Self: How we accidentally built the backend of a bank
By James Garvey, CEO of Self Financial
If you haven’t read my first two blog posts, please check them out:
Back in May 2014, I realized that there is a huge problem that affects millions of Americans: 1 in 5 U.S. adults do not have a credit score. Most lenders require credit history in order to get credit — yet a person cannot get a credit score without having access to credit. It’s a catch-22 for millions of people.
I dug into the Center for Financial Services Innovation’s
and Consumer Financial Protection Bureau’s
research papers, and came up with an obnoxiously simple idea: what if you could lend yourself money to build credit?
Hello, Self (Self Lender at the time).
Funding Self: June 2015–July 2015
After Techstars Austin in June 2015
- we closed $1.5mm in seed funding from Silverton Partners
- signed a letter of intent with a bank partner
- and also lost a co-founder.
With the money in the bank, we brought on an amazing team of engineers and experts with deep financial knowledge.
The technical name for what we’re trying to do is called a “CD-secured installment loan” or commonly called a “credit builder loan.” It’s a common transaction that credit unions have been doing for decades that has both a savings and loan component — simultaneously. Here’s what we wanted to do:
- Customer applies for credit builder loan
- Bank lends customer $1100 structured as a 12-month installment loan; however, the loan proceeds are deposited into a fresh 12 month, certificate of deposit (CD) account in the customer’s name (and earns interest)
- Customer makes $98/month payments for 12 months. Payments are reported to the bureaus
- At the end of 12 months, the customer has paid off the loan for a total of $1176. The CD matures with $1101.10 for the customer to take out
How hard could it be?
Building Self’s Core: August 2015 – November 2015
August 2015 — we still did not have an official deal signed with our bank partner. Partnering with a financial institution takes a very long time for obvious reasons.
As part of the process, we talked to our bank partner’s software provider to map out the software integration between Self and the bank. We discovered that there’s a huge monopoly in the bank backend software market
: four companies control 96% of the software that powers banks.
Your bank may be paying $75/customer per year — or more — just for software
. It’s now obvious why banks do not do small-dollar lending.
Bank software is ridiculously expensive.
Inside Self, the credit builder loan produces a total of $76 in earned interest. When you subtract standard bank software provider core fees, setup fees, documentation fees, and 6-month timelines for even getting PDFs — we had two choices:
One of our core CRM screenshot^^
- To give up.
- Or to build out the backend of a bank, i.e. our own API-driven core to power the partnership for the credit builder loan.
In order to actually launch Self, here are the challenges we have for each customer:
- customer onboarding (e.g. handling customer PII, security, KYC/AML/BSA compliance)
- loan origination (e.g. creating the loan docs)
- deposit account origination (e.g. creating the account docs)
- $1100 loan asset (which should turn into $1176 for us)
- $1100 bank account deposit (which should turn into $1101.10 for the customer)
- loan servicing with payment collection, credit reporting and customer service
- customer relationship management with recorded and tracked calls, emails, texts, chats
- accounting (interest and payment calculation) and payment reconciliation with 100% accuracy
This is why bank software is expensive. We had to build all of this stuff.
October 2015 — we passed an SSAE 16 SOC 1 audit
of our internal processes and controls. Third-party auditing is imperative because it makes us accountable and gives financial institutions and regulators confidence.
At the end of November 2015, we finally signed our bank partner (it took 9 months from “intro” to inking the paperwork).
But that’s not the end of the story.
Regulators: December 2015–February 2016
In December, our bank partner had a federal exam. Thankfully, we invested a lot of time, money and enthusiasm into compliance, security, and making sure that we’re doing things the right way.
We are passionate about compliance and security because we’re working in a highly regulated industry. It’s money.
Compliance is serious @ Self^^
Self Launch to 43 states on February 27, 2016
The customer experience inside Self.inc^^
After 19 months, we finally did it.
Self is the first do-it-yourself platform to get a credit builder loan. We’ve raised about $2mm, have partnered with a bank, and are now live in 43 states.
Building the backend of a bank was blessing in disguise. We didn’t plan on doing this, but it was the right decision. In doing so, we were able to launch Self — while increasing margins in the short term, and opening the door to additional innovative products in the future. Learn more about Self's credit builder loan
and credit-builder card
About the Author
James Garvey is the CEO and co-founder of Self Financial, Inc.