Upstart is an online lending marketplace that provides personal loans by applying machine learning to a vast array of non-conventional variables.

It was founded in 2012 by Dave Girouard, former president of Google Enterprise, Paul Gu, a Thiel Fellow and Anna Counselman, a former Google manager.

THE ISSUE

Upstart quickly found out that there was a mismatch between borrower’s creditworthiness and their access to prime credit.

  • 80% of Americans have never defaulted on any credit product
  • Only 48% of Americans have access to prime credit

This leads conventional players to refuse credit to “good borrowers”, negatively impacting the borrower but also the lender’s bottom line.

HOW IT WORKS

The company therefore developed an income and default prediction model that is capable of determining the creditworthiness of prospective borrowers.

  • As conventional banks, it uses traditional variables (FICO score, income, credit report) to predict creditworthiness
  • It also uses academic variables (GPA, area of study, colleges) and work history to predict a borrower’s propensity to repay
  • All in all, it uses over 1,600 data points to score borrowers

DELIVERING RESULTS

These consumers benefit from lower interest rates, higher approval rates and a highly automated and digital experience:

  • Upstart provides 27% more approvals than traditional models
  • Borrowers benefit from 16% lower rates than traditional models
  • At the same approval rate, Upstart gets 75% less defaults than large US Banks
  • At the same loss rate, Upstart gets 173% more approvals

Upstart doesn’t make loans itself, rather it connects consumers to its network of Upstart AI-enabled bank partners.

ARTIFICIAL INTELLIGENCE FLYWHEEL

Upstart is able to improve its models, cut down its loss ratio and select good borrower more efficiently as the volume its processes increases over time.

  1. It uses more data and better models to undercut competition
  2. Customers get better interest rates, attracting even more customers
  3. With more data, Upstart is able to continuously improve its models

THE PRODUCT

Upstart’s loans go from $ 1,000 to $ 50,000 and concern 3 to 5 years loan terms with an APR range of 6.5% to 35.99%.

  • 99% of these loans are funded in 1 business day after signing
  • Approximately 70% of Upstart-powered loans are instantly approved with no document upload of phone call required
  • Borrowers can pay their loan back early at no cost

Upstart is focussed on personal loans, it thus offers helps customer with:

  • Moving loans
  • Home improvement loans
  • Medical loans
  • Credit card consolidation
  • Debt consolidation
  • Wedding loans

EASE OF USE

Upstart’s tools are delivered in the form of a simple cloud application. This shields borrowers from the underlying complexity of Upstart’s tools.

Upstart’s bank partners can configure many aspects of their lending programs, including factors such as:

  • Loan duration
  • Loan amount
  • Minimum credit score
  • Maximum debt-to-income ratio
  • Return target by risk grade

ATTRACTING BORROWERS

Upstart is referring borrowers to partner banks when these visit Upstart’s website. Borrowers can also interact with Upstart’s tools as a white-label product through their bank’s interface.

1. Directly through the Upstart website

  • As always, a partner bank will emit the loan as Upstart itself doesn’t provide any loan
  • The loan can then be retained by the originating bank, sold to Upstart’s network of institutional investors or funded by Upstart’s own balance sheet
  • In Q3 ’20, around 22% of the loans issued directly through Upstart were retained by the originating bank
  • Around 76% were purchased by institutional investors (such as PIMCO, Goldman Sachs, Morgan Stanley)
  • Around 2% were funded by Upstart’s own balance sheet

    “but that’s for research and development, so we can test new things.” Dave Girouard by Penny Crosman for American Banker

2. Through a white-label product on the bank partner’s own website

  • The bank uses Upstart’s models and algorithms to predict a prospective borrower’s creditworthiness
  • It then makes the loan on its own through its own interface

MARKETING CONCENTRATION

Most of Upstart’s growth and marketing initiatives are focused on bringing potential borrowers to Upstart. It relies on direct mail, podcast advertising, affiliate marketing and online advertising to boost growth. Noteworthy is their dependence on Credit Karma:

