On Shutting Down Funding Community

In December 2013 we announced we were shuttering Funding Community, the debt crowdfunding platform my co-founder and I had worked for the last 18 months to conceive, build and launch.  Because we were early in the shift from the massive growth of consumer peer-to-peer lending into securities crowdfunding and consumer-to-business finance, we were able to build a little excitement around Funding Community.  When we announced the closure there was some interest in what happened so I put together this post to document some of what I learned.

A few highlighted learnings:

  • Partner with someone who shares your views on how to work, but is different enough to challenge a lot of your business assumptions.
  • If the market you are creating has a corollary disruptive market, question what might cause that market to succeed and yours to fail.  Just because peer-to-peer is successful for consumer lending it does not mean it will be as easy for business lending.
  • Businesses in heavily regulated sectors are tough because you either spend all your time and money figuring out how to comply with regulations or how to avoid them. (Obviously if you get it right though the regulatory challenges create a huge barrier to entry.)
  • Don’t underestimate the value of subject matter experts.  We launched without a credit expert as part of our core team and while we could go a long way with consultants and advisors I think filling this hole would have instilled more confidence in our community (and maybe in ourselves).
  • Don’t be afraid to reach out to people you think may be able to help.  A lot of people ignored me, but a few were really supportive and helpful.
  • Don’t get too beat down by the critics, they might have their own agenda in their criticism.

Conception

I left my job as a corporate lawyer five days before the JOBS Act, which was supposed to legalize crowdfunding securities (debt, stock, etc), was signed into law.  While practicing law and doing small startup financings and multi-billion dollar bond offerings it became abundantly clear to me was that massive companies had phenomenally easy access to capital while smaller companies had almost none.  There were a number of companies jumping into the crowdfunding space, but everyone seemed to be doing something similar and the question I couldn’t shake was *why* an individual would want to invest in a company on a crowdfunding platform.  Obviously the idea of earning a massive return was exciting, but what really gets people excited is when they hear about their friend who has a hot new company and they want to invest.  It’s not that dissimilar to the Kickstarter world where the first substantial amount of money in most projects is contributed by friends and family.  I realized that affinity should play a role in whatever investment platform I built.

The most natural place to look seemed to be the university market.  As a Harvard graduate I was acutely aware of the role my university affiliation played in things like jobs, promotion and, of course, interest in investment.  As I talked to friends who had been successful in the business world and wanted to angel investing every one of them told me they would never troll Kickstarter or a general investment crowdfunding site looking for projects, but would absolutely spend a lot of time on a Harvard-based platform.  Call it snobbery, but it made sense.

I started designing and wire framing what was going to become a white-labeled crowdfunding platform.  Basically, we would set up funding.harvard.edu where Harvard students and alums could post their companies and other Harvard students and alums could come in and invest.

At this point I was still working on my own, developing and pitching a product that was not yet fully fleshed out.  I had some interest in what I was working on from a couple universities, so continued to push forward knowing that at some point I would need to bring on a technical partner.  When I met the developer who would eventually become my co-founder the two of us were at WeWork “Summer Camp”, a yearly event put on by the co-working company WeWork in the Adirondacks.  There was only a small area with wifi in the camp and we were both hanging out there one day when we got to chatting.  As it turns out, he was a talented developer building a crowdfunding platform for a Jewish business organization and was interested in what I was working on.  My parting words to him that day were “Well, if you’re ever interested in building a securities crowdfunding platform let me know” and he looked back and said “I’m very interested.”

We met back in New York a couple weeks later and agreed to jump into the project together.  Over time though our goals shifted.  We realized with the SEC was taking its time approving (or even proposing) crowdfunding rules that we were not willing to wait a year before launching our business or even knowing if our business was feasible.  At the same time, we started thinking more about what kind of businesses really *needed* better access to capital.  As a former startup lawyer I knew it wasn’t necessarily your usual tech startup companies.  These companies needed capital of course, but there was a tried and true path for that kind of capital raise, which often involved raising money from experienced investors who couple mentorship with financial backing.  Of course some promising startups would slip through those angel investing cracks, but that didn’t seem like enough to build a business.  Instead we looked at where the real funding gap seemed to be, and that was in the true brick and mortar small business space.

