1. 56
  1. 11

    That BOM to me seems quite high, and sounds a lot like what happens when you fail to convert from general-purpose prototyping components into application-specific circuits.

    “Productization” is a super important step in all commercial projects, and I wish there was more of a recognition around that both in software and hardware. (You of course can go too far in that direction and decide not to cut any corners, but that has its own problems.)

    1. 10

      Excellent tear-down of both the technology and the company!

      There’s also a giant hole [in the back of the case] for some unknown reason.

      To my untrained eye it looks like the sort of hole intended for you to put your finger through to easily pull the back board off, instead of needing to turn the sign upside down and shake it.

      I imagine that in a city of about 8.5 million it shouldn’t be too hard to sell 600 signs in at least the first year. But next year you’d have to sell ~600 signs all over again. And this isn’t including the cost for any other employees, contractors, etc.

      No reason you couldn’t expand to other cities, is my first thought about this. If you hold steady and sell 600 signs a year in a new city every year to cover payroll and costs, then frankly your business is small but profitable for at least a couple decades. Or, in the more realistic case of it being a lot easier to add new cities than create the thing in the first place, cover the top 50 cities in the world in 5 years and make a lot of money. Glancing at Wikipedia, NYC has the 11th largest subway system in the world, so it seems like a pretty decent market if you can go international, even though that might take a fair amount of work.

      But yes, it’s definitely something you’ll need to think about. I’ve recently learned this is one of the things you pay marketing people for: to figure out the answer to the question of “if everything goes perfectly, how much money can we actually make off of this idea?”

      1. 4

        This is a fantastic post, and an interesting run through of the adventure.

        But now this makes me want to make my own sign. I have plenty of RPis and ESP32s to work with. Just need the LED matrix displays. Might have to check mouser or digikey

        1. 1

          Great article, the idea that a company dedicated to selling an IOT device had 15 employees is just baffling. Maybe I’m out of touch (also don’t live in NYC) but it looks like a niche product, not something that everyone and their brother would buy.