Why I'm long UPST

I’m extremely frustrated in that I had composed a rather lengthy post on this subject which somehow just got blown away when I poked the “Submit Message” button. But, oh well. Here’s the condensed version.

Upstart will post earnings Feb 15 after the market closes. They provided revenue guidance for the present quarter with the high end of $265M. I am expecting a beat at least on a par with 3Q21, 6% yielding revenue of $280.9. That would be sequential growth of 23% and 223% for the same quarter a year ago. If I’m right, those are very good numbers. Yes, I know, analyst’s consensus is below the high end of guidance. I think they’re wrong. CEO, Dave Girouard said something to the effect that Upstart will blow everyone away this year. I don’t think he was just blowing smoke. I think he meant it. I think we will see the beginning of that this quarter.

But to be honest, that’s not the only reason Upstart is 8.5% of my portfolio. Unfortunately, the aforementioned lengthy, lost post provided my arguments for holding this large a position. I’ll try to summarize.

A scientifically validated risk assessment model disrupting the 30 year old FICO score which is based on a weighted average of five “common sense” components.

Upstart does not lend money. They do not compete with banks. They partner with banks and collect a fee for service.

An unassailable moat. Not just the AI model, but having pre-addressed regulatory hurdles.

Expanding markets:

  • Hidden prime borrowers of personal loans
  • Competitive rates for actual prime borrowers of personal loans
  • Purchased auto loans
  • Refinanced auto loans
  • Micro loans (disrupting payday lenders)
  • Web interface for Spanish speakers
  • Mortgage loans (on the horizon)
  • Business loans ???

70% of personal loans fully automated obviating the need for borrower to produce copious documents and underwriters to process the documentation.

There’s more, but those are some of the more important ones.

But yes, of course, there’s a downside. Here’s why my position is only 8.5%.

  • No ARR. No subscriptions, no recurring revenue. Every quarter they start over, so to speak.
  • Cross River Bank and Frontier Bank originates about 84% of their loans despite a growing number of partner lending institutions.
  • Credit Karma accounts for about 45% of their traffic
  • Fraud, they beat back an organized fraud attack in Q3, but it was a drain on resources requiring a lot of manual effort.
  • Rising delinquency rate of Upstart loans. This was anticipated and previously addressed, but it is a valid concern. But, the target is profitability, not zero delinquencies.

There’s more, but that’s enough . . . Despite the downside, I think the positives significantly outweigh the negatives. We’ll see if I’m right when they report. If I get disappointed, I will likely sell about half my position and buy BILL with the proceeds. I had kept my position in BILL very small because I had difficulty understanding the company and it seemed a bit like Lightspeed to me (which I also had trouble understanding).

If you want to learn more about Upstart let me suggest you read the linked document below. This link was to posted to the board previously, but I don’t recall who posted it. I am appreciative, but unable to give credit.



Hi Brittlerock,

I like your logic. Thanks for posting it.

Here’s mine-

When trying to explain to myself why I’m going into Upstarts Confernce Call today with a 10% position, I’m very appreciative of all the work presented here for why I should invest in other companies instead.

Perhaps everyone has gone through this exercise in there heads. But, I haven’t read it on this Board. Or, I don’t recall if I have anyway. I’ll admit that this last years influx of newbies has changed my habit of making a point to reading everyone of the posts. So as to not risk wasting anymore of your time…

My thinking like a human led me to believe Upstart is focusing on generalizable factors when determining whose most likely to repay a loan. In order to understand Upstarts business model, I not only need to understand banking. I’ve spent the last six months learning to think like a computer and what the differences are between Strong and Weak AI.

To take the time to respond to every reason not to stay invested in Upstart, at this time, is not possible for me. But I will say that my understanding of the Strong AI, which I believe Upstart has, is capable of navigating everyone of these challenges.

Now remember the following has complete regulatory approval and has been chosen by the National Association of Credit Union Presidents.

Here’s why I stay at a 10% allocation in Upstart:
Per Paul Gu, Upstarts’ AI is not trying to mitigate macroeconomic factors as much as determine the ‘type’ of person that pays back their debts and correlating the significant individual life circumstances for each person that might change their ability to pay. Me here: This sounds like an impossible amount of information about just one customer. How could this scale? It is possible with todays computer algorithms. In fact in addition to initially going deep into the particulars of every customer, most customers will become repeat customers and Upstarts algorithms will need to add newer information to get better with making the decision to lend or not.

We also know, thanks to prior conferences given by Upstart management: Upstart is also correlating information about each customer that helps determine what the due date each month should be to increase the likelihood of repayment. How many other benefits to the lender are possible when Upstart is able to go so deep with each and every customer? Advertising particular loans perhaps.

I haven’t posted this before because it has a similar smell to ZoomInfo (why we believe that the stock price doesn’t correspond to the numbers- aversion to the collection of personal information). Upstart does have what everyone wants, access to money, where ZoomInfo gets you a call from a sales rep. Banking has maintained a air of respectableness for some reason. So that gives Upstart the advantage in one respect anyway, IMO.

I believe given the above ability by Upstart, how could I not have at least a 10% position. Writing the situation here like I did, I believe, addresses every concern raised by anyone. I believe that this directly explains my positions in many other of our companies: Snowflakes Data Sharing ability is similarly justified- ‘regulatory hurdles? Done and Checkmate’. My own assertion about justifying owning Cloudflare? Their individualizing regional regulatory hurdles? Done and, with the number of developers hitting 1 Million this year on their Workers
product, checkmate.

Aside from Snowflake and Cloudflare, do you see the logic in my argument concerning Upstart?




I’m not sure I follow all of your logic, but I certainly concur with some of it . . .

"Upstarts algorithms will need to add newer information to get better with making the decision to lend or not.

This is really at the heart of their moat. It’s not just that Upstart has a head start on the competition and that they one of the key ingredients to refining the ML models is time, it’s that Upstart is continuously improving their models based on the accumulation of new information. They are way out in front of the competition and with each day they advance that much more. This is the same thing that makes it so much easier and more efficient for banks and credit unions to partner with Upstart rather than role their own.

Upstart does have what everyone wants, access to money, where ZoomInfo gets you a call from a sales rep.

OK - but I don’t really agree with this. Zoominfo is not supporting telemarketers making cold calls. They are a BtoB business. Yes, a person gets a call from a sales rep, but the person receiving the call is a buyer of the type of product the caller has to offer. Put simply, Zoominfo helps sellers and buyers who share a common interest with respect to products get together.

To be honest, as an investor with a modest position in Upstart I don’t think it’s necessary for me to have an intimate knowledge of banking. And most decidedly, I do not possess this knowledge. I just need to know enough to understand why partnering with Upstart is a very good business decision for a lending institution.