Author, Publisher, Developer, Stitchcoder


Actually Making Things! Kits Began Shipping Today

My son Connor claims that I know how to do this because I've done it before, but I don't think so. I'm pretty sure I'm just making it up as I go along, because as far as I can remember I've never actually mass produced a product for sale. (I'm not counting quilts because they are made one-off, not in a production line way.) 

18 copies of the Radial Engine kit.

If you're one of the people who have ordered one of my Mechanical GIFs kits, yes they have actually started shipping, and most likely all current orders will have been shipped by the end of the day tomorrow. See, here is a picture with a box full of finished kits in their retail packaging! (The last potential delay cleared up last week with the delivery of 5000 springs for the Pin-Tumbler Lock model.)

The concept of 5000 springs raises many questions in my mind. Like "where do you get 5000 springs?" and "why do you get 5000 springs?" and "is that a big box of springs or a small box?". I mean, that's enough for a thousand Lock kits, and I didn't even know if people would order ten kits, let alone a thousand. Am I nuts ordering that many?

5000 springs doesn't look like much: This is all of them, together weighing less than a pound.

The answer to the first question, in this case, is the W. B. Jones Spring Company in Wilder, Kentucky. They are price-competitive with Chinese suppliers in the range of quantities I was looking at, with far shorter lead times. Yes, we in America can still manufacture things!

Why 5000? Because the economics of the situation push towards the maximum remotely plausible quantity. Purchased individually, these springs are $3.70 each. The five in my Lock model would be $18.50, which is of course nuts! But in units of 5000, they are 6.5 CENTS each, over fifty times cheaper (making the total $0.32 per model). By looking at the slope and intercept of price quotes for 1000, 2000, 3000, and 5000 units, it's easy to see that the formula for anything over a thousand units is $170 plus 3.1 cents/spring (total of $325 for 5000). In other words, there's a $170 setup fee for them to configure the machine to make that particular spring, and after that it costs 3.1 cents each to make them (including their profit). For small quantities they presumably either have a bit of stock on hand, or they lose money making enough to cover the order, plus some more to put into stock for the next small order.

It's not unlike the economics of color offset printing of books and posters. The first one costs a fortune because of setup time, but after that, they are just pennies a piece. So you should order as many as you can justify, to avoid paying the large setup costs again.

So I ordered a lot of them. I'm also pretty sure that I'm going to need similar springs in other models I have planned for the future, so unless the whole concept is a miserable failure, I do expect to use these up over time.

This is what 1000 tiny screwdrivers looks like.

I mention Chinese suppliers because for many things they are not actually the cheapest option. They are the only option. For example, suppose you want a bunch of those tiny disposable screwdrivers you get in some kits, and which I wanted to include in all my kits? It took me a while to figure out the right search terms, but eventually I found them for 8.3 cents each in quantity one thousand ($83 for a thousand, shipping included).

It was, of course, from a Chinese supplier. I don't know this for a fact, but I would be surprised if there are any US manufacturers of tiny disposable screwdrivers. I certainly didn't encounter any in my googling.

In the past my main problem in getting these screwdrivers would have been how to spend only $83 with a wholesale supplier who is literally half way around the world. If I wanted a million tiny screwdrivers for $80,000 there would be all sorts of ways, but in the past, the answer for such a tiny order was "you don't". It simply was not practical because of the high cost of finding and doing business over such distances. 

Today, the answer is or, the outside-China focused websites of the giant Taobao (淘宝网) marketplace owned by the even more giant Alibaba group. At you can buy small wholesale lots from countless thousands of small merchants, who in turn buy from the real manufacturers and resell with very little markup. It's basically a virtual version of the vegetable-stall style of industrial supplies market I talk about in this blog post. At the focus is on larger quantities of serious industrial equipment (but many merchants there will also deal in smaller quantities).

The more I think about it, the more I am convinced that the existence of aliexpress and its relatives is an important step in the restoration of industrial innovation and small-scale manufacturing in the United States. Yes, we need Chinese suppliers to be able to do industry in the US.

The thing about manufacturing is that it only works, in a competitive way, when each company is part of an ecosystem of suppliers and customers. For example, there used to be an ecosystem like that in Detroit to make cars. They could make cars faster, cheaper, and better in Detroit than anywhere else in the world, because everything needed to do it was just down the road. If one of the big companies needed a new kind of headlight, they could just go a mile over and talk to the engineers at the headlight manufacturer, who in turn could talk to the glass molding outfit next door, and so on. That's all gone now, or rather it's moved.

When I first started making quilts I thought about how we needed to price them, and whether we could consider going after mass markets. The answer is "not if we make them in America". This has nothing to do with labor costs, it's because we would be competing against factories in parts of the world where there are integrated manufacturing ecosystems for textile products. We would have to get fabric and batting shipped to us from far away. They can get supplies from next-door factories. They are also next door to the companies that make the machines that make the quilts. (We have one of these machines. There are thousands in China surrounding the companies that make them.)

So we make "fancy" quilts that sell for 3-5 times more than the ones you get at Walmart. Some of that extra price is because there's a lot more stitching, or they are custom designs, or just very cool, but some of it is just because we cannot possibly be as efficient making one or two at a time (compared to, for example, this factory which makes 1350 quilts a day using five of an earlier model of the same machine we have).

With the kits I'm trying to keep the cost as low as possible. It's more realistic to be competitive because the parts are much smaller, so shipping in components isn't as cost-prohibitive as with quilts. And I manufacture the highest-value-added parts in-house (using my large laser cutter) from raw stock (acrylic sheets) that I can get at a good price close enough (Chicago) that I can economically drive up there and pick up new stock in person from time to time.

If the kits are successful online, in museum shops, and through educational distributors, I think it is realistic to keep making them here in Illinois. But only because I have access to a wide range of wholesale parts through aliexpress/alibaba. It really is the difference between happening and not happening. It's an example of what people mean when they say the world is shrinking.

Some people dream of restoring the kind of geographically-concentrated, vertically integrated manufacturing regions we had in the past (like Detroit). But that's not going to happen, and it's not the right goal anyway. We should instead look towards the inevitable future where the whole world is that region, for everything. We could be leaders in the game of bringing together the best from everywhere by embracing rather that fearing the merging of our destinies with those of our friends across the ocean..


Theodore Gray1 Comment