Q&A About a Hardware/Data Startup | Backblaze's Gleb Budman

Other ways to do that. Yeah, so the question is, this idea of buying these external hard drives that aren't too much more expensive, maybe that was only going to be a short-term solution and it wasn't going to continue going for the long term. And we were absolutely concerned about that. So when we were faced with this hard drive crisis, we brainstormed a variety of different paths, right? And we said, one option is, raise prices on customers. Like, if we have to do that to stay in business, that's better than going out of business. We went through kind of a variety of different scenarios. One scenario that we actually started considering is, should we raise funding in order to have cash in the bank as a buffer? So we brainstormed a variety of different scenarios, but we started executing down the path of the one that we thought was best, which was farming these hard drives. As it turned out, we were able to do this long enough to where the internal drives started coming down in price and becoming available again, but it was a few months before we saw any of that, and it was a year and a half before they kind of came down to where they should have been. But certainly, no one knew for sure what that future looked like, so we brainstormed a number of options, picked the best one, executed against that, and then had these other options as kind of backup plans. I was wondering what's your concern for raising the funding at that time? What was our concern for what? Yeah, so what's your worries and concerns about raising some money from VCs? Oh, what's our concern about raising money from VCs? So it's not just about that time. In general, there is a question of do you want to raise funding or do you want to bootstrap your company and there are advantages to both. So some of the things that we didn't— some of the reasons we didn't want to raise funding. One is raising funding takes time, like months of focus on going out and raising funding. As a startup team, you have 2 people, 3 people, 5 people. You take a couple of those people off full-time to do nothing but raise funding, you're not spending time building your business. So it's incredibly distracting from that perspective. If you successfully raise funding, now you've got a board and it's an external board that has somewhat different objectives than you do. So now you have to spend time managing that board. At our previous company, at Mail Frontier, we would spend one week leading up to every board meeting, which was once every quarter, of all the executives' time preparing for the board meeting. And then there are these other things which are unrelated to the VCs themselves, but when you're building a company, if there's someone else that just hands you a few million dollars of cash and that's how you start, then the culture inside the company becomes one where you expect money to kind of fall from the sky, not to come from customers. One of our VCs at our previous company said, The problem is that people start to look at the VC as the customer. It's like, what can I do to get more money from the VCs? What can I do? Which VCs might be the best ones to get money from? And it's like, no. A business gets money from customers. It might temporarily get money from VCs to survive, but if your whole culture is focused on the VCs as the place money comes from, then you've built the wrong kind of perspective, and it's hard not to do that if that's where your lifeline comes from. Other questions I can answer? Yeah. I just wanted to say thanks for doing all those studies on your blog, putting up the data on reliability for hard drives. Sure. That's really useful and interesting to read, in addition to being a fan of your service, but are you going to keep doing those? Are you going to keep doing those? The studies? Yeah, so the question of are we going to keep doing these studies. So one of the things that we've done is we've got about 40,000 hard drives in our data centers. And so we know a little bit about which drives work, which drives don't work, what the failure rates are. And so we've been plotting which drives die at what rates on our blog. So we absolutely plan on continuing to do those. We actually— I don't know if you might have seen, we just recently released the raw data, so you can actually download literally every single hard drive's data that we have, when they failed, why they failed, all their smart stats, et cetera. And yes, we absolutely— our plan is to continue doing that, and we're hopefully going to expand the amount of information we share on that. Cool. Other questions about bootstrapping, running a company? Yeah. Yeah? Do you actually see iCloud and Dropbox as your competitors? Do we see iCloud and Dropbox as competitors? It's funny, they offer a service that's different, but in the mind of the end user, they sometimes are viewed the same. So what's that? iCloud the same. Yeah, so it's a really interesting thing because from a competitive set in terms of like if you do feature function kind of comparisons, They're very different products, but from the perspective of a customer trying to decide what to use, it's confusing, and so sometimes they'll say, oh no, I have iCloud, so I'm fine, and so they effectively are a competitor even though we're offering different products. The funny thing is, so I personally, I use Backblaze, obviously, to back up all my data. I use iCloud, because it's the best way to get stuff from my phone into the cloud and onto my computer. I use Dropbox because it's a great way of sharing some files with some people. I use Google Drive because we use Google Docs internally, and so it's a good way of using— of sharing the various different kind of spreadsheets and things like that that are on there. So I actually use, you know, kind of— a number of these different services because they each have different purposes. So a common misconception is that iCloud backs up your Mac. It actually doesn't, right? Like your Word files, your PowerPoints, your spreadsheets, like almost nothing on your computer is backed up with iCloud. But it's a common misconception that it is. What Backblaze does is it takes all the data off of your computer all the data off of any external hard drives and is constantly backing all of it up. And so they really are different services, but from a competitive perspective, sometimes they're confused in the market. I just want to add one thing to that, because as a customer, maybe we're just thinking back up some of our data, but not all the data is necessarily needing to be backed up. And also, I'm thinking since For me, one service