On Deck Follow Up: Interview with David Booth and Erik Torenberg

I sat down with On Deck's leadership to chat about their past, present, and future.

Welcome to a special edition of The Flywheel! This post is a follow up to Tuesday’s article about On Deck and its potential to disrupt the MBA. If you haven’t read that one yet, it might be worth doing so before jumping into this one! But totally up to you.

If you’re new around here, read more about The Flywheel here. And if you haven’t subscribed, click below to join 1,692 of your smartest friends and peers who receive The Flywheel in their inbox every other Tuesday (plus on special occasions like today!).


Hi everyone 👋🏻! This is the first of what I hope will be an ongoing series of founder interviews. Please reply to this email with your feedback! I’m relying on it.

If you know a founder who I should interview for The Flywheel (or are one yourself!) please let me know. Onto the interview.


An Interview with Erik Torenberg and David Booth

I sat down with On Deck CEO David Booth and chairman Erik Torenberg to discuss all thing On Deck. My questions are bolded, and their words are lightly edited for clarity. For more about On Deck, you can read my insights from earlier this week here.

On the Origins of On Deck

Tell me about the early days of On Deck and why you made the original pivot away from dinner parties towards an online community?

Erik Torenberg: The difference [between where On Deck was at the time and other cohort type models like universities or accelerators] is that they compound. On Deck, when it started, was kind of an events business, or events community. We would have events all over the world, but we didn't really see continuity, we didn't get stronger over time.

When you look at institutions like Stanford, or [startup accelerator] YCombinator (YC), they are very hard to disrupt. We think of these as ‘curation businesses’, because they are selecting who can participate.

I think people underestimate the moats that these types of businesses have. One of the reasons why universities are hard to disrupt is that the alumni have so much incentive for the credential to mean even more, and for the network to remain valuable. The more you pay forward, the more the network gets valuable, the more it's valuable to you, and so on.

On top of that, they are enormously hard to get off the ground. You need to either partner with existing mega brands, or find some novel approach that gets great people in the door. At first, people are sort of taking a brand hit by joining your thing instead of the more established path, so your approach needs to offer corresponding utility to make up for it. And for us, it was really a sort of in between phase of people’s careers, that was sort of our wedge. I wrote about this more in a Tweet thread:

On On Deck’s Expansion into New Horizontal Categories

Let’s talk about On Deck’s evolution from a one-program company to now nearly 10 different types of fellowships. These days you have a new fellowship launching every week it seems!

ET: One of the biggest lessons I learned from Product Hunt (editor’s note: Erik was employee #1 at PH) is if you expand into things that don't make the existing things stronger, or more precisely, are uniquely enabled because of your first category, then it's like you're building a new startup. And so we did products, like gaming products, or podcast products or books, and it wasn't the same audience. We weren't uniquely enabled by our existing business. They neither made the core offering better nor benefitted from our existing core audience. We were just building different startups. So that's why we're really precise here on what are things that are uniquely enabled by your existing fellowships, existing audience, and what makes it stronger.

David Booth: I think about supply and demand. Every program we launch should bootstrap supply for an existing program’s demand. We have founders in On Deck Founder Fellowship (ODF). Great, well what does a founder need? They need capital, they need distribution, etc, etc. Let’s start On Deck Angel Investor Fellowship (ODA) which is a supply of capital for the early demand of founders. And so on.

But it’s also two sided. What do angel investors want? They want to invest. And ODF founders are starting companies.

On Prestige vs. Scale in Curation Businesses

What lends ODF prestige today, since we haven’t yet seen that mega-unicorn come through?

DB: We accept less than 10% of people who apply, so if you get in it’s worth bragging about. That prestige has immediate value — it’s like a credential, helping fellows hire, fundraise, or explain why they quit their jobs (“I’m not unemployed, I’m on deck!).” There’s also utility in the education component — peer to peer knowledge sharing, expert-led sessions and cohort-wide resources.

However, prestige and education alone aren’t sufficient to satisfy the real reason many join On Deck. 

Many come in looking to find co-founders or early hires. The more people there are in the network, the more prospective co founders or hires, people to learn from, or support you along the way etc. So for On Deck, our balancing act is to carefully measure our expansion such that any drop in “prestige” associated with an increase in network size is offset by raw utility value. Take LinkedIn as an example, at the extreme end of the spectrum. Having a LinkedIn profile is not prestigious — but it’s incredibly useful. Anyone can do it. Everyone has it. Everyone has a profile and they use it to search and everything else.

ET: One is the utility of the program. Did you get value out of the program? Did you meet great people? Did you have a good experience? Two is who's done it. We have had for example a VPs of Product at places like Flexport and dozens of people who have been founders at big companies, saying, ‘ you know, this is worth paying $2,000’, and being really involved in the community.

