March 23, 2021
This week we’re having tea with Sahil Lavingia, the founder & CEO at Gumroad, an online platform that enables creators to sell products directly to consumers. Prior to Gumroad, Sahil was the second employee at Pinterest. He recently launched a $5m rolling fund using Twitter and a Notion memo; his limited partners include Arlan Hamilton, Tim Ferris, and Naval Ravikant.
This interview has been lightly edited for length and clarity.
Should CEOs be running funds on the side?
Is there a base level qualification you need before you’re ready to start a rolling fund?
What is a syndicate? Do they work alongside rolling funds?
How does Sahil think about warm intros vs cold outreach?
In a traditional fund, you have to raise capital and then you deploy the capital. Then if you want to raise more money effectively, you start a new fund. Traditional VC funds will have fund one, fund two, etc.
With a rolling fund, they're both happening at the same time. You're raising money, you're deploying capital quarterly. You're taking a one-off event and you're turning into a rolling subscription model.
It's pretty new. You can go on the internet and say, "Hey, I have a fund. I'm raising money for it," which you couldn't do before, to protect unsophisticated investors from getting scammed effectively. You still had to be accredited. There are still these qualifications, but even before, you weren't able to tell anybody that you were raising a fund.
The world was quite different for early-stage investing not very long ago. Doing a deal costs like $30k. Software has made it so easy that I can do a deal literally in two minutes. It's meaningful now because you can effectively do this at scale, by lowering the barrier to entry.
I didn't want to be a VC because when someone said VC, I had an image in my head. It included this traditional process and meeting people I didn't know. These traditional LPs are typically university endowments, family offices, pension funds, that require networking. It would take me 6 to 18 months to do that. Doing all of that stuff I described, while you're running a company, would be quite difficult. Whereas with rolling funds and with the 506(c) rules, I can massively scale that because I can tweet about it.
So much of my LP base came to me because I just tweeted about it. It scales well. I think it was cognitive friction of, “Oh, I can just start at $100,000 a quarter, do one deal a quarter.” That's already going to let me invest more than I would have been able to do personally.
It’s not taking away anything from me running the company the way I am. I feel like I'm a better CEO. In the past nine years of running Gumroad, some of the worst periods of my time as CEO was working all the time on Gumroad. Basically, you do productive work, and then you finish it, and you still have all this time that you want to feel you're contributing to Gumroad. So, you just pester people. It should be like, "We're good. Come back tomorrow. You've worked enough."
One of the unique things that I was able to do, compared to other CEOs, was test the waters. My investors might not be pleased with the outcome exactly, but I was very upfront with what happened there. A lot of them are LPs in this fund. I was just able to be like, "Look, this is what I'm doing." I don't have to pretend that I have this fund. CEOs have been doing this for a long time. This is the world now and people appreciate it. The truth is, founders will want founders on the cap table.
The number one benefit is that you have more empathy than anybody else would. I'm not saying that non-founders are bad. I'm not saying founders are better than anybody else. But if you want someone who's going to understand what you're going through, someone who's gone through, founders are going to be able to do that better than anybody else. There are certain things that I know I would not have been able to understand if I had not started a company; for example, hiring being so difficult. You learn over time that hiring is maybe the hardest thing about building a company.
Probably not. Founders care about who they're going to be spending time with, who's on their cap table, and who's going to be helping them. Ultimately, it doesn’t matter too much what the legal structure of the deal is.
Something that does matter is if I can tell a founder “this is my exposure to the deal.” If I write a $100,000 check from my fund, it ends up being something 35%. That's the equivalent of me personally having invested roughly $35,000, which by the way, I would not be able to do. I don't have that much capital.
Not knowing that is a little bit weird and I suspect that over time, transparency in the venture world will increase. However, that has nothing necessarily to do with rolling funds. That can be done for traditional funds as well.
