Selling models that have never been printed

I can't print a copy of every variation of every model I've made, so, if I offer them for sale, I would be leaving it up to customers to be the first to try printing some of them. I imagine others do this already. Does it cause problems?

Yes that's problematic, because there's always the chance that something is going wrong.

I also have a lot of items in my shop and can't do test orders and test prints:
- that's to costly
- and it needs to much time

Sometimes I have some users which are willing to do the test in their order.

But it's still an open problem how to test pieces effectively.

Woody64

I normally just put a note on the bottom of the description saying that i have not been able to test the model. I figure if someone DOES buy it, then emails me with the problems, would willingly fix the problem and offer the model to them without a markup.

You dont want unhappy customers, you have at least try to make them feel like you solved the problem, even if they dont order the model a second time.

It depends a lot on what is being printed. An art object, for example, is generally tolerant of small errors. For stuff that has to fit together or with something else, it can be critical. Since all of the stuff I sell is in the latter category, I always order everything in my shop before putting it up for sale. This means that I am always my biggest customer by an order of magnitude. But I don't want people to get upset that my puzzles don't work.

Thanks for the advice. The variations I make to my basic design are not likely to affect the printability of the model, I don't thnk,, and i have tested the base model, so I'm hopeful that no problems will arise. Still, as suggested above, it would be a good idea for me to note in the description whether or not the model has already been printed successfully (assuming I do end up offering untested models). Thanks again.

Hello, I'm usually dead keen to make my models public, because I'm excited about them, I would never put a markup on a model that I hadn't paid for at least once myself, after all, why would I expect a stranger to buy something that I don't even think is worth having. And in fact, I'd rather debut a model along with a nice photograph of the finished article, rather than a rendering or something. That's very hard to trust and risks devaluing the process and the product. Waiting a few weeks isn't so hard.

I think there's actually some danger associated with having a flood of untested designs for people to spend their money on - they might become disenchanted with 3d printing as a whole and that damages the entire industry, or might come to see it as only suitable for frivolous, novelty items. In this sense the shapeways site (a bit like thingiverse) hinders attempts to differentiate on quality or investment. It levels the playing field very much, so a random 3d doodle of a gadget that might or might not fit together has just as much weight as a thoughtfully designed zen box that has money, time and attention invested in it.

I apply the principles of a responsible designer, and a responsible programmer: Make changes, improvements, but put your money where your mouth is and accept that your design is an extension of yourself, and you (and your industry) will suffer if you do not live up to your promise.

Saying that, as they say in software testing, if you see no risk, then don't do testing. If you know enough about your process and your own fallibility, and you can conclude that adding an extra hole won't spoil your design and won't spoil the experience of the buyer (it won't surprise the buyer who saw photos of the old one), then I don't see any reason to do a test just to tick a box.

I'm rambling now. I'll stop!

Sandy Noble

Almost everything I draw is something I would want. Sometimes I have random ideas and offer them to who ever wants to print them. The ones I really want I order. The ones I offer that I have not tested, either don't require testing, or are noted that they are untested, and if someone decides to order it to please let me know how well it did/didn't work. I 99.9% of the time release it before testing it. Recently with my ball joint test I didn't release until I tested it, and I have a design soon to be on it's way that will remain unreleased till I get it, because I hope it will be big. All great inventions were never proclaimed before they were actual. Imagine if Edison sold light bulbs before finding the right filament and than found out it was impossible. I think some would be mad.

I had this happen with my turtle earrings when I forgot to put them on show only.

I was still waiting on my own order for them in gold plate when I received a mail that someone ordered them in stainless.
It got sent a week earlier than mine too! I got really nervous about it. They are really small and so their tails/earring loops are very, very thin. Making me worry they might not close properly or something and now my customer will get bad earrings and started checking the forum a few times a day to see if someone might post pics of them.
I was very relieved when I got mine and the tails work just fine.

I don't think I will let that happen again, I will put mine on visible to all and add some renders of the model to the product page but only available to order for me until I have it in my hands.

I might make exception to anything that is well within the limits for the intended materials and stable because of simple budget reasons.

But never again with any model I'm "exploring" wall limitations with or are otherwise uncertain about ^^
It's just a hobby at the moment but I do want to be professional about it and agree with Sandy and your "rambling" was a good read to be honest

Would it be possible to design safer things that do not have to be tested before selling them? To have two categories of things:
a. dozens of designs that will sell and b. a small number of experimental things that push the envelope.

Or do you feel you must test everything because you owe it to your customers? Or do you feel that Shapeways does not deliver the quality or consistency you need?

