Why Low-Priced 3D Printer Companies Often Fail

Counting that cash

Counting that cash

Last week we told the story of XMachines and their extreme financial challenges. The failure to deliver product is usually just a result of bad arithmetic and planning.

There have been waves of low priced 3D printers projects attempting to reach the market, but many of them fail miserably. Yet they often attract a great deal of attention because their low price is seemingly attractive to thrifty buyers. 

That market segment is highly competitive and thus requires very careful attention to pricing, as margin levels will be very tight. 

For example, let’s say you’re considering selling a budget desktop 3D printer for USD$300 per unit. The parts margin on such a machine is certainly going to be very slim, say only USD$30 per unit (or maybe even less). 

Now because of the low price, you receive a massive pile of orders for say, 10,000 units. But for all the work of building, and shipping 10,000 units, you only profit USD$300,000. 

And from this USD$300,000 you pay your staff. At say, a budget salary USD$40K per staff member, you can pay for 7.5 people for a year, or more likely, 5 people for a year and a half. 

Now, can five people produce 10,000 units? In a year? Properly? 

You’d think you might be able to produce 833 units per month for  a year, with each of the five making 166 units, or about 8-9 per day. 

One per hour. 

But wait - you can’t use all five people to build machines. You have to do a bunch of other things to keep the business running. You must: 

  • Handle marketing, advertising, web sites, social media, communications and the like
  • Handle financials, bookkeeping, payments, loans, investments, receivables, etc.

Each of these can take a person. Three left. 

But wait, this all assumes you have a machine design that works perfectly and needs only to be assembled. That’s almost never the case. There will be bugs, and different versions of the machine will emerge. So you also have to: 

  • Handle hardware design & development, maintenance and customer enquiries and problems
  • Handle software design & development, maintenance and customer enquiries and problems

One left. One person worth of cash to assemble 42 machines per day, all year long. 

Nope, that’s not it, because SOMETHING will go wrong. It might be a shipment delay, or a bad batch of components. Or a bug requiring already-built machines to be fixed before shipment. Or, or, or. 

Yes, you could attempt to sell additional machines to raise more cash during the year, but guess what - you end up with even more machines to build, and more people time tied up doing the sales. 

How can you possibly handle all this? 

You do so with cash. Cash for additional arms and legs to execute quickly. Cash for emergency shipments of replacement parts. Cash for additional testing to ensure things work properly. Cash for contingencies. 

You get cash by having more margin. You don’t have cash if you have a thin margin. 

That’s why I am always skeptical of low-price desktop 3D printers. You have to be incredibly savvy in business planning to survive such circumstances. One very successful company in this regard is Printrbot, who produce low cost machines. They are incredibly diligent in their financial planning to ensure they have the best price - but also sufficient cash flow for continued operations. 

When I see a product launch with a price of say, USD$1,000, I can be reasonably assured they’ve allowed enough margin to actually do the job. On the other hand, a USD$250 machine from a fresh startup is pretty questionable. On yet another hand, a USD$250 machine from a large, established company is likely a sure bet, because the larger enterprise can subsidize operations. 

The bottom line here is this: the lower the price, the more skeptical you should be. 

General Fabb

Kerry Stevenson, aka "General Fabb" has been writing Fabbaloo posts since he launched the venture in 2007, with an intention to promote and grow the incredible technology of 3D printing across the world. So far, it seems to be working!

+