Lessons in scaling a consumer hardware company. My conversation with Boris Sofman, CEO @ Anki
From my years of living and working in the Bay Area, I was quick to realize the strong sense of community within the startup ecosystem. Part of this, is the the willingness of entrepreneurs and investors to share their experiences and help guide their peers. I love this attitude and was lucky enough to experience this recently when I sat down with one of SVB’s clients, Boris Sofman PH.D, the Co-Founder and CEO of Anki, Inc. We discussed his experiences of running a consumer hardware company and his expectations for the future. In this blog post, I have summarized some of the highlights from our discussion.
What is your advice for entrepreneurs who are trying to sell their products outside of Silicon Valley, and get visibility and sales across the U.S.?
In a lot of ways it’s like crossing a chasm. All these companies, whether it’s through Kickstarter or some other launch, actually find it pretty easy to get early adopters at some scale and get that PR out of the gate, but it’s also easy to get seduced by that early press and not realize that it has a really quick tail on it. If you don’t follow it up with a meaningful investment and a thoughtful strategy on how you’re going to raise awareness and establish yourself in the market, you can actually get into a lot of trouble. There is a graveyard of companies that have that nice pop out of the gate, but that really have trouble scaling afterwards. It took far more time and resources than we expected to actually build that awareness season after season and build a reputation and meaningful ability inside the company to be able to sell in a way that’s scalable versus optimizing for the first launch.
Given the seasonality of your business, forecasting inventory build vs demand can be challenging. How have you guided Anki through that, balancing motivations of investors to grow quickly but also not wanting to crater the company by building excess inventory?
We’ve been on both ends of it. We’ve massively underestimated demand, and we’ve also over forecasted and had to figure out how to deal with it on the tail end. The challenge doesn’t stop with predicting orders from retailers because you can sell a lot of product into the channel, but if you don’t actually sell through, you have a headache on your hands. We’ve done a few things that have helped our forecasting a lot over the past few seasons, partially based on the learnings of the past. We’re far more data driven, we spend a lot of time studying trends in premium price products in our category, we look at trends from previous years, and look at sales we saw for earlier products or earlier seasons that are indicative of what we might see today. As for advice to earlier stage companies — there is far too much emphasis on high revenue numbers. At the end of the day, none of the numbers are that meaningful in your first year. You’ll be spending far more on just running the company than you’ll ever get in revenue. All these things will improve. It’s far more important to launch a product, show that there’s demand for it, show that customers are happy with it, and use what you learn to quickly iterate year after year, than it is to take a bigger risk on the upside where that swing isn’t even that significant in the big picture, but you potentially leave yourself with a bunch of challenges that will hold you back. Not being able to refresh the hardware is a constraint that you might have to deal with.
Do you think investors have gotten smarter when evaluating consumer hardware investments?
I think so. When we first started fundraising in 2012, hardware wasn’t sexy. It was the year of “Social Gaming”. Zynga IPO’d at the end of 2011, so everything was about social gaming — no one wanted to do hardware. That was a different phase. When you get to 2013 and 2014, Nest sold to Google, there was all this excitement about hardware, Internet of Things, connected devices, etc. Suddenly it swung in the other direction and every company that had a Kickstarter that raised a million or two dollars was raising these huge Series A’s. I think there have been enough challenges in going from that to getting to the next level, where now investors are more skeptical, and showing proof points beyond the early adopters into slightly broader levels of adoption, and they’re looking for signs that there’s a deeper fundamental business under the hood. You can’t just make a product that sells a lot of units and raises your revenue if there’s no fundamentally good business model under the hood. And so it feels like investors have started asking better questions on that front over the last few years.
Over the past 12 months, we’ve seen a number of big names in the consumer electronics space fail. Why is scaling a consumer hardware company so difficult to get right?
We’re certainly not out of the woods. That’s one of the other surprises. Every year I remember thinking “this is so tough, it will be so much easier if we can just get through this year — it’s going to be a lot better next year. That’ll be the inflection point, and then we’re off to the races”. But every single year the stakes just get higher and it still feels like life or death. It’s like poker, where it’s like “oh you won, well now you have to double up again”. The nice thing is that it’s less random, we feel like we have more control of our destiny, and when you diversify with more products and more countries, you have more opportunities to make up for a surprise downside in one place which can be offset by a surprise upside in another. The mindset we try to take is that we’re inherently optimistic, we’re inherently aggressive, but always try to think about both upside and downside and we never do things that will put the company at risk. Even if we think that there’s really strong evidence that we could have a great season, we think about a reasonable level of surprises that we might encounter and make sure that we’re positioned to be ok if that happens. We’re not happy about it but it won’t kill the company. Part of that forces comprises where, you launch a complicated new product, you’re aggressive but you don’t swing for a grand slam, and if that means you run out of stock that’s a better problem to have than the other way around.
Thinking about the future, as you think of Anki and robotics in general, what is the home going to look like in 5–10 years and how will robots going to play a role there?
There’s a few dimensions to this. There’s a huge arms race on voice interface with Alexa, Google Home, etc., and that’s important because that’s the one really important interface that will be the digital hub for all of these other elements that are digitally accessible, whether it’s your calendar, external events, data, or some control of things in your home … That’s why a product like Cosmo, even though entertainment is the underlying root for that product, the ability to start to develop character and personality in a way that feels comfortable and feels intuitive, that becomes a tool for extensions down the road, because at some point, when things in your home become intelligent, if the creators have zero thought towards personality or character, it’s going to become unnatural, or even scary because you have something that physically moves and has sensors and capabilities, and there’s no human element to it whatsoever. In some sense it also becomes a handicap from a functional standpoint because the moment something has some degree of character or personality to it, people become more forgiving in any limitations of its abilities. That becomes a way to give yourself a wider margin of error and start having an impact before you’re 100% bullet proof in terms of your ability to function.
I would like to thank Boris for taking the time to talk to me. If you work for a Robotics company, or any startup working with Frontier Technologies, please feel free to contact me at mtrotter@svb.com. I would love to get your perspective and see how we can help.