I do test prints for a few reasons:
1. Printablilty - does it survive a print without getting distorted or broken, does it polish well etc.
2. Presentation - so I can get photos of the item. This also covers differentiation (from the many products that aren't presented well), and proves that I'm happy to invest in my own stuff.
3. Design quality - is it actually as good, once realised, as I thought it would be (things look very different on screen than they do in your hand) - _does it work as an object_ rather than as a 2d pattern. Have I got the scale or proportion wrong, are the points going to catch on things, etc. For stuff that fits together this'd be really important.

Point of fact is that I don't test everything - with co-creators it wouldn't be possible, but I have already tested that the basic design worked. But this too is open to problems - I had an issue recently where I had an order (from outside shapeways) for a really small polyoptic ring, about half the size of an "average" one, and once I'd had it printed, it was had a very different quality than a larger one. Not just in the sense that it was smaller, but that it seemed so much cruder because of the resolution of the print was relatively lower, and it hadn't been polished very well because (I guess) it was hard to hold since it was so little. It wasn't all bad, but my heart really fell when I saw this little lump of metal drop like a rotten tooth in my hand, instead of a big lovely shiney futuristic cloud like I expected. I expected just a scaled down version of the other ring, but the process doesn't scale like that, and I didn't (and don't) know enough about it to know when things will work or won't.

But it did mean I had a reason to go out and buy some good polishing stuff and eventually got it looking nice, but it taught me a little about what caveats I should add when making promises about being able to make things "in any size", and what the limitations of the process are.

For metal, I think consistency is a problem, but that's not something we can really test for. I ordered two rings a few weeks ago, and I imagine they were printed in the same batch, but one is good and one isn't so good. Not bad, but just not as good as the first. I'd be _really_ interested in being able to specify a finish quality (how polished), and wouldn't mind paying a little bit more for a good silver grey print, well polished. If I have surface details I realise this is impractical, but I don't.

In the plastics, I'd have no fear at all about consistancy: Everything I've had has been outstanding.

So the key word here would be predictability?

If a rendering would tell you exactly how it looked, would that be adequate?

Or will you always have to touch something you sell?

Joris

For me it depends. If it has to fit something in the real worlds, I want to be sure I'm not off by a fraction of an inch before i sell it. But if it's stand alone than it's not as important, such as a ring or bracelet.

I admit at first I was confused about how I feel about this selling stuff without seeing it - I was still in the mindset of "I personally am selling a physical product to a customer", which left me feeling a bit nervous about trusting someone else with such a big piece of the process.

Very recently I suppose I have come round to realise that I'm not selling an item personally to the customer, I am selling a design to shapeways and it's up to you guys to deliver on that promise! I don't feel as nervous about it now, because I don't think getting an unsatisfactory print would reflect that badly on me personally. And actually, because of the overall low cost of printing, then I'd imagine it wouldn't even reflect badly on anyone.

I'm certainly happy to sell without holding things in my hands first for that reason - the risk is split, in fact, you guys seem to take all the risk!

But you're right Joris, for me, (and again this has only ever been an issue in metal), it is totally about predictability. If I know that prints will always be yellow and rough, then I can do the sums and make an informed choice about how much I need to charge for the post-production (or even if I can do it at all). If I know they will always have that exquisite polished fingerprint texture and the delicate, almost imperceptible champagne colour, then I can try and work out how much that's worth as a feature. If I can confirm it will always be grey and satin, then I can stay where I am. I would be very happy indeed if I could get something like you say, where it'd be polished to a pre-agreed level.

I've deliberately attempted to steer clear of subtle surface details so that it can be more easily polished, and it's true that it's the models with edges and corners that come out roughest - I assume because the person doing the polishing doesn't want to grind all my lovely sharp edges away. The models with large curved surfaces are usually slightly more polished, and better looking.

Cheers!
Sandy Noble

I'm with YouKnowWho4Eva here; if it has to co-exist with something in the real world then as a designer you'd better damn well order it and test it. Moving pieces (puzzles) should be ordered before being sold, too. It all comes down to what it is you are designing/selling. Most of my pieces are jewellery which I feel has a lot more leeway when it comes to "testing". I have ordered a lot but I cannot possibly afford to order everything I sell in multiple materials...

Glenn

But a character model for example or a vase, would that have to be tested also?

I'm trying to discern what is a lack of confidence in the repeatability of the production process from what is just pride in one's own work.

