
The 2008 UK financial crisis was a rollercoaster. If I hadn't completely broken away from it, I would have fallen away from the belief that I can overcome any difficulty. So McHugh ended up taking on the riskiest thing in business: starting a startup.
Finbourne may have its roots in how McHugh and other team members tackled the challenge of building more efficient data services in banking, but he also developed ideas that reflect and shape the way financial services companies buy IT today. Just as companies with broad sales operations may use Salesforce or a competing platform rather than building their own software, Finbourne is confident that financial companies will increasingly do the same. That means, rather than building your own software, you're working with an outside company for the tools to run your operations.
This inevitably dovetails with how banks and other companies in financial services are increasingly leveraging AI.
The company's current products include the LUSID operational data store. A portfolio management platform that tracks investments and ledger (used for asset management analytics) positions, cash, P&L and exposures; and data virtualization tools. McHugh said Finbourne is also helping companies manage how they handle data for training models, which are likely to involve more engagement.
The main takeaway here is that there is no clear leader and banks do not want to share their data with other banks, so they are putting in place training on how to prevent that from happening. This process allows the client to have tighter control over the results and prevent “hallucinations” from seeping into the picture. Open source plays an important role in providing more flexible options to end users.
“What we’ve seen is that customers don’t want the models we write or use to be trained on other people’s data,” he said. “We see that very strongly. “The reason we do that is so the models are less likely to hallucinate because they’re not allowed to use other people’s photos.”
Finbourne currently has a number of competitors. For example, asset management competitors include Blackrock's Aladdin, SimCorp, State Street Alpha, and Goldensource. Alternative asset management competitors include Broadridge, Enfusion, SS&C Eze and Maia. BNY Mellon Eagle, Rimes, Clearwater Analytics, and IHS Markit all offer tools for asset owners. Asset services include FIS, Temenos, Denodo, SS&C Advent and NeoXam.
So much so that someone is getting Fidelity International, London Stock Exchange Group, Baillie Gifford, Northern Trust and Pension Insurance Corporation ( PIC ).
