On Particle, IoT, and solving hard problems

Particle is an IoT platform. That’s a very jargon-y way to describe us, and it resonates well in the tech-forward Bay Area. But a better way to describe what we do, in plain English: we help companies who manufacture physical products bring those products online. We add connectivity to their products (Wi-Fi or cellular), we provide security and device management, we let them push out software updates, and generally we help them overcome the many hurdles and technical challenges associated with the exciting new industry that we’ve started calling “the Internet of Things”.

But let’s rewind a bit. What is the Internet of Things, exactly, and why does a company like Particle need to exist?

The Internet of Things has been around, in some form, for decades. In the early days, we used to call it telemetry (literally, “remote measuring”), and later we started calling it telematics when we started adding remote control. Then, in the ’90s, we started calling it machine-to-machine, or M2M. The term Internet of Things, or IoT, was coined in 1999, and really hit its stride in the early 2010s. Veterans of the industry will tell you that these terms all mean slightly different things, but for all intents and purposes, we just keep rebranding the space as the market expanded.

Other things have changed, besides the name, over thirty years. Wired connections are becoming wireless. Local networks and servers are being supplanted by the cloud. But perhaps most importantly, the building blocks of IoT — computing and connectivity — have gotten a lot cheaper. Adding a computer and a radio to a device costs a small fraction of what it cost even just a few years ago.

It also helps that we’re now all carrying around smartphones in our pockets — little portable interfaces for all our connected “things”. Since connectivity is cheap, and we’re all carrying around screens, it feels like it’s time for IoT to explode.

But while a handful of things have come online — like Nest thermostats, Tesla electric cars, and GE wind turbines — it doesn’t feel like everything is connected. Instead, it feels like you have a few market leaders creating top-notch IoT products, surrounded by a bunch of worthless garbage. And at the same time, for every “thing” that is getting connected, there are thousands of “things” that aren’t.

It often feels like IoT is overhyped. Why? What’s keeping this revolution from happening the way that it’s supposed to?

Here’s the dirty little secret of IoT: it’s really, really hard to build a well-functioning IoT product.

It’s hard to nail down exactly why it’s hard to build a good IoT product. That’s because there are about a thousand things you have to get right, and any one mistake could sink you. You might trip up over antenna design, or security implementation, or reliable networking. Your servers might experience downtime. Your setup process might work on iOS 10, but get borked in iOS 11, or the Android version might work on all phones except the Samsung Galaxy S8. You want over-the-air firmware updates to futureproof your device, but any reliability issues could lead to bricked devices, and any security issues could turn your fleet of devices into a botnet.

What happens is that companies underestimate the complexity of building an IoT product. They invest limited resources and set tight deadlines, and then their IoT team fails to deliver. Gartner reported in 2016 that 75% of IoT projects will take twice as long as planned:

Gartner expects three out of four IoT projects to face schedule extensions of up to 100 percent with the consequent cost overruns. The more ambitious and complicated the project, the greater the schedule overruns. For some projects, compromises will be made to keep them on-schedule, leading to significant weaknesses in performance, security or integration into existing processes. In the mid-to-long term, these compromises will require that the IoT project be refactored and perhaps even recalled and redeployed.

This estimate has also been backed up by a recent survey from Cisco: of 1,845 business or IT decisionmakers surveyed, 74% saw their IoT projects fail to see the light of day.

A stock photography representation of how it feels to try to launch an IoT product in 2017 without support.

Particle launched in 2013 to make it easier for people to build IoT products. After our own failed consumer IoT product, we started launching development tools for engineers building IoT products. We took on nasty challenges like implementing software-based security that fits in 10K of RAM in order to run on a resource-constrained microcontroller, reliable over-the-air firmware update tools, and bandwidth optimization for cellular-connected devices. We built a supply chain to provide IoT hardware. We built an MVNO to provide SIM cards and data plans. We built the hard stuff so that our customers don’t have to.

It worked. Particle’s ease-of-use has made us pretty popular. We’ve built the largest community of developers and engineers in the industry — more than 120,000 people are building IoT products with Particle. Our customers get their IoT products to market in 1/3 of the time at 1/10 of the cost, and their products are secure, reliable, and “just work”.

The company that we’ve built as a result is a little bit… odd. Yes, we’re a cloud platform. And also a hardware company. And also a cellular carrier. We do professional services. We build custom solutions. We do it all.

This sounds antithetical to the common entrepreneurship mantra of focus. When you build a start-up, you’re supposed to do one thing and do it well. But if we only did one thing, we wouldn’t be solving our customers problems, because then they would have to deal with all of the complexity of integrating all the various layers of the IoT tech stack. We take on these challenges because our customers need us to.

An Opti stormwater management system, powered by Particle

These days, Particle is powering a wide swath of the emerging Internet of Things. You can find us in stormwater systems, making sure that reservoirs don’t overflow during a heavy downpour. We’re under gas stations, making sure that the fuel tanks aren’t leaking. We’re on natural gas rigs, monitoring methane emissions to comply with EPA regulations. We’re on industrial motors, making sure they are functioning properly. We’re in coffeemakers, in pet toys, in cars, and in gardening systems.

You’ll find us in prototypes built at IDEO, Lunar Design, and Mindtribe; at the Met Gala in an IBM Watson-powered dress; at Twilio SIGNAL in the badges; and at Google tracking their campus bicycles.

You’ll find us in the R&D labs at half of the Fortune 500, and in 90% of Fortune 500 IT companies. We’re at P&G, NASA, and everywhere in between.

We’re in an exciting place at an exciting time. IoT is just starting to take off, and Particle is quietly playing an important role. Now it’s time for us to get a bit louder.

Today, Particle is announcing that we’ve raised $20 million of new investment, led by Spark Capital — early investors in companies like Slack, Twitter, Oculus, and Warby Parker. Spark has earned a reputation for investing in great founders with stellar products and a massive vision. I’m grateful they considered us worthy of their time and energy, and delighted to have them joining our board.

Raising this capital is an important milestone for us — it allows us to significantly invest in growth. We’ll be using this capital to solve more problems for more customers; we’re growing our team so that we can engage with and support more companies, and we’ll be expanding the scope of what our platform does to further simplify what it takes to build an IoT product.

Right now it costs millions of dollars to bring an IoT product to market. Our goal is to make it cost thousands of dollars. When that’s the case, entire industries will be revolutionized. What the web has done for industries like retail and publishing, IoT will do for industries like manufacturing and logistics. And we’ll be out there, leading the charge.

To learn more about Particle’s Series B raise, read the investment announcements from Andrew Parker (Spark Capital) and Patrick Eggen (Qualcomm Ventures).