Best Practices for Crowd Funding Electronic Hardware/Devices

[1276-2449-1-ND]

A while ago there was a thread here that had a few posts lamenting how people were structuring or running their crowd funding campaigns in a way that made it harder for them to pull it off if they got funding -- and the more funding they got the harder it became for them (and of course I can't find that thread now). There are lots of examples of successful campaigns that ended up folding even though they had a viable product.

What do you think should be "Best Practices" for running a hardware-based crowd funding campaign?

[EEVblog]

Make a dozen real pre-production units with your own money first.

[xaxaxa]

Keep the reward tiers simple; don't offer too many variations and stretch goals. This can easily double R&D resources required or more, and add headaches to production.

[LapTop006]

Do some rough math to work out what a "success disaster" would look like, and don't forget that practical production yields will be less than 100%.

For example, if a particularly expensive part is only available in full reels of 250, what happens if (ignoring yields & spares) you need to deliver 251 units?

Also, your full prototypes should include everything you plan to include, and a real shipping box both to prevent surprises in how much packaging can cost, but also to get you a much closer idea on what shipping will actually cost.

[SiliconWizard]

Make a dozen real pre-production units with your own money first.

I second that.

Many people try getting funding for their project way too early in the design process. A lot even have nothing more than just ideas and pretty 3D renderings and videos. That's the projects that are bound to fail IME.

That's pretty much the same pitfall for startups. Common sense and experience suggest making first prototypes on your own funds (and/or family/friend funding) before trying to get external funding.

[Stray Electron]

  I would say that "Best Practices" and "Crowd Funding" are nearly always mutually exclusive.  Unless you include the word "Fraud" in there somewhere.

[xaxaxa]

The other thing I would mention is there are realistically only two countries you can do shipping from, which are the US and China. For anything medium value ($20 to $200), customers will expect full tracking (and you will want it too), and the national post in most countries just won't offer that to international destinations for a reasonable price. I would rule out couriers like DHL or Fedex too unless you have >100 units per month of sales (and can therefore sign a contract with them to get a lower price). $50 shipping on a $100 product just ain't competitive. For anything low value (<$10) even national post without tracking is usually too expensive (reason I never bought a raspberry pi).

If you live in a first world country I would seriously recommend crowdsupply which will do the final shipping for you. You ship the entire batch to them and they ship to individual backers via USPS.

[biot]

If you live in a first world country I would seriously recommend crowdsupply which will do the final shipping for you. You ship the entire batch to them and they ship to individual backers via USPS.

Respectfully, that's terrible advice. I bought the EEZ H24005 power supply (much discussed right here on EEVblog) via Crowdsupply. The unit was made in Europe, shipped to Crowdsupply in the US, and then shipped back to Europe for me. That took several *months*, and €114 in customs fees.

Crowdsupply is just a terrible deal if you're not in the US.

[janekm]

Keep the reward tiers simple; don't offer too many variations and stretch goals. This can easily double R&D resources required or more, and add headaches to production.

I'll definitely second that being currently involved in the fulfilment of a Kickstarter campaign... Which had very modest extra tiers but even just an extra poster or patch ends up being a lot of hassle during logistics. Consider making higher tiers multiples of the base tier, or perhaps something that can be digitally delivered.

[pylo]

Make a dozen real pre-production units with your own money first.

This.

Many campaigns, especially some of the larger ones like Ossic, Fontus, PocketDefi, SuperScreen etc. all fail because they underestimate the needed R&D costs until the product becomes ripe or even just functional. A common factor to their initial success then sudden failure is they give the fake impression to backers as if they already had a working product, but in reality they were just using empty mockups or computer renders in their marketing videos. In reality all they had at campaign launch was an idea, a designer, and some photoshopping skills. Then when money is there and they actually have to deliver, they realize how much they underestimated the hurdles (not just technical) ahead of them.

So, I'll reiterate Dave's words: Build a bunch of prototypes, and only go into campaign once they are actually working. Don't worry, you'll still have lots of problems when going into volume production and entering the market

[LiTTT_up]

I've done precisely that in my crowdfunding campaign. Dave / Admin couldn't be more right about this. If you don't believe and trust in the design of your product enough to shell out for some pre-production units from your own money... you may be limiting how smooth everything will go after that.

Pre-production units offer a lot of learning opportunities from design challenges to logistics. You'll discover changes that need to be made and might even find a way to save money in the process!

In fact, Kickstarter now requires that you have working prototypes BEFORE you take anyone's money. I'll post my crowdfunding project in the appropriate category on the forum here, but here's it is below for reference on the prototypes. I just launched it last night, so I'm not sure how well it's doing yet.

http://kck.st/2z5hhzw

I used FDM/SLA and installed the copper myself. Working on the tooling for sheet metal now.