Now, if you’re anything like me, you have very little interest in owning equity in a pizza shop.  The company is likely to be barely covering its expenses, and the best you can hope for is that the pizza shop will throw off small dividends every once in a while.  What this means is that your best option for investing in that pizza shop is actually through debt.  But why would you want to lend money to a pizza shop, except if (a) you were getting a huge return on your money or (b) you personally loved and supported this pizza shop’s success?  We actively wanted to avoid being another merchant cash advance company with high interest rates and fees, so instead we made a commitment to keep interest rates low and focus on helping individuals lend to their favorite local businesses.  At the same time we DID want our lenders to be able to “invest” in their communities, which means that the Kiva model of 0% interest loans was just not right for us.  (Plus, we wanted to be very transparent and Kiva’s model of getting American’s to lend money at 0%, which Kiva could pass on to a local moneylender in the third world to re-lend at 40% just wasn’t appealing to us.)

This approach came with some challenges.  The major one being that it is violation of US securities laws to make a public offering of securities (including debt) except under certain circumstances and we did not have the benefit of one of those exceptions.  Instead, while my co-founder was building the technology I spent months reviewing securities law, case law and SEC regulations to try to find something that worked in our favor.  Finally we got a glimmer of hope.  We found a Supreme Court case discussing the definition of securities, which included a few exceptions, one of which was not ideal, but would work for us.  One of the major components of this exception was the length of our loans, they had to be 9 months or shorter (among other things) in order for our loans not to be considered securities.

Hurricane Sandy

Then Hurricane Sandy hit. Our office building was shut down for a week while crews pumped water from the basement and we had to rethink whether we could have any real sales strategy at a time like that.  Although we knew hundreds of businesses were going to need money, instead we turned our attention to a side project for a couple weeks and built relief.fundingcommunity.com, a small app that allowed any relief organization to better organize its needs for both volunteers and products.  While the intention was altruistic, this project had the added benefit of starting to get our name out there and associated with community support.

With our legal structure organized and this new project humming along, we hoped to launch as soon as possible to help out small businesses that were hurt by Sandy.  In fact, my job for weeks following Sandy was to walk around neighborhoods like DUMBO, Red Hook, Alphabet City and the South Street Seaport to see who I could talk to and how we could help them out.  We were not going to accept any fees on these deals, we just wanted to get these small businesses on the platform where their customers could see them asking for help and support them however they wanted.  We figured customers would be willing to donate to for-profit businesses only for so long before they needed another reason to support the business.  Suffice it to say, this process was grueling. Half the businesses I stopped at were not open, most of the other half had no idea what they were going to do about the money they needed re-open, and were not yet willing to make any sort of a decision.  On top of this, because I could only walk so much and talk to so many businesses in a given day I was not making a huge amount of progress.

Sale Strategy

Earlier that fall I had attended a seminar put on by an old acquaintance Scott Britton (http://life-longlearner.com) on startup business development.  One of the tips he mentioned in his class was a gmail plugin called Rapportive, which shows you all sorts of information about people emailing you (LinkedIn, Twitter, etc).  The great thing about Rapportive is that you can use it in reverse to figure out almost anyone’s email address by typing it into the “To” field and seeing if Rapportive populates with useful information.

So while I was trying to figure out how to get our message out in a time and cost-effective (and *actually* effective) manner I decided to reach out to someone who had done just that, Andrew Mason of Groupon.  (Andrew and the Groupon team get a lot of flak for the expense of their sales team, but the fact of the matter is that they became a small business juggernaut in just a few years by developing an extremely effective sales process).  After just a few attempts I figured out that andrew@groupon.com would work, sent an email asking for help on this problem and 2 minutes later got a response “Hi Alex - introducing you to [someone very high up at Groupon], he would be happy to help.”  After chatting with this Groupon executive, who informed me that no Groupon salesperson *ever* set foot in a customer’s shop, we figured out our own scaleable sales strategy.  We found various data sources, particularly those that aggregated daily deal offers and scraped them for information on the business.  The way we thought about it, a company that had done a daily deal (a) needed money, (b) needed customers and (c) was at least nominally familiar with operating online (which was very important to us). 