DB: I realized that we're actually perhaps being a bit misleading holding out prestige as being so important, because think how many people join On Deck, and don't tell anyone. They deliberately don't want anyone to find out, because they don't want the boss to find out for example. And so why do those people join and pay $2,000? The only conclusion is utility, because it helps them figure out what to do next.

On the Business Model in the Future

I want to ask about scalability. As the community grows, there’s an upper limit to the number of members, at which point the community breaks, presumably. And today the business model is to charge the price of admission. So how do you think about scaling it, and will the business model sustain?

DB: if you look at a school like Harvard Business School, only about 1/3 of their revenue is from Tuition. Their largest revenue stream is HBS publishing. 

So one is I think we are by no means bound by tuition fees as the end of our model. But that said, I think it’s an incredible incentive alignment. People pay and then they want to get the value of it. And the act of getting value out of it creates value for everyone else. So yes, we will continue to charge tuition fees. But like other institutions, once our utility function rises past a certain point in the future, we will add additional products.

I’m glad you brought up incentives. I wanted to ask about it because one of the interesting things in education now is ISAs, and the story there is that it aligns the incentives perfectly between student and institution.

DB: What I dislike about ISAs broadly is that it’s a very risky/expensive form of debt. If you think about debt, it has cost of capital, and the safer the borrower, the lower the cost of capital. And also the more rigid and secured the terms, the lower the cost, right? So a corporation can borrow at a cheaper rate than a family restaurant.

An ISA is debt, which is unsecured, inconsistent, and has unknowable revenue streams, usually extended to somebody who has a higher credit risk than can otherwise get debt. And so by the laws of finance, the return profile has to be far higher than if you get somebody who has the ability to pay cash for a service up front.

If you offer an ISA alongside a pay-up-front option, you're going to get incredible adverse selection selection baked into that. The only way we can make it work is if we say like $2,000 up front or $5,000 ISA. The people who will do that are going to be the least likely to be successful.

So I don’t think ISAs work where the alternative is paying cash, because the cash prices will have to be way cheaper. And I don’t think it works where there’s more of an intangible shift. For Lambda School it works because it shifts someone from a lower trajectory to a higher one very clearly. For On Deck, if someone joins having been an engineer at Stripe, it’s much less clear how much of their future trajectory On Deck is responsible for. They will always be a former engineer from Stripe.

Makes sense. The other option is instead of doing an ISA for earnings-power boost, you could take equity in companies?

DB: The number one reason people don't go to YC is they don't need to. In a lot of cases they are experienced second time founders, and capital is ubiquitous. YC is very expensive in terms of equity you give up, and taking equity on sub-market terms is itself a form of adverse selection. If what we want is the very best people to do On Deck, we don’t want to force anybody to give us equity or any other thing that will adversely cause them to self-select out.

ET: On equity, one challenge is that it lop-sides your organization. If you have equity in every company, you are much more incentivized to help the top 1 or 2 companies in your program than the rest. By orders of magnitude. So it’s not straightforward that it really aligns incentives for everyone.

And I reject the premise that we are less incentivized to care about people’s On Deck experience because the moat for us and every other curation business is alumni NPS. That’s all we have. It's a powerful moat when done well, but that's all we have, so we are thoroughly motivated to help fellows beyond their program. Unlike universities, we don't get gov't support. We rely solely on customer love to scale our business.

On the Hiring Flywheel

Last question about hiring from the community. I've seen you guys do this now a few different times. I think that's also a part of your flywheel, right? You bring all these cool people in, they have a great experience, they want to work for On Deck, and then you can take your pick of who to hire. How much of this is a happy accident and how much of this is deliberate?

ET: Very deliberate. At Product Hunt we also hired from within the community a lot. It’s always great to hire from the community, because they're more loyal to it, they just have more context on it, you've gotten time to work with them already. Anytime we prospectively want to hire somebody, we say go through this program, so we can sort of get to know each other a bit better.

In ODF, we saw that people were starting a writing channel, we saw that people were starting an angel investor’s channel. We looked at what people were asking for, and then we built specialized services for it. And then we can hire the person who started it and let them run with it.

Last comment from me: Erik had a great tweet about people taking chances on people in their careers. I feel like you guys just took a chance on me by having this conversation and giving me your blessing to write about On Deck. Thank you very much!


That’s it for this edition of The Flywheel. Thanks so much for reading. Please send your feedback on this new format and let me know who else I should talk to.

If you liked this article, smash that like button and share with a friend! Let me know your take on Twitter here.