[coppercone2]

be sure to pull a communist stunt that would make uncle Joe proud. The latest AVE video is like that. The channel is supposed to be about like, lower-tech/low barrier to entry stuff, typically old beat up cheap mill and skilled use of hand tools etc... but he takes his patreon money and buys a 5 axis CNC machine, then proclaims its a 'community device' that is basically collectively owned by the commune.

Exactly why I refuse to fund random people online unless there is a very specific plan/cash flow.

My advice would be to make a road map and try to do honest estimates of where the money is going, what avenues of R&D its going into.. so the purse does not feel cheated. You could go as far as saying how many billable hours went into various tasks, like I paid myself 15$ an hour to work on the graphical design and product videos and shit, worked for this many hours so it does not feel like your throwing your money into a black hole... however many business people would consider such transparency suicide imo.

[coppercone2]

no but most of his videos are cool shit that makes it interesting to watch, which tool is the best bang for the buck, how to do fairly complex things with less effort and tools.. etc, but now you have a jump to something that looks like it costs at least 50-100k. It would be like if EEVBlog decided to buy a Cray and pushed code optimization for fluid simulation on us or something. I would stop watching it.

when you put money into a channel you think its gonna be more of the same stuff that you find interesting and useful.

I don't see why dave would do a test service, the idea behind his channel is that he is an expert that can offer insight into advanced design techniques.

[coppercone2]

with AVE though, if he just uses the CNC as a time saver, and still makes the projects so that you can replicate them without weird advanced tools and does not start thinking in a way that basically makes everything exponentially more complicated to suit his 5-axis CNC capabilities, that's different and the channel will still stay close to the roots.

But the whole 'community owned' shit he was talking about just made be laugh because it sounds like 50's commie propaganda, maybe its honest but it sounds awfully like a 'oh shit what the fuck did i just buy with your money... yea.. its not mine... i spent your money to buy YOU something' defense. I am sure we all had parents do this to us, i.e. spending present or vacation money on home improvement or a car and sugar coating the fact that we were gonna go to camp and instead dad looks better at work because he got new wheels.

[coppercone2]

I mean, you have a PhD right? 6 years ago or whatever when I first watched the blog it was pretty advanced for me, now I already know alot of the stuff because when you do electronics full time you end up just getting this stuff.

[coppercone2]

But how are you gonna do community owned by mail? Is ave going to become a low-cost CNC operator? Does he even want to manage a business of mailing people parts?

If you are making people sub assemblies for cheap how do you ensure that someone is not taking advantage of your labor?

I think the EEVblog u-current and meter were different because IIRC Dave Jones was not crowd-sourcing the design or anything like that, I mean he asked like 'what would you like to see' but it was pretty basic (i.e. the battery powered power supply) in terms of discussion and stuff... he just decided to make and sell some stuff, and what we provided him with did not really have anything to do with it (since he had a list of projects going back way before in magazines and stuff I think).

With ave:
I can't see someone complying with "yea you got some money on patreon, I will mail you some aluminum, make me 50 of these" without being paid nicely. With a CNC what can AVE do for the community other then provide selectively applied really cheap labor?  In that case you stand to make money off of him if he likes you or your idea. How can custom parts benefit a community as a whole?

[coppercone2]

With ave:
I can't see someone complying with "yea you got some money on patreon, I will mail you some aluminum, make me 50 of these" without being paid nicely. With a CNC what can AVE do for the community other then provide really cheap labor?

Time is NOT FREE. He has all the rights to ask to be paid for the time he spent on programming and grinding metal.

What's more. his Patreon money IS HIS MONEY. He can blow them up on Louis Vuitton if he likes to.

He's being kind on feeding back the money to the community on building someone's dream. That's at least better than buying handbags, right?

yea but then how is it a community thing if you don't provide cheap labor? Most older CNC places probably already paid off the NRE for their multi axis machine. Like why would the community go to him for parts rather then just getting it milled else where?

If you read my post carefully you will see that 'i can't see anyone complying with', meaning that I agree that working for free would be unheard of.

A makerspace is different because you have code which you bring but you still end up babysitting the machine, loading it, etc.. ave would have to do all of that.

 Once you factor in the physical distance there is almost no difference to mailing your 3d file to a machine shop. It's just AVE's machine shop.

Maybe its cheaper because its one guy, you have basically no turn around time requirements (a business spends alot of money to organize itself to be prompt).

personally with AVE i think if he spent all that money on beer it might make for even more entertaining television, more so then with a 5 axis CNC lol

He can't even work that thing while drunk because a typo is going to destroy the cutter, its more sensitive to errors then a hand operated machine.

[xaxaxa]

Also don't confuse patreon with kickstarter.

Patreon money is strictly a gift. Once you donate someone money it is up to them how they want to spend it. They are not obligated to do any work for you or even keep producing videos. It's up to you to stop giving him money if you don't agree with what he is spending it on.

[langwadt]

Also don't confuse patreon with kickstarter.

