You're familiar with the number of documents required to approve a loan application if you have ever taken one out.
It's a lot.
This process can be tedious and time-consuming. In many cases, it is manual.
Ocrolus, a startup, aims to change this with its automation platform. It claims it can analyze financial documents with over 99 percent accuracy. The company, which is based in New York, today announced it had raised $80 million in Series B funding to help lenders automate their underwriting processes.
Fin VC was the financier. This financing values Ocrolus, a New York-based company at over $500 million. It brings Ocrolus' total funding raised since 2014 to more than $100 million. The latest round also included participation from Thomvest Ventures and Mubadala Ventures as well as Oak HC/FT FinTech Collective, QED Investors.
The company is transparent and open about its financials. According to Sam Bobley, co-founder and CEO of Ocrolus, Ocrolus' revenue has increased from $1 million to $20 millions in annual recurring revenues (ARR) between the first quarter of 2018 and the second quarter 2021. It has been able to attract a variety of high-profile fintech clients, including Plaid, PayPal, Plaid, Enova and LendingClub.
Bobley stated that the company spent approximately $10 million on total marketing and sales expenses.
It is now ready to pursue traditional financial institutions as well. The startup will use its capital to build more products for the banking and mortgage lending industries, and expand its U.S operations.
Ocrolus employs a mix of technology including OCR (optical characters recognition), machine learning/AI, and big data to analyze financial documents. Bobley says that what makes it stand out is the Human-in the-Loop (HITL), component, which is also used in document processing for further accuracy.
Bobley explains that the company's goal is to assist lenders in making faster and more data-driven decisions. It is clear that the COVID-19 epidemic has led to digital acceleration in many industries. Financial services are the most affected. Bobley notes that less than 1% worldwide loans were made online before the pandemic. However, the demand for digital lending technology has increased dramatically since the COVID-19 pandemic.
He said that COVID-19 has forced financial institutions into evolution, so every bank and lender has to offer customers online options.
Logan Allin, Fin VC's managing general partner and founder, agrees. He points out that banks and fintechs still have to deal with mountains of paper and digital documents when extracting financial data for loan applications.
Ocrolus is a pillar of fintech and has been solving these problems using OCR and AI/ML. He wrote via email. Ocrolus' platform is still in its early stages, and we are excited by the unconstrained TAM.
Ocrolus isn't the only player in this space. But what is helping to drive its growth, Allin believes is its compliance overlay and fraud detection capabilities as well as its customized analytics and benchmarking capabilities.
Ocrolus was founded with the intention to automate long-term care Medicaid applications. It discovered that professionals who were well-paid spent their time looking through pages after page of documents, line by line.
We began to investigate the issue and discovered that the technology available was simply not accurate enough to make it useful. Bobley shared this information with TechCrunch. Bobley stated that OCR products are offered by tech giants like Google, Amazon, and Microsoft. However, many people have difficulty reading text from PDFs or images, especially if the documents are not structured. Machines have a hard time understanding all of the formats.
We wanted to find a better way to do this. He said that we created a machine-learning-based platform that also includes humans. No matter how the submitted document looks, whether it's a Chase document or a blurry image from a Kansas community bank, we will be able process it with pinpoint accuracy.
The solution breaks down any tasks that it can't perform automatically and routes them to its own team, which includes analysts and quality control specialists who will verify the data. To ensure that its workers are doing the job correctly, the company uses a series algorithm checks.
Bobley stated that we provide perfectly structured digital data for each file we process.
The company was founded in 2012, and the team soon realized that lending was a more lucrative market than Medicaid.
Bobley shared his belief that we entered the fintech lending market at the right moment and the right place.
The company launched its official product in 2016 and was now generating revenue.
Bobley stated that one of the many benefits of our software was its ability to scale fintech lenders. Our software is 10x more efficient than manually processing the loans, Bobley said. We processed loans in minutes or seconds, whereas other lenders took hours or days. People started coming to us, even big fintech lenders, without our having to have a sales team.
Bobley believes that the startup does more than speed up the loan approval process. It's also about financial inclusion.
Bobley stated that our platform allows lenders to automate underwriting and smartly leverage income and cash flow data for credit scoring. Ocrolus makes it easier for lenders to quickly analyze different sources of financial data and provides increased access to credit at lower costs.
The company will also make use of its new capital to hire more people, with a particular focus on the machine learning and data science departments. The company plans to also open a data quality control facility in Florida for financial institutions and government agencies with onshore data needs.