I know much more about the materials and things now than I used to, so I'd have far fewer qualms about not testing. I have a much better idea in my head about what something will be like, what it'll look like anyway. How something feels, and how flexible it is, that kind of thing is a lot more difficult to predict, but I'm getting there.

Again, in plastic I'd be doing tests out of curiosity rather than necessity (or to take pictures). I haven't noticed any variability in the quality of the plastic prints at all. If I was going to do some stuff that relied on different colours or different materials that I hadn't used before, then I'd want to do a test so that I could accurately describe the finish or the item or the colour - Colour is very subjective, and what you guys call black (black strong flexible) I call purple. I also think the interesting qualities of 3d printed stuff means that it can look very different in different light, and that's one of it's selling points for me, so I like to be able to write about it as a feature.

Metal is a different kettle of fish, it's more hit and miss. I'd be testing in metal to make sure it was printable without distorting, and ultimately for comfort when wearing.

If I was making something that didn't have to "fit" something else (like a character or a vase), then I would be happy not to test. Unless I was unsure about how certain tricky details were going to show up. If I'd avoided tricky elements I'd be very happily confident. I'd still be curious of course, but if someone commissioned me to come up with a design, I wouldn't necessarily need to hold it in my hands before I signed off on it.

Parable: I designed some very simple, very basic hollow lozenges type objects to be used as tests for some dyes I got. I hoped they might be developed into pendants or earrings eventually. Everybody I showed the design to saw nothing unusual. Got some printed in WSF. The instant I saw them I realised they looked _exactly_ like tampons, and I haven't been able to take them seriously since. Rather embarrassing... Luckily I spotted the similarity before I took lots of pictures and put them everywhere...

As they say: "many a slip, twixt cup and lip".

Sandy

Some great points here. Testing is important, and is absolutely essential with functional items, but there are some things we just can't predict. It is not reasonable to order a model in every material, then order (and photograph) all over again every time a change is made. We are still not guaranteed success with risky models due to production variations. I think we are covering our bases by disclosing everything a potential customer might want to know before ordering a model, and doing what we can if there is a problem. Fortunately Shapeways has excellent policies in place to help us keep customers happy, and a lot of care and attention is given to avoiding problems before they happen. One feature I would like for model pages would be a separate tab or a "More>>>" drop-down extended description that shows product specifications, including its print history.

I think our obligation as designers is to make fairly safe design decisions to avoid gambling with our customers money, and if we want to innovate and experiment, to make sure the customer is aware and is a willing participant. I have been making my models available before I receive the print, and I'm now comfortable doing so as long as I make that clear in the description. A model does have lower sales until the photos go up, but the fact that a model can be created, printed and delivered in such a short amount of time is one thing that sets 3D printing apart, and to close off that option removes value from what we are offering.

One Shapeways user put his faith in me and ordered my Time Keeper, even before I did. His son was born a few days after I released the model, and he ordered one to commemorate that. Our stainless-steel Time Keepers shipped at the same time, and depending on his location he may have actually been the first to receive it. There is a whole lot of magic happening here that would simply not be possible if no one was willing to take risks. I took a risk, Shapeways took a risk, and the customer took a risk. We were all richly rewarded.

Happy Birthday Royce.

Personally, I don't open anything up for public sale until I've tested it for several reasons.

First is quality control (mine, not Shapeway's!). I'm overly conciencious, and as I'm usually working on the bottom end of the minimum detail/thickness level (as I'm doing model starships), I'd rather do any testing first myself, concerning printability issues. I've had several occasions where I've missed thin parts (usually engine exhausts), and I'd much rather that happen to me than to the customer (mostly because it's less of a pain for all involved.) (For this reason, I'be started sharply restricting the materials my stuff s printed in, limiting it to WSF and occasionally detail, since those are the specs I design the model to - and the stuff made for WSF doesn't always transate to detial!)

Secondly, it gives me chance to photograph and/or paint the models for the gallery (which gives a little bit better idea of size, important for starsip miniatures).

Thirdly - and importantly since I'm still fairly new at this - it gives me a chance to see and make any adjustments. (That said, minor adjustments I don't worry about re-printing, now I've got my eye in. Fortunately, as I'm designing via a CAD package, I've got an accurate idea of what size small bits are if they need fixing, so on the whole, I don't tend to have too many problems in that regard.)

Finally, there is still a fair bit of "Mine! All Mine! Ahahaha!" going in the back of my head with regard to starships, and on top of all the previous, I still have to be able to satisfy my "Me First!" Since I will of course actually be using everything I produce myself!

Actually, I can't think of an occasion where I'd design something that I'd not want to prototype myself first, not even with something really basic like say, flight stands.