  • In 2020, 52% of consumers that applied for a loan on Upstart.com learned about and access Upstart.com through the website of a loan aggregator, Credit Karma:

    “The percentage of loan originations that were derived from traffic from Credit Karma was 28%, 38%, 38% and 52% in 2017, 2018, 2019 and the nine months ended September 30, 2020, respectively” Upstart S1 Filing
  • Only 12% of loan originations came from direct mail, down from 36% in 2017

    “[…] the percentage of loan originations that were derived from direct mail was 36%, 28%, 23% and 12%, in 2017, 2018, 2019 and the nine months ended September 30, 2020” Upstart S1 Filing

BUSINESS MODEL

Upstart counts 10 bank partners and generates around 98% of its sales with fees paid by these banks. These are charged for:

  • Referral fees for each loan referred through Upstart
  • Amounts to $ 400 to $ 500 per loan on origination
  • Platform fees for each loan originated
  • Amounts to $ 200 to $ 300 per loan on origination
  • Loan servicing fees as customers repay their loans
  • Amounts to 0.5% to 1% per loan

Upstart’s bank partners include Cross River Bank, Customers Bank, FinWise Bank, First Federal Bank of Kansas City, First National Bank of Omaha, KEMBA Financial Credit Union, TCF Bank, Apple Bank for Savings and Ridgewood Savings Bank.

CUSTOMER CONCENTRATION

Of these 10 banks, Cross River Banks generated 65% of Upstart’s total revenue in 2020, down from 81% in 2019. Upstart’s second largest customer generated 15% of Upstart’s sales in 2020.

“for our first four or five years, we just worked with Cross River Bank. We came to a decision point about two years ago: Do you become a bank and pursue a bank charter, or do you decide to be a friend of banks and fan out from Cross River? We chose that second path” Dave Girouard by Penny Crosman for American Banker

Larger banks are not yet working with Upstart as they are figuring out whether they should build the technology in-house.

“The company isn’t yet working with any of the largest banks, as those institutions have to figure out if they want to partner or try to build the technology themselves. But he’s having conversations” by Ari Levy for CNBC

THE MARKET

Upstart now controls 5% of the personal loans market, a market that grew by 8% in 2020. The company is further set to benefit from the growth of the digital lending market which is projected to rise by 11.9% a year over the 2020 - 2025 period.

1. According to TransUnion, there were $ 118B in unsecured loans from April 2019 to March 2020 - growing at 8% year on year

  • Upstart facilitated the origination of $ 3.5B in unsecured personal loans, accounting for 5% of the total market

2. According to Mordor Intelligence, the digital lending market is expected to grow at a CAGR of 11.9% over the 2020 - 2025 period

  • Driven by the proliferation of smartphones, advanced in machine learning and cloud computing

    “Also, technologies like Artificial Intelligence, Machine Learning, and Cloud Computing benefit the banks and fintech as they can process huge amounts of information about customers.” Mordor Intelligence

3. According to the Federal Reserve Bank of St Louis, from April 2019 to March 2020, there were:

  • $625 billion in U.S. auto loan originations
  • $363 billion in U.S. credit card originations
  • $2.5 trillion in U.S. mortgage originations

Upstart’s also ambitions to apply its methodology and models to adjacent markets such as auto, home and credit cards.

“[…] by applying our AI models and technology to adjacent opportunities, we believe we are well-positioned to address the U.S. auto loan, credit card and mortgage markets.” Upstart S1 Filing

This would considerably expand Upstart’s TAM from $ 118B to $ 3.6T.

“In June 2020, we began offering auto loans on our platform, and in September 2020, the first auto loan was originated through the Upstart platform” Upstart S1 Filing

FOUNDER-LED AND EXPERIENCED MANAGEMENT

Upstart is a founder-led company as Dave Girouard is CEO, Paul Gu is Product Lead and Anna Counselman is People And Operations Lead. Management has deep experience in technology, consulting and finance.