Unfortunately, we were struggling with some delays getting the platform completed (even in a basic, but secure state) so we were not able to open up to the few businesses that had decided to come on in the wake of Sandy.  At this point we felt it was better, given that we were a financial services platform dealing with peoples’ money, to complete the technology and not launch until everything was in place.

Pre-Launch 

Finally, leading up to the end of April and beginning of May it looked like we were getting close to launch.  We ran an alpha test with some friends and family, while raising money for Funding Community and learned a ton.  For example, we were trying to find a way to make our payment processing both inexpensive and simple.  Many ACH (bank transfer) companies wanted to charge us high fees because our type of transaction was considered “high-risk” for chargebacks, so instead we started working with a relatively new payment processor called Dwolla.  Now, Dwolla does ACH transactions cheaper than just about anyone else, but that cheapness comes with a price.  For us, it was ease of use.  When we launched our alpha Dwolla had not yet created embedded forms for payment processing, which meant that we had to send our users off-site to make their transfer.  At the end of the process Dwolla did not re-direct the user back to us, but instead kept them on a large advertisement for Dwolla itself.  Dwolla had agreed to give us $10 in credits or $2 cash for each user we signed up, but we did not want to sacrifice platform functionality in favor of  signing up Dwolla users.  At the launch of our alpha we saw a 30-40% drop off rate when people had to go to Dwolla’s site - partially because of the process, partially because of the lack of a redirect and in large part because Dwolla would freeze on our users very frequently - this drop off was with friends and family who had already committed to help us out!

So we were stuck on what to do.  We had already changed payment processors once (we had tried using a company called Zipmark because of its guaranteed speed-of-light delivery, but found it had too many other kinks for us) and really did not want to have to change processors again.  Luckily, late one night while scouring Dwolla’s API documents I came across a new API that no one in business development had told us about (presumably because it had just been released).  It had exactly what we wanted, a secure way for us to embed a form on our end and let Dwolla know to run the transaction, without customers having to sign up for Dwolla or even visit Dwolla’s site.  When we reached out to Dwolla about it they seemed shocked we’d stumbled across that new API and let us know that it was actually designed for a bank that was going to be testing out some of Dwolla’s features.  We quickly jumped in and got them to let us test it ahead of the bank, which required a lot of legal documentation on a tight turnaround, but made our product so much cleaner and simpler.  (As a side note, we were shocked that they wouldn’t have been super excited to have us test the product for them.)

The Launch

With our transaction processing wrapped up we wanted to pick a date to go live so we could organize a little press and excitement around the launch. Given that we were launching a marketplace and could not totally “fake” the first couple months this was important to us.  So with that we picked May 8, 2013 (coincidentally, my 31st birthday).  There were a number of pieces that should have been ready by the time we launched, but just weren’t, so we went ahead and went live with what he had.  Anyone who has launched a product knows this feeling, the concern that not every piece of what you built is quite ready so you are just going to have to fake part of it until it is.  I was not terribly comfortable with some of it, but eventually we just needed to set the platform into the wild and see what happened.

What happened was a partial success.  We went live with four companies, including an additional loan for ourselves.  Now, anyone paying attention should have known that making a loan to us on our own platform was probably the best deal you could find.  The chance of default on that loan was effectively 0% since if we were to default the company would go under and we would end up with a horrible reputation.  The idea of including ourselves on the platform was to get more lenders on board in a way that made them as comfortable as possible with what we were doing. 

Post-Launch Glow

Right after our launch a few exciting things happened.  First, the US Treasury reached out to chat with us about small business lending.  We launched on May 8th and first spoke with them on May 10th.  One of the first things they asked was “so how long have you been around?”  Of course, I guess they didn’t expect my response would be “Two days!” That conversation led to us being invited to Washington, DC to a summit on small business access to capital along with the major players in the space: Lending Club, Prosper, Capital Access Network, OnDeck Capital, Kabbage, Dun & Bradstreet, Intuit and others.  The main discussion there was over the concern that a small player would pop up in the peer-to-peer space and act in an inappropriate way that might bring down the entire industry (yes, this is foreshadowing).  A few weeks later we got a call from the Federal Reserve Bank of Boston asking for some help as they were thinking about small business capital as well.  Heady times to say the least!  