The biggest difference is you can temper your deployment. If you want access to my deal flow, or you feel I'm going to be a good investor and you want a part of that, you can commit. It can be $25,000 a year, $50,000 a year, $100,000 a year. You should think of it as an Evergreen deployment.
If you want to think about it as a traditional fund, you can also do that. You can do 100K spread over eight quarters, like a traditional fund would typically be deployed, and then stop. Then there'll be somebody else there to take your place. It's a conceptual change. This is the new model, but for new people, they love it because they're like, "Yeah, this is how I think about investing in stocks. This is how I think about my mortgage." It's very common when you have these big purchases that you spread out over time.
Anyone that feels like they should, should go through the process. Even if you're not thinking about doing a rolling fund today, I would recommend everybody start thinking about starting. It can be tiny checks, like $5,000, $10,000, even less than that, in people that you believe in, just to start building that mental model and getting some of that skin in the game to show investors that you can get into deals. You'll learn more from investing by investing, than basically anything else.
If you've made a few angel investments, you have some sort of track record. It takes time to build that portfolio.
One of the nice things about me, in terms of how LPs look at me, is they know what they're getting. They can go on Twitter. I have a carbon footprint on the internet. There's tons of stuff about me. You can go listen to a podcast when I was 18. The number one LPs look out for, and the same thing when investors look at companies, is “is this person legit? Is this person a fraud?”
It's pretty clear for most people that I'm probably going to be in this game until I die. I want everyone to think about this as a long-term thing. I want to do this investing thing for 20, 30, 40 years. Establishing proof that those are your intentions would be helpful.
Your early backers are typically going to be people who know you, believe in you, and have worked with you potentially before. That's always the case, even if you do a Kickstarter campaign or start a company. Then you grow and can take that proof and be like, "Hey, all these great people invested in me, who are sophisticated and understand this stuff."
The Twitter audience does help. Ultimately, the stuff that I think I could add value-wise to the founders is the stuff I would have been able to do without a Twitter account. It's my experience working with Gumroad and all of these other things that I can tweet these pithy statements that go viral.
But that's not that beneficial to the founders. What is beneficial is I might get deal flow. I might get founders coming to me first because I'm top of mind, because they see my tweets. People talk about how VC is a commodity and it has become a content marketing game. If that's true, I can play that game and do quite well.
They take care of almost everything. They have that page that people can go sign up directly on if they want to become an LP in the fund. They can have some information about the fund and the terms, and all of the fine print. They take care of doing all of the wiring and all of the signing.
In terms of what I have to do, some basic tax stuff for the fund. I'll have to get a lawyer on retainer to help me think through stuff over time. If I want to throw events, I'll have to do that myself. Any financial-related stuff, I do myself. They handle the vast majority of the back office stuff.
There is flexibility. Let's say you've spent half the capital, it gets rolled over into the next quarter. Trying to deal with that with human beings would not be fun. It would be dangerous because that's some complicated auditing stuff. One great thing about AngelList is they'll deal with all of that for you and figure out all the distributions correctly. That just gets rolled over as if you had just deployed it in the next quarter, and LPs in that last quarter will just get a little bit of exposure into the next quarter.
AngelList has to approve all the deals. When you do an investment, you fill out a little form and it says what's the company, what are the terms roughly, have you received a financial gift, etc. So far, everything that I've asked for to invest in, they've said yes. I don't think they think about it strategically. It's mostly a legal thing.
Rolling funds are syndicates on a quarterly basis. Ultimately, as a founder, I don't like working with syndicates. I want people with skin in the game, with direct exposure, who've invested in the company directly. People do syndicates because they don't have any personal money to invest, so they're able to utilize their network. Rolling funds are basically a blind pool syndicate. It's better for the GP. You can move faster. It's better for the founders.
