Founder of the Week #1: Daniel Chang (Belay)

During the “Founder of the Week” series, I interview startup founders about their experiences building their companies. Today, I talk with Daniel Chang, a co-founder of Belay. Belay provides support services to gig workers (including rideshare and delivery drivers).

To start with, can you tell me a little bit about what Belay does and what need it fills, for someone who’s unfamiliar with the company?

Absolutely. Belay has gone through a series of pivots. I can start with the original idea and how it’s evolved. The mother of my cofounder, Kai, was a massage therapist and some of his greatest hardships were due to the inconsistency in his mother’s income. We had the idea that if a gig worker’s income has peaks and troughs, we’ll pay them during the troughs and take a cut during the peaks. The first place we started was Uber drivers because the availability of data was greatest there. And we were able to reverse-engineer Uber’s APIs and scrape data such that we could predict the future income of Uber drivers to provide a guaranteed hourly wage on a weekly basis. So that’s the first iteration of the product.

Our second iteration of the product was kind of the same idea, but at a larger timescale. We found out that Uber and Lyft drivers get deactivated a bunch, and get injured a bunch as well. Taxi and delivery drivers are both in the top 10 most dangerous jobs in the US, and we realized this was the area that hurt the most, as opposed to the week-to-week income stabilization.

That’s really interesting. You mentioned the injuries were more impactful than the lack of constant income, how did you come to that realization? Through user research?

You always have to talk to users — that’s basically how we got our first direction. Also, we realized this through logic — say that on average, someone makes $25 per hour, but it varies between $20 and $30. This might be fairly impactful, but not as impactful as someone losing their income for two months because they got deactivated on Uber.

The other aspect of that was that we found that between people with the same average hourly rate, the ones with more volatile hourly incomes were the ones who drove Uber less. They would get on Uber and drive for 2–3 months because there was a gap in their employment, and would then find either another form of gig work or another full-time job. They would essentially use it as a stopgap between jobs. As you can imagine, that’s an issue for us, if the main audience our product appeals to is only able to be serviced for two to three months.

The injury and deactivation resonated more broadly across the platform, so it was a change based on talking to users.

Thank you. You mentioned your cofounders, and I was curious how you first met them.

My co-founder, Kai, I met online. I didn’t think I would meet my cofounder online, but we met on a platform called Ladder, I think now it’s been rebranded to Nova. So we met in a fairly uninteresting way of doing it. We soon started bringing on other people who were essentially cofounders, who’d been there since Day 1. Those were people who I’d met throughout my entire college career — people I met on the first day at campus, people who I met through internships.

When you were evaluating them, what were some of the qualities you were looking for as you brought people on?

So I’m a technical co-founder, so I was looking at someone who’s good at business, as well as someone who’s good at selling the mission. During my intro call with Kai, he was a very convincing person, which transfers to when you’re talking to businesses or industries. Besides that, just engineering talent. I just look for the best people in my network that were available. I had to convince some for a few months to get them off their current project or to extend their workload.

That makes sense. On a different topic, founding and running a startup is obviously a great deal of work. So what motivates you to put in that work, both initially and on a day-to-day basis?

So I can talk for myself, and also about a few patterns I’ve seen in general. For me, I would say it’s two things. One, the initial work is interesting to me. It was a large of large-scale data engineering to be built out into a pretty large platform. And not only is it just data, the platform also tries to dodge stuff like rate-limiting and token expiration. So there was a lot of work put in there to build this scalable and robust data platform. There was also the time series and machine learning aspect, that I tend to be interested in.

The other aspect of it is working with my friends; when we bleed and sweat together. It’s an interesting thing to do when you’re young, and you don’t have as many obligations.

The last thing that I’ve seen motivate people is the actual mission. So Kai, of course, grew up with this issue and that motivates him to try to change the world.

That’s really cool to hear. So you’ve been working on this product for around a year now — what’s been your largest success in that year? What are you most proud of, so far?

I feel like we navigated our pivot pretty well. In certain projects that I’ve worked on in the past, you really need to have momentum. If people leave, get demotivated, or things get broken off, it makes it really hard to get back on track. So I think that’s one main thing we executed very well.

Right now, I’m less proud of fundraising, although it has less correlation to the eventual success of the business compared to any other part of the business. We were able to raise both from angel investors and institutional investors. We received a terms sheet from a private equity company, but they wanted 60% of the company. We eventually said no, for a variety of reasons, including wanting to maintain more than 40% control.

I was going to ask about that — what was your general fundraising strategy? What were you looking for, and what was your philosophy around fundraising?

I would approach fundraising in a very different way now. The way I would approach it now is to first get a very solid business model, solid economics, and solid product market fit.

I’m able to raise small checks of around $10k to keep the lights on. We had a masterclass on fundraising through the Village Global Accelerator, and the operating partners were able to walk us through and connect us to an entire pipeline of investors.

Could you talk a little bit more about how the Accelerator or other fellowships you’ve done have helped with getting your startup off the ground?

I would say the most important one is the Village Global Accelerator; it has since been rebranded into the OnDeck Accelerator. We applied to a few different accelerators, including YCombinator, but we went with Village Global because we liked the people we talked to throughout the interview process and we thought we’d get more one-on-one mentorship.