Dave Girouard

  • Co-Founder and Chief Executive Officer of Upstart
  • Formerly President of Google Enterprise. Before that, he was a Product Manager at Apple and an associate in Booz Allen's Information Technology practice
  • Graduated from Dartmouth College with an AB in Engineering Sciences and a BE in Computer Engineering. Also holds an MBA from the University of Michigan with High Distinction

Paul Gu

  • Co-founder and Product and Data Science Lead at Upstart
  • Previously worked in risk analysis at the D.E. Shaw Group and has been recognized as one of Peter Thiel's 20 under 20 Fellows, Forbes 30 under 30, and Silicon Valley Business Journal's 40 under 40
  • Studied economics and computer science at Yale University

Anna Counselman

  • Co-founder and People and Operations Lead at Upstart
  • Previously led Gmail Consumer Operations at Google and launched the global Enterprise Customer Programs team. Received a White House Champion of Change award and was recognized as one of Silicon Valley Business Journal's 40 under
  • Graduated Summa Cum Laude from Boston University with a BA in Finance and Entrepreneurship

Sanjay Datta

  • Chief Financial Officer and responsible for leading financial operations and capital markets efforts at Upstart
  • Formerly the VP of Advertising Finance at Google, overseeing the global finance organization. Also held various international finance leadership positions in Asia and Europe. Prior to Google, worked at private investment group Artisan Capital, and began his career as a valuation and data specialist at Deloitte Consulting
  • Has a joint honors degree in Economics and Finance from McGill University in Montreal and an MBA from Stanford University

TAKE A BREATH

So… This is a lot of information. Let’s summarise:

  1. The company developed a model that relies on non-traditional data to predict a borrower’s creditworthiness
  2. It started with the personal loans segment and already serves 5% of the market and it also plans to enter the auto, credit card and mortgage market, unlocking a $ 3.6T market
  3. Upstart is expanding its partner banks’ base in order to reduce customer concentration
  4. However, most of its Upstart.com-generated loans come from one single source, Credit Karma
  5. As for most data-driven players, their flywheel effect is built on increasing the data they can access by increasing their customer base which is done by providing a better (user experience) and cheaper service than competition

FINANCIAL CHECK

  • Sales grew 44% in first 9 months of 2020 to $ 164m
  • Spent $ 80m on borrower acquisition, verification and servicing costs
  • Recorded a contribution profit of $ 63m in first 9 months of 2020 up 151% year over year
  • Despite a large contraction in lending during the pandemic as its partners paused lending
  • Operating expenses reach $ 145m up 32% from $ 110m a year earlier
  • Recorded a profit of $ 4.6m in the first 9 months of 2020 up from a loss of $ 10m a year earlier

BENCHMARK'S TAKE

The Good

  • Upstart is a fast-growing digital lending player that uses machine learning to predict borrowers’ creditworthiness
  • It is able to provide 27% more approvals than traditional players and 70% of its operations are automated, providing instant approval for borrowers
  • It started with the personal loans market which grew 8% year over year and plans to enter the auto, credit card and mortgage markets
  • Upstart doesn’t hold the risks related to lending as it simply refers customers to loan providers
  • Dealing with customer concentration, Upstart is expanding its partner bank’s base and having conversations with larger banks
  • Given its total addressable market, management and financial performance, it is reasonably valued at 15 to 17 times its sales

The Bad

  • The lending market is strongly correlated to the business cycle, negative prospects might reduce borrowers’ propensity to apply for consumer loans as they cut non-essential expenses
  • A rise in interest rates might negatively influence Upstart’s growth as borrowing would become more expensive for consumers
  • Over 50% of Upstart’s referrals come from one single source, Credit Karma
  • Around 80% of Upstart’s revenue is generated from 2 large customers

Disclaimer

Please note that this article does not constitute investment advice in any form. This article is not a research report and is not intended to serve as the basis for any investment decision. All investments involve risk and the past performance of a security or financial product does not guarantee future returns. Investors have to conduct their own research before conducting any transaction. There is always the risk of losing parts or all of your money when you invest in securities or other financial products.