We're building a new kind of bank - Yes, an actual real bank. We're starting with a small acquisition and re-building the underlying tech from the ground up. The ultimate goal is creating a bank consumers would actually love to use.
Think of paying your bills via "push" vs "pull" so instead of simply paying someone on autopay and authorizing them to debit your account at their will we'd actually say something like "hey, PG&E wants you to pay $55.86, would you like us to send that over now" when you respond they'll get your money but not your actual account info.
Basically instead of using the typical system of handing out your actual account numbers we generate an account number that works for only that vendor and if that account number ever got stolen it wouldn't matter because its locked down on a unique vendor or amount basis.
The main pain point for us right now that we'd like to talk about is re-thinking savings and loans and what the largest consumer pain points are in modern (or lack thereof) banking.
I think there's a huge need for an actual new bank that is built on technology. It's one of those obvious big need problem.
One of the big problems founders usually run into when doing this, though, is dealing with the mass of regulatory requirements involved. How are you thinking about this? Which of the many different services offered by a bank will you target first? Is it going to be that payments piece you are describing?
One of the hardest parts of the regulatory requirements is the Capital on hand and getting a federal charter.
Because we've targeted a handful of very small institutions with those charters in place and the banks are in good standing it actually allows us to skirt some of the pain points of trying to charter one ourselves but also allows us to hit the ground running on the technology front.
We think the bill pay/secured payments could be the one to target first because we think there is value in just being a great traditional savings and loan bank. However, the main thing we're thinking through now is if this is a big enough segment to gain real traction with consumers or if we need more of a WOW product for mass adoption.
I was thinking about this recently - there is a need for a meta banking service - basically a p2p banking service much like Venmo -- but also allowing for vendors.
The actual funds can still be housed in a traditional bank - but you can still have an account or CC number which you would give a vendor.
Whenever the vendor attempts to charge your CC number the amount goes on escrow hold only in the meta-space until you the user will auth it to actually come out of your account.
It sits between the real bank and the vendor and you and gives one more control over their money.
Thats exactly the idea that got us to this point Sam.
To be honest we had no intent of sharing this much publicly before there was more to play with out there but we couldn't pass up getting some advice from Aaron and Michael as we've been going through this process and trying to figure out more pain points for the end users.
Cn I contact you and give you all my personal pain points, I have actually been thinking about this for months -- but only internally. I'd like to vent and give you some context as to why.
Ill gladly sign an NDA - you actually dont need to tell me too much - but I'd like to just brain dump on you in the hopes that my issues will be addressed by someone.
I want to opt out of the traditional banking system 100%.
I want to have integrated books. why can't a bank just sell me the invoice, payroll etc. services so I don't have to import them. Rules for employee spending. employee X's debit card can only buy from these stores fort this amount. Employee Y has limit of $500 per day. and I should be able to go to my smart phone to change the rules.
The way we've figured out so far is really learning from a failed past experience. We built a product in the past which was a kind of Yodlee or Plaid competitor (http://spout.co) and learned alot of what consumers cared about along the way while never hitting an actual nerve with clients who were willing to pay us money.
I think we started like most people would here, by partnering with existing banks. So we have a signed pilot with one major bank and are working on a partnership with another one. But knowing how slow they move we started looking at other avenues that could make a more immediate impact.
The pilots are around lost or stolen cards and being able to have your autopayments as proxy cards so when the number of the card on your Target or Home Depot account gets stolen the number of the card in your wallet doesn't have to change and vice versa.
What we're working on now is a way for people to identify bills in their email that we can then process and remind them when they're due and how much money they may have left in their accounts when paying them. This allows us a couple of opportunities.
1. To get consumer eyeballs on something and to see if people would like to pay bills via a simple Venmo-like product.
2. Building a base of potential users for an actual bank launch.
In that vein some of the underlying tech behind an actual bank itself becomes the most interesting part of the problem once a transaction is approved.
But until we get behind the kimono we really don't think theres a way to know how much we'll have to build or improve with the underlying tech and systems but we suspect what is built can be leveraged in several ways.
What we do expect is a complete questioning and re-thinking of how every single thing is done from how customers deposit money to how we issue new accounts and how underwriting is done on loans of all types.
That seems like a pretty good way to start getting relationships with customers. How is it going? What have you learned from seeing how they use your product?
Well since we started with the pilots I’ll go there first. They progressed very quickly at first. One meeting turned into three which turned into 5 more and then we actually got wireframes and mockups made. And then everything seems to go into super slow-mo.
When that happened with one bank we used the first relationship to leverage into a second one with another large bank, after going through a near identical process we realized that big banks all move much slower than we can (even the ones that claim they can move very quickly).
At that point we started pursuing relationships with smaller banks and those seem to be significantly faster albeit still very slow in comparison to a startup.
The most recent revelation with the bill pay stuff has literally been over the course of the last 7-10 days because we threw out what we thought was a hairbrain idea to buy a small bank to an investor we knew well.
He got excited and offered to fund the entire deal if we could find one that we thought could work well.
That kind of kicked ideas into hyperdrive about what we could build to attract users now instead of waiting, and the tool to identify bills within emails was thought up and began progress on just so us and some of our friends can try it.
It’s not a launched product yet, but its close to something we could stick in the chrome store as an extension and see the feedback we get.
I'm curious about your model versus something like Aspiration [1] that is partnering with Radius Bank? Do you need to own a bank to provide the service?