Since then, I’ve done a few fellowships: 406 Ventures, ZFellows, and the Inspired Capital Fellowship. There are a variety of entrepreneur/operator fellowships that I’m eligible for because I’m still a student. These help us expand our network, raise capital, and do research into generative AI applications (prior to it being cool). That’s something I’m developing a thesis on.

On the opposite end, what’s the biggest failure or challenge you’ve had to navigate so far?

If our biggest success is navigating the pivot, our biggest failure is not pivoting soon enough. We had gotten pretty attached to making the product, so we didn’t pivot soon enough.

You also mentioned the fact that you’re still a student. How do you think founding a company while still in University impacted your journey?

When I founded my startup, I basically took the entire year off. It’s difficult to dedicate a lot of time to something when you have something else in mind. In my senior year, I had a lot of “float”, so I was able to take some time off, and if I did have to take classes, it would be half of what a student normally takes. If anyone wants to do it, and they didn’t have this amount of “float”, it would be very tough, but there are signals you can use towards product market fit. Once you start gaining traction, you can drop out or take time off with a greater sense of conviction.

Do you think it’s easier to find resources if you still have that University network?

I don’t think I relied on my Northwestern University network that much, outside of people that I know. But when you are in University, there are a bunch of programs that are available to you: Pear Garage, Neo, Unusual Ventures, and Floodgate. And, of course, you can apply to YC or OnDeck.

There are a lot of programs that are open to college entrepreneurs. And I would say that’s a pretty recent phenomenon, I’ve seen like a 20% increase for the past year, and that trend has been going strong. I would say it really depends on the economy, as well, for example, OnDeck had some financial issues. But now is certainly not a bad time.

Speaking of the economy, how has the recent tech downturn impacted you or the startup space?

Generally, fundraising is down and valuations are down, from what I hear. We didn’t try to fundraise during the worst of it, which I would say is now and a few months ago as well. From what I hear in the Village Global network, they are seeing some depressions in deal flow and valuations. So if startups are currently well-financed, they won’t be looking for financing now because it’s not a good time.

On the contrary, if you are developing anything in the Generative AI space, that’s quite hot, regardless of the economy. The sectors that are being hit hard are areas that are more capital-intensive, like legal tech.

Was founding a startup something you always knew you wanted to do?
Not really, when I was a freshman I wanted to work at Jane Street. I did go throughout college working on small projects and realized that was an area that (1) I’d be good at and (2) that I would be able to leverage my existing network on. If you’re looking for a pure, intellectual challenge, then, as you know, quant is a great place to work with smart coworkers and interesting work. But if you’re going more for breadth — if you want to think about product, think about business, in addition to all the similar technical challenges — then startups is a very interesting space.

And do you think working on projects and startups has increased your knowledge of how businesses have run?

It’s a crash course for sure. I don’t know in what situation you wouldn’t learn about business when doing startups. You could be the most academic researcher, doing the most academic research, and sometime in the first year, you’ll face a challenge that can’t be engineered out of.

We sort of touched on this already, but in general, what advice would you give to someone thinking about starting a company?

Even prior to formation, think about choosing your cofounder very, very carefully. Have it be someone you’ve worked with for a very long time; it will make life a lot easier.

Second, the thing you should be most focused on is not engineering, product, fundraising, or anything like that. It should be working towards product-market fit, and having the common metrics and signals that people want what you’re trying to build.

What would you say that the most important signals for product-market fit have been for you?

You can’t lie to yourself about how the market reacts to what you’re putting out there. If you are developing a consumer product, you should see organic growth. You can jump-start that with a $5 or $10 incentive, but you can’t lie to yourself.

On the B2B side, you probably won’t see organic growth, but you should have a relatively easy time communicating to your ICP (ideal customer profile) what value you’re trying to provide. Most likely you should be able to get it with one or a few tries. If you’re trying to communicate a problem to your customer that they don’t have, that would be a very bad sign.

We touched on this earlier, but besides the rideshare space, what’s another area that you think has a lot of room for new innovations?

In terms of infrastructure that enables applications, which is a pretty common framework to apply, generative AI is at a very exciting place right now. I’m currently building products for clients that leverage it. You can look at the capabilities or go on the playground to try things out, and realize it will have a very large impact in the future. There’s a startup called Jasper is a very light wrapper over GPT3 for copywriting that’s doing very well and growing very fast.

A fun question — how did you decide on your company name?

Kai came up with it — the Belay is the thing that catches you when you fall when you go rock climbing.

What’s one fun memory that you’ve made so far with your co-founders?

I think it’s always fun and interesting when you “Wow” your non-technical co-founders with your technical ability. I was able to reverse-engineer the Uber API and print out a bunch of summary statistics, and they were pretty impressed with how that was even possible. I like being creative when it comes to datasets, either re-purposing public datasets or reverse-engineering APIs. It’s sort of like a growth/data hack if you’re creative with what methods you’re using.

What’s your least favorite city to drive in?

San Francisco, because of traffic and how aggressively my friends drive when I’m in the car.

Reach out on LinkedIn if you have feedback, questions you want answered, or if you’re a founder that wants to be interviewed.