I had the call with Naval mid-June. He said “Put together a memo, send it to a bunch of people, see how much interest you get.” Then I had a call with AngelList. They started setting that up. It took two to three weeks. My guess is it's probably faster now. I've heard it can be much faster, 48 hours. Then I tweeted about it a few times, did a webinar, and made some Notion docs. Anytime someone hits me with something, I just say, "Go read this Notion doc, and then if you want, here's a Calendly link, if you want to talk about it." The majority of people just sign up directly.
I want to back things that I feel proud about talking about. For example, I really believe in rolling funds being a great product, so I can go out and feel comfortable talking about it. In general, I like when really hard problems are solved elegantly with software that I'm excited about using.
Typically, that means the founders are technical. It doesn't mean they need degrees, but they need to be proficient at building stuff. I don't care about pre-revenue, post-revenue stuff like that. If you have metrics, I'll want to see them, but they're not necessarily.
The one thing I get a lot of is ideas, and I always struggle with ideas, because for me, it's not that much work to go from an idea to a prototype. It’s always great to have an answer to why hasn't this already been built by somebody else? Why now? Why you? These are important questions to have answers to.
I do both. I'm not going to say no warm intros, but I don't think there's an inherent advantage to it. If anyone emailed me totally cold, I'm going to take you just as seriously, because that's my job as an investor: to find those diamonds in the rough.
Time diversification comes up a lot, because if you only invested in 1999 to 2000, you might've lost all your money. You don't want to do that. You want to spread it out longer term. Have you tried timing the stock market this year? You probably did worse. I always tell my LPs, "Look, it's a four quarter minimum, so I want you in for at least a year, but you should think about it longer than that."
I actually think the rolling fund will actually incentivize me to be better, because instead of doing 30 deals on demo day, I can only do a few because I have a certain amount of capital to deploy right now. It's better to have that discipline: I can do X amount of deals versus I have this much money to spend over three years, and then within 18 months you deploy it all.
I believe the answer is no. Technically, if you're not a GP in your own fund, you don't have to be accredited. You can be an unaccredited investor and actually become accredited through that process.
I think it would count.
AngelList just made the fees way lower. It's 1.5% per quarter forever. That's just a one-off cost. Let's say I have a million dollars. It's going to be $15,000. And there's a minimum of 20K for the four, except that's forgoed for the first two quarters while you're getting spun up. So, it's actually incredibly affordable. A traditional fund is much more expensive. Sometimes it doesn't seem as much because it's free. But then you realize all these admin costs add up hourly over time.
My guess is that AngelList will have competition. They certainly have a branding head start, but AngelList has had a lock on early stage marketplace connecting, LPs and syndicates, and startups for a long time, 10 years or more. If those people are good, then it's good. If they're not good, then it's not good. If AngelList said tomorrow, "Our fees are now 30 times higher," I'd be screwed. The beauty of this thing is, as AngelList has said, it's just a legal structure and some software. You can always leave.
In terms of where the future goes, I'm making a bet on AngelList. My guess is that rolling funds will enable all sorts of interesting, new stuff. You'll have much more liquidity, and when you have liquidity, you have opportunities to build more software for that group of people. Even just stuff like the 99 LP limit, like what if that goes away? If I went on Twitter without a 99 LP limit, I would have launched with a $10, $20 million a year fund, which is like going from a cold start to having virtually one of the largest early stage funds on the planet, with an incredibly diversified LP base. It would be game-changing.There's a reason those rules are in place, but my guess is just like the 506(c) is new because of the internet, it's like the new use cases are going to enable the regulation to support them, and that's going to enable more new use cases and there'll be this flywheel effect.
What is the problem? What is the solution? Why are you building this? What's the origin story here? Why now? Why hasn't this been built before? Why are you, specifically, going to solve this problem? What have you struggled with? What have you learned through this process?
Teach me. I almost always say yes to emails that teach me something. The emails that are the worst are like, "Hey, I have a startup. We're raising money. Click here to see the deck."
You can also send us a note at [email protected] We’d love to hear what you thought of the episode, or who you’d like us to have tea with next.
The Mercury Team