Ask HN: I want to build/sell a product. Steps?

0. Hack together a prototype.

1. Go get some users for which your device stops pain. Get them to sign NDA's if necessary, and pay them a small stipend for their time.

2. See if your design really works. You can tell this if they do not want to give the device back.

3. Iterate.

Once you have something that is really valuable then get the patent. We made the mistake of getting designs done way too early and it cost us heaps in terms of ability to iterate and money.

Customers don't care what it looks like as long as it works. We had the most "customer success" with a grey $2 box when prototyping. Iterate early.

I can help.

Depends on your project, but designing a couple iterations in 3D CAD is usually the first step. Life like renderings, stress testing can be performed on the models.

Prototypes can then be made directly from that via 3D printing, CNC machining, low run injection molding, laser cut sheet metal, etc...

Do a few iterations until you get a something functional, attractive and can be manufactured for a little as possible.

Then source a manufacturer either local or abroad - whatever is most cost effective.

In my experience, patents are a waste of time unless its absolutely necessary. Get it to market as fast as possible because there are 100 other people thinking the same thing and 5 may have the means to do it.

You can contact me here: casey (at) elevationlab.com

Good luck!

-Casey http://www.elevationlab.com/

Thanks for the info–I'll be in touch.

-2. buy google adwords for the benefit / pain your product is alleviating. 1. measure searches/clicks to see if it's worth building a product. 0. all the steps above

I really like this idea. Felt it was worth more than an upvote. Great way to get a little bit of market research without much cost or effort.

You should check out contract manufacturing. Many of these types of small jobs can be handled by a CM in China. The problem with any physical device is cost--cost in materials, machining, tool building, assembly, etc. This means you'll probably need to get some form of financing. You may be able to do a small run for not too much (10's of $k), depending on the device. This is why I always opt for bits over atoms these days.

By the way the first place to look for money to pay the manufacturer is the manufacturer itself. They need you as much as you need them and will often provide cash for equity if order from them.

I don't have direct experience, but on the purely practical level:

a. you could do worse than learn to use sketchup. I can't tell from your description how complex or mechanical your 'thing' is. If it's beyond your capabilities, advertise for someone to do it, but have them sign a 'work for hire' agreement which includes a rigid confidentiality clause.

b. it's hard to say because I can't tell whether it's a widget (handy-shaped thing) or a gadget (handy single-function machine). If the former, look into Rapid prototyping and/or 3d printing. If the latter, research similar products, find out what class they are described as, then investigate manufacturers in China.*

* Who will probably ignore your confidentiality clause, foolish roundeye.

I'm sorry this is so hand-waving, but I honestly don't know whether you're making a plastic resistor drawer or a self-regulating soldering iron.

It might be worth advertising for an electronic engineer to do some consulting, executing the above-mentioned work-for-hire agreement, and asking them to evaluate the functionality of your prototype. Best of luck.

For the drawings, its best to contract it out to a Industrial Designer/Mechanical Designer. Find a freelancer if you can or search through LinkedIn to find a good reference.

For production run the best bet is get the PCB's made, source the components yourself (try DigiKey) and then hand assemble and package them yourself or with your friends. There are some services provide PCB's in small quantity. I dont have the link handy but you should be able to Google it.

There are some electronic enthusiasts forums too where you can find some help.

0. Figure out if your market is B2B, B2C, etc.

1. Find a way to talk to your prospective customers according to their channels. Sometimes this is easier in B2B.

2. Talk to them about your ideas (ideally with some mockups) and see if it's something they are interested in.

3. If so, do whatever it takes to get money in exchange for something that you can build as quickly as possible.

4. Iterate based on their / market's feedback.

5. Worry about patents, etc. once you have an idea that you can actually make money with.

Edit: this is my approach, anyway. FWIW.

I am not a lawyer but it is my understanding of patent law in America that if you sell an item that you want to patent the patent office might be able to deny your claim because of prior art. Let me know if this is off base.

http://en.wikipedia.org/wiki/Prior_art

My understanding is that you have 1 year from the first time you publicize the invention to file your provisional. For international patents, though, this is not the case; you need to do it right away.

I am also not a lawyer.

Perhaps voting down is one way of saying that the above information is incorrect. It would be more helpful, however, to provide the correct information.

According to a couple of books I've read, the 1-year rule applies in the US, but if you're interested in foreign patents, you have to keep your design under wraps until after you have applied for the patents, at least in several important jurisdictions.

Go straight from 0 -> 4. Get feedback before spending more money. If you can't find customers with just your prototype, you either won't be able to sell it later, or maybe just no-one wants it.

I've found that most of my product ideas fail in the "marketing" step (which you omitted): either I can't clearly identify the customers, or the cost/difficulty in reaching them is prohibitive. In either case, it's a fail.

This book has some good advice:

http://www.squeezedbooks.com/book/show/1/how-to-bring-a-prod...

As I said in the comment though, the summary isn't great because the book is not a hand-wavy business book, but has a fair amount of advice. Worth a look at the library to see if you like it.

1) can you not do any hardware yourself, and choose to license it to someone else? 2) if you must do hardware yourself, can you have anyway of scoping product, doing google adwords, anything to assess demand and product fit?

The hardware business is BRUTAL. It is going to be exceedingly tough to get financing for your project, given high capital requirements and low profit margins.

You forgot to mention one early step:

Research market/niche viability. Are there enough likely customers out there for what you want to build and sell?

It's a good idea to do this up-front before sinking hundreds of hours into a project, only to find that A: Few want it or (more infuriatingly) B: Many want it, but in a different form.

There's definitely enough likely customers ... and it addresses a pain point for me personally ... and I assume for others too