We also set up a booth at the LendIt conference where I ended up on a panel with Funding Circle (now live in the US through their merger with Endurance Lending), Sino Lend (founded by a LendingClub co-founder) and Candace Klein of SoMoLend.  Now, the last person created and interesting dynamic.  Earlier in the conference I had come out of one of the panels and found Candace berating my co-founder that we were violating securities laws.  He kept his composure incredibly well and politely explained that we had spent a lot of time thinking about this.  I came over and assured her of the same, but she was definitely not convinced (and I’m not surprised, Candace is a bright woman, a securities lawyer herself, and had initially tried to do something very similar with SoMoLend as we were doing at Funding Community and could not find a way within the regulatory regime).  This meant there was more than a little tension on that panel!  The interesting end to that story is that a couple months later it came out that the Ohio Securities Commission had sent a cease and desist order to both SoMoLend and Candace just before LendIt. Put this in the category of “you never know what’s going on in someone else’s life to make them act the way they are.”

So after our first couple weeks we had just about funded all of out first set of loans and we were furiously trying to get new ones signed up.  Here was the rub though.  We started to get a little bit of inbound interest, but frankly most of those businesses were in rough shape.  When we looked at a small business making gross revenue of $1,000 a month looking for a $10,000 loan we just could not see how our lenders were going to be repaid.  Of course this was not every business we were looking at, but it was a huge percentage.

Learning

This is when we learned an important distinction between what we were doing and what Lending Club and Prosper were doing in the peer to peer lending space.  The vast majority of Lending Club’s loans are for “refinancing” existing debt (61%) while another 22% are for “credit card payoff”, which is another term for refinancing.  So, a total of 83% of Lending Club’s loans are for refinancing existing debt.  How amazing is that?  You don’t have to convince someone to take on new debt, you just have to be able to convince them you can offer a better deal than their current debt.  On top of that, you can piggy back on other lending companies’ credit analysis.  Our company was built on the idea that the credit markets were too lean for small business, which means that we were built on the idea of originating new debt as opposed to financing old, a much more challenging (and expensive) proposition.  On top of that, we had already limited our pool of potential borrowers since we were dealing only with borrowers who needed, and could repay, short-term capital.  Even more restrictive, because we were trying to open up the ability of non-high net worth individuals to lend, we were dealing with small dollar loans, which meant we needed a high volume of lenders.

This is probably a good time to talk about our pricing model.  If you are reading this may know that Lending Club and Prosper charge an origination fee on each loan and also charge a servicing fee to each lender upon repayment.  This spreads the cost out, and also puts a little skin in the game.  If a loan Lending Club originates defaults Lending Club does not get its servicing fee.  Lending Club’s origination fees are between 1.11% and 5% (mostly in the 4-5% range) and its servicing fee is 1%.

Because all of our loans were 9-month loans and we were trying to keep total cost down to borrowers we felt we could not charge origination fees nearly that high.  Remember, Lending Club and Prosper are each dealing with 3 year and 5 year loans, which means a 5% origination fee does not hurt nearly as much as it would on a 9 month loan (which would effectively be 6.67% on an annualized basis).  We were charging a 2.5% origination fee on the first loan in a 24-month period and 1.5% on subsequent loans in the same 24-month period.  This meant that if a company were to take on multiple loans we could make more on origination fees than Lending Club or Prosper, but there was no guaranty our borrowers would take multiple loans.  From an economic perspective this was a huge challenge for us.  We wanted to keep our borrowing costs down, but also needed to find a sustainable business model.

Speaking of business model, one aspect of our model we were pitching to investors was that small business lending is very different than consumer lending.  When you make a loan to a consumer you really don’t have a lot more to offer that consumer (except maybe more loans).  When you make a loan to a business or theory was that you then had the ability to sell a lot of ancillary products to that business.  As I already mentioned, we were using Dwolla for payment processing and they would pay us to sign up users.  What I didn’t mention is that we actually required all borrowers to sign up for Dwolla to allow us to collect our repayments.  This was our first testing ground on selling ancillary services and actually worked quite well.  That said, we were never able to get far enough down the road to see if our ability to cross-sell would work with other businesses.

