OnDeck uses Applied Big Data to help small business get lower cost debt

OnDeck is one of three companies on my Fintech IPO Watchlist and a pure play in the most broken market out there – small business lending.

If they are heading to IPO (intention announced, not yet publicly filed), they must be doing something right and that must be based on some “secret sauce”. (I use the term secret sauce for any combination of technical, process or business model innovation that gives them an edge).

First, I wanted to eliminate what they are NOT doing. OnDeck Capital is not offering financing based on:

  • Factoring




  • Merchant Cash Advance.


Small Business Owners out there who have suffered through Factoring and Merchant Cash Advance are heard muttering,


“well, thank goodness for that!”


OnDeck is not a cheap source of lending, but small business lending is such a totally broken market that anything that brings competition to this market is to be heartily welcomed. So the interest rates that you get from OnDeck might seem usurious, but today they are competitive in the world of small business. In simple terms (the kind of terms that a small business owner wants), I would characterize the OnDeck value proposition as:


“The ease and speed of Merchant Cash Advance with interest rates that are closer to Factoring.”


That may not sound like totally disruptive breakthrough proposition, but when you create lending platforms like Lending Club and OnDeck you have to keep both parties happy – Borrowers and Lenders. You have to give both a bit better than they are getting at the moment. That is how you make money as a lending platform.

Lets look at this first from the POV of the small business owner.

Most small business owners consider Factoring as a necessary evil, not something to aspire towards. So, who on earth would want to aspire to the interest rate that you get from Factoring? The answer is any small business owner who has paid 70% or more for Merchant Cash Advance.

The beauty of Merchant Cash Advance from the POV of the lending platform is that it is highly scalable and automated. The small business owner gives up the right to future revenue in return for cash now. It is scalable because the money comes back automatically via the credit card payment networks. Lots of Merchant Cash Advance platforms are growing like a weed. On the negative side, the interest rates for Merchant Cash Advance (MCA) are so high that many small business owners are giving up or going broke. Merchant Cash Advance is the entrepreneurial equivalent of Pay Day Loans. One small business owner confided to me that the reason that he sold his business and went back to being a salaried employee was when he realized how much money he was losing on Merchant Cash Advances and he figured that the treadmill was not worthwhile.

Paying 70% APR is OK if you use it once or twice. When it becomes a dependency, then you really are paying 70%. The MCA lending platforms make money when you become dependent.

Factoring is fundamentally the same business logic (in B2B, while MCA is more B2C). You are giving up the right to future revenue in return for cash now. With Factoring you pay APR that is more like 24% to 36%. That is a lot, but seems like a screaming bargain compared to the 70% that you pay through MCA.

The problem with Factoring is that it has not been scalable. It is a labor-intensive business (for both the Lender and the Borrower) with lots of small Factoring companies. Inefficiency means high cost.

OnDeck is NOT doing Factoring. However they are offering lending rates that look pretty similar to Factoring but with a scalable, automated process.

OnDeck is using Big Data. Yes, that term is heading on the downward slope of the hype roller-coaster towards the “Slough of Disillusionment”, but this is the application of Big Data to a real business problem. This is the sort of use case that maybe referenced when analysts describe how Big Data eventually climbed up to the “Plateau of Productivity”.

OnDeck is automating what we used to do manually through Banks. As this article in Forbes puts it, “impress the Algorithm not the Loan Officer”. OnDeck is using Big Data to automate the collection of all the piles of paper that we used to bring to meetings with Bank Managers, such as:

“business plan, income tax records, personal financial records and even a résumé.”

OnDeck can automate the data collection and  processing and that is where the clever algorithms come in:

“On Deck crunches that process into a couple of hours. After filling out a short online form, applicants upload 90 days’ worth of cash flow data. “From cash flow alone we consider about 100 different variables,” Breslow says, including transaction frequency and volume, seasonal sales, expenses and returning customer revenue.”

Automating the data collection and analysis is only one part of what OnDeck does. The other part is automating the repayment. This is what makes it similar to Merchant Cash Advance. You do not send money to OnDeck for principal and interest payments, they extract these payments automatically from your bank account.

The big data analysis and automated repayment mean that OnDeck is keeping default rates as low as 5%.

The cost to the Borrower is pretty high:

  • 2% origination fee. Banks have always loved origination fees, as they encourage roll-over.

  • 18% to 36% on 3-to-18 month loans (based on secondary sources, I have not yet tried lending through OnDeck).

This is for loans ranging from $5,000 to $150,000 with an average of $30,000.

My conclusions so far:

  1. OnDeck has a great business.

  2. Lenders are getting a screaming bargain (18% to 36% with 5% default with Libor rates close to zero…).

  3. Borrowers are not yet getting a screaming bargain, but these are early days and competition will do what competition always does.


9 thoughts on “OnDeck uses Applied Big Data to help small business get lower cost debt

  1. I was wondering if you could elaborate on the applied big data that Ondeck uses.

    Do you happen to know what data analytics process or platform they use?

    How many data points do you imagine they use?

    And how many do they retrieve externally?

    Why is it that they are not offering a rate as competitive as a bank if they have more data on any given borrower than a bank?

    It seems that applied big data would assess the risk of a loan more accurately than traditional lending institutions (that should be the innovation), yet their rates are extremely high.

    Great blog.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s