Gleb Budman bootstrapped Backblaze for five years with no salary or funding, targeting unlimited backup at $5 a month. When Amazon S3 was too expensive, the team built custom storage hardware from scratch, eventually creating a 45-drive server for one-tenth the market price. That constraint-driven engineering now stores over 100 petabytes and gives Backblaze a lasting edge.
Chapters
So today I'm going to chat with you guys about bootstrapping. We actually bootstrapped Backblaze for about 5 years. We bootstrapped it to millions in revenue and to profitability before ever taking funding. So I'm going to tell you a little bit about some of the things we learned kind of through that process. So going back to where it started, so in 2007, this is Lise. She's an author and a journalist, and she had a MacBook that died. And she called up Brian, who was a friend of hers that helped her with IT, like many of you guys probably do to your friends and family, and she's like, help, you know, my computer died, you gotta help me. And he's like, no problem, we'll get you all set up, we'll get you a new laptop, we'll get you apps installed, where is your backup? To which she starts pounding the table and going, look, what I don't need now is a lecture, what I need is for you to get me my data back. And he's like, I can't help you. If you don't have a backup, there's nothing I can do. And so we started talking about it, and we're like, how many people are actually not backing up their data? And as it turned out, a lot. How many people here back up your data? About half. About half? OK, so statistically speaking, about 50% of people lose data every year. So the half of you that don't back up, I'd highly recommend it. But we asked people at that point, what do you do for backup? And we got a lot of like, eh, I kind of mean to, I'll get around to it and stuff. And we're like, this is crazy. Every photo, every song, every document, everything you do, it's on the computer. And when it dies, it all vaporizes. So we said, this is a big problem. 5 of us together and we said we're going to quit our jobs and we're going to try to solve this somehow. We didn't exactly know how, we just said it was a problem. And so we were talking about how do we want to start this company? Do we want to raise funding the way we did for the previous company? Do we want to take seed, angel, et cetera? And what we decided was that for 1 year we would do nothing. We would not raise any funding, we committed to each other that we wouldn't raise funding and we wouldn't leave. Because obviously another bad-case scenario would be we all quit our jobs and then the next day 2 of the people go, you know what, I'm out. So we said we're going to commit to each other for 1 year, no salary, we're going to put a little bit of money in, that is effectively going to be the seed round for the company. And then we'll try to figure it out and come up with some way of dealing with it. So we went through a variety of iterations, but we decided that online backup made sense. We would just make it really simple. We'd back up your computer into the cloud. You could get it back on your phone or by downloading, or we'll even FedEx you a hard drive if you needed your data back. So that was the idea for the company. It was pretty straightforward. And then we asked people, well, how much space do you need to back up all your stuff? And people said, ah, you know, I've got a 300 megabyte, gigabyte, megabyte hard drive, about that much data. So they're off by a factor of 1,000, right, in terms of how much data they actually have to back up. So we're like, we can't ask people questions they can't answer, because that is a huge friction. So we have to say, it's just unlimited. You don't have to worry about it. And then we asked people, well, how much would you pay for a service that looked kind of like this? It's online backup, we back up your computer, it's unlimited, what would you pay for that? And some people said, well, it's on the internet, it should be free. Everything on the internet is free. Right? And then, so that was one extreme, and then there was the other extreme of people that were like, oh my God, that is so incredibly valuable, I'd pay $50, $100 per month for it, right? So those were the two extremes. But then there was this huge bell curve when we asked people, they're like, you know, if it was like $5 maybe a month, it'd be a no-brainer. So we said, okay, we're going to target that big middle ground where people said $5, I would just do it and we're going to offer it a month. So this was the service. And so we're all sitting there, we're working on this, we're going to build this. And so then came the first big challenge. Which was where do you put all of that data? And so originally we're like, yeah, we'll put it on Amazon S3. This was maybe a year or 2 into when S3 started. So we're like, yeah, we're a group of software people and we'll let them do the heavy lifting of actually storing the data for us, but we'll use them for it. The problem was that for $5 a month, which is what we were going to charge the customer, we could only put 30 gigs of data on Amazon. And we figured most customers would store more than that. So we're like, that's going to be hard to make up in volume. We're going to lose money on every single solitary customer. How do you do that? And so then we were like, well, we'll buy equipment. We'll buy it from EMC, NetApp, Dell. There are a lot of companies that make hardware, so we'll just buy some equipment for them. But all those systems are like $1 million and they're really expensive per terabyte. So if we had raised funding, what we would have done is said, well, we've got $5 million in the bank and so we'll do one of these 2 things for now and then we'll figure out later how to lower the cost on it. We'll just lose money on every customer for some period of time. Since we were bootstrapping, that wasn't an option. Like, we couldn't just go, yeah, sure, well, here's $1 million, well, I've got a credit card here somewhere. It just wasn't an option. So our options with bootstrapping were go out of business or do something else that we could afford to do. So we tried coming up with different ideas, and we were originally thinking about, well, we'll take USB hard drives and we'll plug them in and we'll daisy-chain USB hard drives together and we'll store all of your data on there. And what, maybe FireWire, something. And we ended up coming up with this. We had our own server design for how to store data really, really cost efficiently. By the way, it's a little hard to tell from the picture, but this is a plywood server. What about— So the first server design— To prototype quickly, we prototyped it out of plywood. And we actually deployed this in the data center. So we needed to make sure that it fit. We needed to make sure that it worked, that airflow went through it. So this plywood server, we kind of had to close our eyes to the part in the data center which said nothing flammable in the data center. Plywood's mildly flammable. But we deployed this server in the data center and tried it, and it worked. And so this might be a path. And so then we iterated with a bunch of different components. And we had some cases cut for us in units of 1. This wasn't some mass production run. Again, we still didn't have millions of dollars to spend. This was all done on a complete shoestring budget. And in the end, we came up with this design, which is a 45-hard-drive storage server for about 1/10 the cost that anyone would sell it to us. And so today we've got over 100 petabytes of data that we store on what used to be a plywood box. And today this is actually a huge competitive advantage for us because it's dramatically lower than anybody else's storage system. And the only reason we have this is because we had no other option. Because we were bootstrapping, we had to come up with a lower-cost option or we were simply going to go out of business. If we had raised funding, we would have certainly gone on Amazon S3 or EMC or something and we would have tied ourselves to those more expensive platforms and it would have actually been very, very difficult to extricate ourselves back off of them. Like we would have never— once you're at scale and you've raised millions of dollars and everything and you've committed yourself to some platform, we've never gone back and said, oh, maybe we'll try a plywood box and see if that works. So bootstrapping in a way actually forced us into a path of success that we wouldn't have been able to go down.