charging for money, one service not charging for money makes a huge difference to me. So I personally would prefer maybe using Dropbox or iCloud. So I was wondering, what do you think is the biggest differentiator for you? So I think the thing is, with iCloud, you can't actually back up any of the data on your computer, almost any of the data on your computer. So it's the kind of thing where it's just a different service to do a different thing. So if you care about backing up data on your computer, like your Word files, your PowerPoints, your spreadsheets, your code, your modeling, any of that stuff, iCloud doesn't help you with that. It won't go up there. You can use Dropbox for some of those things, but you have to kind of pick and choose individual files and remember to put them in there. Backblaze just makes sure it's all safe and not have to worry about it. Yeah, go ahead. Just as Chris said, customers are sometimes confused about Dropbox and your product, and do you have some very simplified message just convey to the customer that, oh, you must use our product? Yeah, it's actually complicated. I actually wish that it were really simple, and I get a lot of these in conversations with people. A lot of times what I hear is, yeah, yeah, I've got that cloud stuff, so I think I'm good. And so it's like, that cloud stuff. It's like, there's so many different things that use the term cloud. It's like, there's a lot of confusion out there. The simplest thing that I can say is, we're the only company that completely backs up all of your data, completely unlimited, for $5 a month. Everything else you have to pick and choose and remember and figure things out. So if you forget something, to put something in there, and your computer crashes, it's going to get lost. With Backblaze, you don't have to worry about it. It's just going to be there. Yeah? When would you say, like, would you say you should always bootstrap for as long as you can, or when would you say there's an indicator that, hey, maybe you've been bootstrapping, but you should start to raise funding, or, hey, your company's this way, so maybe you should just start raising funding? As long as you can. So the question is, should you always bootstrap or as long as you can, or should you actually raise funding sometimes? It's a good question and it varies by scenario. So I'm a big fan of bootstrapping, but it's not to say that it's the right thing for every company or every person. Let me give you a couple examples of when I would definitely raise funding for. If you're in a market that is going to move very, very quickly and it's winner takes all, bootstrapping is probably a bad option because it's hard to get to scale fastest by bootstrapping, right? So in some markets, social networking is potentially an example of that, where having scale matters a lot, it's really hard to bootstrap and get to the scale first. Other scenarios are markets that are incredibly capital intensive. So ours is a little bit of a weird scenario because it's actually a pretty capital intensive business, but if your business relies on building out massive data centers or buying up tons of land or building spaceships or something, things that require huge huge amounts of capital to simply execute on it, it's hard to make a lot of progress without any capital. So there are these certain scenarios where you should just raise funding, try to get as much money as you can in the door as early as you can, and then pursue that vision. For lots and lots of other businesses, whether you are second or third or 27th is not the issue. It's whether you've done a better job at it, it's whether you understand the customer better, it's whether you've executed better, and you don't necessarily need to raise the big funding. If that's the case, bootstrapping is great because I remember about 6 months into starting Backblaze, I had this kind of like that weird feeling in your tummy where you're like something is wrong, like something— I'm not doing something, I'm like forgetting something. And what I realized was what was bothering me was I hadn't created a single PowerPoint presentation. Like, we had been at this for 6 months and I hadn't opened PowerPoint. Whereas in the previous company where we were raising funding from the beginning, it's like every single day I was spending half my day in PowerPoint because I was creating a PowerPoint presentation and then we would go pitch some VCs and then like I'd come back and I had gotten feedback from something in the presentation and then I I'd spend half my day rejiggering the presentation, then we'd go pitch some more VCs and it was like I was spending all my time on PowerPoint pitching. And here I was spending day in and day out focused on customers and product and kind of building out the business. So bootstrapping has a lot of these other advantages, but there are definitely scenarios where it's not the right answer. Cool, thank you. How often does the average customer recover his data from their— From Backblaze? Yeah. How often does the average customer recover data from Backblaze? So what we found is that 50% of customers recover data once per year. So it's not like nonstop, right? It's not like you're constantly losing data all the time. But it's more often than we thought. Why is it? So when we started the company, like many of you, we prioritized what we thought was the most important first, and the most important was backup. So we finished the backup application and how it would store data. We hadn't built the restore functionality yet, and we joked that we could probably just launch the company because no one's going to need to restore data for like for a year or something. I mean, it's going to be a long time before someone's computer crashes and they actually need a restore. We should just launch the company with backup only. Okay, it was a joke, but it was— that was kind of what we were thinking. And the reality of it was, day one, people had installed the product in the morning, and by the afternoon, someone had lost data and used us to recover the amount of files that they had backed up that day. So it does happen, I think more frequently than we expected, in part because a hard drive crash is far from the only reason people lose data. They lose data because their computer is left in a taxicab. They lose data because their house was flooded. We had a guy in New York that told us about that case. They lose data because the computer is stolen. stolen. They lose data because they hit save instead of save as on a Photoshop file and it flattens all the layers. You know, like, there are just lots of ways people lose data.