Security Engineering: Inside the Scooter Startups

A year ago, ridesharing scooter startups were gearing up for launch. Workers at Bird, Lime, Skip, and Spin were busy improving their app, retrofitting scooters, and most importantly, figuring out the logistics of distributing thousands of electronic scooters along the sidewalks of the Bay Area. These companies were gearing up for a launch in early summer, but one company — nobody can remember exactly who — decided to launch early. First mover advantage, and all. Overnight, these scooter companies burst into overdrive, chucking scooters out of panel vans onto the sidewalk simply to keep up with the competition.

The thing about San Francisco, and California in general, is that it’s a very direct democracy masquerading as a representative government. Yes, there are city council members and a state legislature, but the will of the people will rule. No one liked tripping over the scooters littering the sidewalks, so the scooters ended up at the bottom of a lake. Or in trees. Or in the trash. In time, city permits were issued, just like a hot dog cart or any other business operating on a public sidewalk, and the piles of electric scooters disappeared. Not before hundreds of scooters were vandalized, that is.

It’s still early in the electric scooter rental startup space, but if there’s one company leading the pack, It’s Bird. they’re getting the most press, the CEO was formerly at Lyft and Uber (which explains the press), and they’ve raised nearly a half Billion dollars in funding (which explains the press). Bird is valued at two Billion dollars, and it’s one of four major ridesharing scooter startups. had nothing on this.

Despite how overvalued you think a scooter startup might be, they’re still a business, and they’re ruled by the bottom line. Bird has grown a lot in the past year, and with that comes engineering challenges. The Bird scooters must be more resistant to vandalism. The Bird scooters must be harder to steal. Above all else, they must remain in service longer. This is the teardown of how Bird managed to improve their bottom line and engineer a better scooter.

The Economics of Scooter Rentals

The economics of Bird scooter rentals has been covered in-depth, but came mostly in the middle of last year during the increasing distribution and meteoric descent of ridesharing scooters. The best assessment of Bird’s economic model from this time comes from Haje Jan Kamps of Bolt. The unit economics of Bird scooters actually makes sense, and a scooter rental company can actually make money. The key is how many times the scooters are used before they become unsalvageable.

There are problems with this economic assessment; the assumptions are wrong, and the back-of-the envelope calculation that Bird was spending $400 to put a single scooter on the road was incorrect. According to The Information, in May — a month after Jan Kamps assessment — Bird was spending $551 per scooter. This affects the numbers, but not the calculations. If a scooter can last long enough to generate a profit, Bird will make money. Everything else is secondary.

And so we come to the engineering problem. How do you improve the design of a scooter to last longer? Any problem with Bird’s financial plans aren’t tied to laws or city permits. The core problem is making the scooters last longer, and over the past year they’ve been improving their devices.

Four different versions of Bird scooters. Image credit: reddit user ITTVx

There are in fact two (or three, or four, depending on your interpretation of ‘wide distribution’) different versions of Bird scooters. The first, and original, is a rebranded Xiaomi Mijia M365 electric scooter. This is not the ‘secret sauce’ Bird keeps close to their chest; resources for Bird Mechanics (people who are paid to repair Bird scooters on the street), refer to this version of the Bird as the M365. The parts are identical. Whatever can be gleaned from the mechanic guide and the actual manual of the Xiaomi M365 reveals these scooters are identical. Since the M365 is a consumer scooter, you can also buy replacement ‘dashboard boards’ to replace the Bird brain, allowing you to take garbage off the street and turn it into your own personal scooter.

As an aside, concerning the M365 scooter, the most valuable component in the scooter is the battery pack. You can buy an entire scooter for about $400. A replacement battery pack containing thirty 18650 cells, with the associated connectors, thermistors, balance charging connectors, and a handy wall-mountable frame, costs $220. This reinforces my belief that the ultimate use of discarded Bird scooters is not as liberated scooters, but as the raw material for a Tesla-style Power Wall. Even considering the cost of raw cells, thirty 18650 cells would cost about $100. The batteries are the golden egg inside Bird’s scooters.

The second main model of Bird scooter is the Bird ESB / ESX. These scooters are manufactured by Ninebot, a subsidiary of Segway, and are vastly more substantial scooters. They weigh more, they’re sturdier, and more importantly to Bird’s bottom line, they are vastly more secure. However, being more robust comes at a cost: the retail price of the Ninebot scooter is approximately $150 more than the Xiaomi M365.