When we launched we were in an interesting place.  We were on the very early stage of individual-to-business loans, but because the peer-to-peer space was already red hot a lot of peer-to-peer lenders “got” what we were doing.  This was a double-edged sword.  It meant that we didn’t have to explain the model to these people already in the peer-to-peer space, but it also meant that they had certain expectations.  This became very apparent as we were closing our first loans, but did not have a fully functional dashboard available to show the status of loans-in-funding, repayment dates etc.  We started to get questions and concerns about what was happening with peoples’ money.  It was all safe and was all being put into the loans intended, but because the transparency was not there some lenders became very concerned.  Had no one done what we were doing before this might not have been as much concern, but because Lending Club and Prosper had 7 years of operating experience and had fully-functional and very transparent platforms people’s expectations were that we would operate in the same manner from day one.  To quote one lender “The start-up excuses are BS.”  That kind of hurt… 

So there we were.  For regulatory reasons we could only offer 9 month loans, which led to a sales issue of only being able to provide capital to a small percentage of small businesses, and this in turn meant that our relatively lower economics could not be made up as easily by scale.  On top of those issues, we were struggling with some technological issues (including the dashboard problem) that made me less comfortable pushing our product to lenders than I might otherwise be. 

On a personal level, my relationship with my co-founder was starting to deteriorate.  We were agreeing less and less on where the product should be headed, on how we should handle sales and customer relations, and even things like how to work together as a team. I’d heard it said many times as a startup lawyer that “team” was vitally important to a company’s success and that co-founding a company with someone was like marrying them.  Just like a marriage, when it is going well it feels so natural and doesn’t feel like “work”, but when it isn’t that all changes.  Also, just like a marriage, if other things are going well (like your relationship with your children), you can focus on those and gloss over the issues you have with your spouse.  Had Funding Community been wildly successful in its first couple months we likely could have glossed over a lot of issues.  Eventually more and more of my mindshare was going towards focusing on our internal dynamics and away from our core business.

Heading through August and September I needed to make a decision on what I was going to do.  We could not keep working together, but enough of the platform was still being run by hand that I could not run it alone or just bring in another partner or employee.  After a few weeks of holding off posting new borrowers to the platform while I tried to figure things out, we sat down to have “the talk”.  For those of you who have gone through this before you know exactly what is about to come.  “It’s not you, it’s me” and other similar platitudes.  Yes, it was a breakup.  After a long discussion we mutually decided that the best course of action was to stop accepting lenders and borrowers on the platform, but that opened up a new can of worms.

If you have an iPhone app you’ve launched and decide to stop supporting it, the app just continues on in the ether while people either stop using it or continue using it without support.  However, we knew that because we had all sorts of financial obligations outstanding we could not take this approach with Funding Community.  At the same time, we knew the perception would be that once we shut down lending operations we would no longer be incentivized to actively service our outstanding loans.  While we knew we would continue to service our loans, we wanted to avoid all appearances of impropriety and decided that we would repay all of our lenders all outstanding principal and interest owed at the time on each loan.  Remember how I said the main concern at the Treasury was that a little guy would bring down this whole industry?  Well, we wanted to ensure that we were NOT that little guy.  So with that, on the date that we picked to repay all our loans (December 27, 2013) we accreted the next monthly payment, so if a payment was due on January 4th, the lender instead received that payment on the 27th, a couple days early.  We then repaid all principal remaining on that loan, the idea being to put our lenders into a better position than they would be had we not done anything.

 We made the repayments on December 27, 2013 and then notified each lender of what was happening and began processing our withdrawals.

So, what’s next?  Well, my co-founder and I have mostly gone our separate ways.  We’re still tied together by a corporation with loans outstanding and a few users who have not yet collected their money from us, but we mostly just do what needs to get done to keep the ship upright.  I am doing legal work for a few startups and consulting with a couple of crowdfunding companies helping to share some of the things I learned along the way.  Anyone interested in chatting with me about this should feel free to reach out to me at alex[at]bnkly.com.