Patreon money is strictly a gift. Once you donate someone money it is up to them how they want to spend it. They are not obligated to do any work for you or even keep producing videos. It's up to you to stop giving him money if you don't agree with what he is spending it on.

afaiu patreon have now banned giveaways to patreons, because they are concerned that it might violate gambling laws in some countries

[LiTTT_up]

That's good advice. One issue I've run into is getting larger companies with mid to high volume manufacturing capability to take you seriously unless you've got money written all over you. Even if you're a privately held upstart with proper branding and a web presence - I've found you've got to press quite a bit to get you're RFQs taken seriously. Many of the companies consider an RFQ on tooling or assembly, etc that shows up out of blue to be a null "no lead / no quote".

It was a fight getting my LED frames/connectors quoted. In fact, the best I get is a window on the tooling cost because they don't want to write themselves into a corner and commit until there's a surety or trust deposit exceeding 5K to 10K. For an established company that shouldn't be a huge problem, but for an upstart it is. I've got every detail of my crowdfunding project hammered out down to the dollar, on everything except tooling which I had to take the median estimate on since their window has a $10K margin on it. The reason for such a margin is tooling, and more than one quote shows that among three manufacturers I could even get to respond, they all had a similarly large margin on the assembly.

Just for reference, I'll list what I'm talking about here as just a link: http://kck.st/2z5hhzw

So that's my project in the link above. The injection molding with multi-cavity dies was no problem to quote. The overmolding had a tight and safe margin on it, but assembly before overmolding is still a small risk factor, mostly for me since it's tight enough that I can still deliver the goods, just with less profit if they hit the high end of the quote in the real world. With three hard-won quotes under NDAs, I've gone middle ground when pricing the backer packages. Product fulfillment wasn't hard to estimate either. But it seems to me, there's heavy capital companies moving in on the Kickstarter scene, coolermaster, ziplock, other big names with very serious pre-launch marketing campaigns that force their projects to the top of the list on kickstarter. Sounds like normal capitalism, but kickstarter didn't start out that way. It is inevitable though. Wherever there's money to be made the deep pocket sharks move in with a fury.

Reference image below:

[mikeselectricstuff]

Just for reference, I'll list what I'm talking about here as just a link: http://kck.st/2z5hhzw

Spring contacts onto non gold-plated PCB pads seems like a recipie for trouble.
Also if you're going to this effort, it should also provide connection points for cabling to reinforce the power connections as the voltage drop along a LED strip can get significant.

[LiTTT_up]

Spring contacts, whether solid core or stamped sheet metal is currently the main method employed for electrical continuity between connectors & flexible LED strips. There are also the 4-pin insert variety (which are gold plated) but require the female socket to be soldered anywhere a termination connection is desired

Although there are a few connectors out there that have a sharp point which partially punctures the strips pads/layers for a better bite at the cost of some wear and tear on the strips.

Gold plating the pads on LED strips is practically unheard of in the consumer electronics world. There's just not enough money in them for that. Perhaps in aerospace applications, but these strips don't sell for enough to warrant such an upgrade. If the pads wear out, then the strip gets replaced.

Unless you're talking about mixed metal reactions., although I've had white LED strips in parts of our office running for ages on the same old cheap compression pin connectors, they had a nickel allow in them if I'm not mistaken.\

Power injection is possible with LiTTT™ Snap! connectors, it's already part of the design. Pigtail connectors can be added to any connector - anywhere in the net, that is to say - at any point down the line to reduce voltage drop over long runs.

I've also got the designs of the PCBs modified sufficiently to help out a little in that regard. Thanks for the input so far, I honestly do appreciate it. It's been a fun project and we're looking to improve the design in any way that matters in the long run.



http://kck.st/2z5hhzw

[coppercone2]

i soldered my LED light strips I did not see a point in connectors if you use reliable LEDs and a good power supply. the main failure mode is probobly gonna be cracks on the PCB anyway so your going to have to fix it while its mounted with a soldering iron

shady interconnects are a possible fire hazard. a good solder joint is probobly going to be pretty good.

https://www.digikey.com/products/en/cables-wires/single-conductor-cables-hook-up-wire/474?FV=1440045%2C10480004%2Cffe001da%2Cf80015%2Cf8001a%2Cf80009&quantity=0&ColumnSort=1000011&page=1&pageSize=25

https://www.digikey.com/products/en/cables-wires-management/heat-shrink-tubing/483?k=adhesive%20heat%20shrink

[PlainName]

Your average LED user wouldn't know one end of a soldering iron from t'other. Come to that, I am not exactly a n00b when it comes to soldering, electronics or LEDs, but I'd prefer a simple clip-on connector to make like easy (now and when it's inevitably moved). A good-looking system like this would be preferable to an indestructible but fugly lash-up. (Hmmm, can a lash-up be indestructible?)