Brain Comparison.

The Bird Brain 1.3, found in the Xiaomi scooters.

The first generation of Bird scooters, based on the Xiaomi M365, used what can only be described as a prototype-level electronics board. The process of converting was simple enough: rip out the Xiaomi controller board, put in a Bird Brain loaded up with a Particle Electron dev board, add a standard, off-the-shelf GPS breakout board, and add a few connectors so it will drop into the Xiaomi frame.

There is no secret sauce to the first versions of the Bird Brains. This is a PCB populated with breakout boards. In terms of engineering, it’s a quick way to get a product out the door, but it is expensive; a better, cheaper solution would be to find a System-on-Module (SoM) that can do cellular and GPS. Tighter integration leads to lower costs.

Although the first version of the Bird Brain was marginally acceptable from an electrical engineering standpoint, mechanically, it failed. The enclosure for the Bird Brain was insufficient, and there are videos of Bird users smashing the Brain box with a rock to disable the scooter. With proper tools (a single screwdriver), the Bird Brain could be completely disabled and discarded in less than two minutes. Again, how-to videos for how to disable an M365 Bird scooter exist.

Replacement controller boards Xiaomi scooters costs $30 which means that stealing a Bird scooter costs $30 (two minutes of time). I haven’t seen a moniker for de-braining a Bird scooter to turn it into your very own personal transportation device — so I’m calling it Bird Boxing.

The V.2 Bird Brain, designed by Mobilogix. Image credit: FCC

The new Bird scooters, on the other hand, are more integrated and more secure. The design of these brains were contracted out by Bird to Mobilogix, manufacturers of asset tracking and IoT devices.

Notable features of the Mobilogix boards include a much, much more integrated design. Instead of Particle dev boards, this board features a Quectel wireless module with an integrated GPS receiver. The connectors for this board are not designed to fit the wires of an existing scooter, instead the scooter and brain board were designed in tandem. Overall, the V.2 Bird Brain is vastly more integrated, a cleaner design, but still slightly overkill. There’s no reason for Mobilogix to use a Cat-1 modem for a device that is sending a few dozen bytes to Bird servers every few minutes.

But Bird learned a lesson with dozens of their scooters finding their way to the bottom of Lake Merritt. The mechanical design of this enclosure is robust. The plastic is polycarbonate, instead of the cheaper ABS found in the first Bird scooters. There’s a gasket meant only for security, as the holes for the buzzer/beeper allow for water intrusion. The installation of the V.2 bird brain requires security Torx screws, where the first Bird Brain was held together with Philips screws and snap-fit plastic. This is much more robust, and the only apparent reason is to deter theft.

A More Expensive Scooter Makes More Money

There is no doubt in my mind the rev 2 scooters, these are the ESB /ESX made by Segway/Ninebot, cost more than the Xiaomi scooters. The tighter integration found in the ESB / ESX and the Bird Zero Brains might save a bit of money, but the ESB / ESX chassis is much more expensive, nearly doubling the cost. The plastic enclosure for the ESB / ESX scooter is more expensive, and the fasteners are more expensive. Bird makes up for this with more riders per scooter lifetime.

This view is reinforced in an interview with Bird CEO Travis VanderZanden. At a summit that’s so hip the chairs were made of Astroturf, VanderZanden said, “Clearly the unit economics didn’t work on those [early] scooters, but that was a test anyway… We quickly scrambled and started creating our own scooters. What we see on the unit economics of those, it’s like night and day.”

What we see here is more evidence Bolt’s assessment was correct: if the scooters live long enough, the company will turn a profit. This becomes a case of Security Engineering, the field that’s responsible for making things tamper-proof and vandalism-resistant. Human Factors also come into play; it’s harder to throw a 30-pound scooter into a lake than a 15-pound scooter. These are subjects we don’t talk about very often on Hackaday, for understandable reasons. Generally, we want to know how to get into something and where our box of security bits went to. We’re not usually designing against vandalism. This, though, is a perfect case study. While Bird is a company that’s only worth Billions on paper, they have managed to do something that doesn’t immediately make sense: they’re spending more on their product, and they’re making more in return.