The short answer to this is that by partnering with a bank (we have before and will continue some relationshops we've already started) you are limited by their underlying technology platform in the data you get, the data you pull out and anything you'd like to do with that information. Even how you'd like to show it to the customer.
Everything is vetted through their counsel and their management team and your hands are always tied by their wishes.
While you might be able to find a super innovative bank that allows you to have some creative freedom its really the difference in owning a home vs renting a home (even if you have a very lenient landlord).
When you own it decisions are yours to be made, good bad or ugly, when someone else does there is always someone to answer to and there is always a period of time that has to pass between thought and action.
That period of time in a small and nimble startup we believe is the difference in success and failure
Not very thoughtful 2c: success or failure of a bank is driven more by where you choose to invest your capital than trying to improve the consumer experience and secure deposits. Due to the way the system is set up, the consumer is almost irrelevant in the equation. So while their might be some margin to squeeze in improving the consumer experience, it's probably insignificant vs. say, picking the right bundle of mortgages to buy.
What about building a frontend as a service and selling it to other banks? That sounds like a better way to focus on a tech/product edge.
we are buying an actual bank, its not a credit union we're just buying a small one in good health and with a charter already in place.
The idea is that we can expand from there, but this is not something we are trying to form and have a new charter issued, this is continuing on an existing bank that we hope to be able to use as a testing ground for our ideas with actual customers
We tried this at Daltic and I know Standard Treasury (YC) tried this as well.
Buying a small bank with a healthy federal charter is definitely the way to go over trying to start a new charter (FDIC hasn't given out a new charter in years and even if they do, de novo status is very cumbersome). The real challenge is getting around the Bank Holding Company Act with respect to raising capital. Basically, if any entity has a significant stake in a US bank (5-10% can trigger this) they are subject to BHCA which includes scrutiny from federal regulators (Plural!) over your books and a legal duty to bail the bank out. Our idea around this, consulting with a top financial attorney and consulting firm, was to get a consortium of wealthy individuals who would not exceed the limits individually and wouldn't be "acting in concert" with each other (which also triggers BHCA).
But even after this, any major shift in the bank's operations, including a re-architecture of technology is subject to regulatory oversight, similar to de novo status. Also, not to mention background checks from not only the FBI but also the CIA. Replacing management (which you probably want to do) is also a major change in bank operations. You also need a CEO who the regulators approve, and has previous CEO-level experience at another charter. We also had someone like this lined up.
We got pretty far in this process, but decided to put our efforts elsewhere. Seems ST did as well. Curious if you've found the same things, happy to chat further about what we learned (email in profile).
Edit: Of course, all the fun begins once you successfully acquire a charter. The real business in acquiring a bank is not to make money on loans. It turns out, despite the monopolistic access to cheap capital from depositors, non-bank lenders do a much better job at making high margin risk-adjusted loans (see https://foundationcapital.com/fcideasfintech). I believe the real opportunity for a tech-first bank is to be a whole sale and commercial bank, and tackle the $300B+ treasury services industry.
They make a mechanism for certain types of users to create accounts -- lets say only those with a valid HN account with a karma and age of over X + Y
I create the account and get the account and routing number
I then use this to direct deposit some portion of my pay into this account.
I have no access to the account aside through their meta app.
I give full autonomy to their bank and app to the funds in that account -- and I only get reports on usage and spend through the app - and I gate keep all transactions with a yay/nay -- but I never have access to the account itself in the chartered bank.
Instead, I just must commit to depositing say $1K per month in the account.
In the app - I can then only sign up for transaction types or vendors that fit within that particular deposit level.
They don't raise money for the bank at all.
They raise money for the app only.
(needs tweaking -- but the seed of what I mean is there)
unfortunately there are regs about chartered banks doing non-banking activities ... we'll talk more over the weekend, but I can give you way more details
to be honest Cory the last thing we have thought about so far is fees.
In all likelihood we'll slash current fees and get down to the nuts and bolts of what makes our customers tick, we'll re-think our lending practices and try to find risk-averse loans that make sense for us and our customers and allow us to continue to be a stable bank while becoming something the average customer is proud to work with
Lets hope you can innovate on the "overdraft business model" and offer people to buy into a credit card as opposed to egregious graft on over drafts.
Further - you should have full transparency in charge ordering.
The method by which banks order charges to algorithmically produce the most overdrafts should be a capital crime.
I used to bank at a CU and they posted their monthly overdraft profits on the wall: ~$250,000 per month they charged their customers in overages. When 2008 hit, this number balloned to ~$500,000 PER MONTH that people were paying in overdraft because everyone was getting fucked so hard by the banks.
On paper I make a lot. I would literally pay a sizeable amount to have a true banking partner that actually had my best interests at mind and wasnt trying to rob me.
A friend of mine recently complained that she had dropped to $22.00 in her BofA account, which had a $25 min balance req. They then charged her a $35 fee for being under $25 - this caused her to go negative, and then for being negative they charged her an additional $35 PER DAY.
Think of paying your bills via "push" vs "pull" so instead of simply paying someone on autopay and authorizing them to debit your account at their will we'd actually say something like "hey, PG&E wants you to pay $55.86, would you like us to send that over now" when you respond they'll get your money but not your actual account info.
Basically instead of using the typical system of handing out your actual account numbers we generate an account number that works for only that vendor and if that account number ever got stolen it wouldn't matter because its locked down on a unique vendor or amount basis.
The main pain point for us right now that we'd like to talk about is re-thinking savings and loans and what the largest consumer pain points are in modern (or lack thereof) banking.