 Next talk, the title is From Project to Kid, will be presented by Jenny List. She is a director at the Oxford Hackspace in the UK and Contributor of Hackaday, the site which you probably know. Very interesting articles. And the talk is about, imagine you have an electronic design that you made and it works and you're happy with it and you think, hmm, might other people be interested in it? What can I do to sell this, to make the design I made into a product? How can I find customers? How can I sell the stuff? This we will hear in the talk. I'm very curious about it and let's give a warm welcome to Jenny List. Hello everybody, good morning. This is a talk based on my experience starting a small-kit business. I used to work in the publishing business, used to work for the Oxford dictionaries and when I left the dictionary I set up in business on my own and one of the things I decided to do was make little amateur radio kits for the Raspberry Pi. I wrote up, if you want to read this in more detail, I wrote it as a series of articles for Hackaday last year but I think it was worth distilling into a talk here. So, if you want to sell electronic kits, the first thing I'd say is don't give up your day job. There are people who've made very successful businesses out of selling electronic kits but if you get lucky that's great, but in most cases you should expect to sell a reasonable number of kits, get a following, be able to pay for toys, bring in a bit of money but in most cases it will take you a year or two to build it into anything like the sort of business that you might earn a living from. Some people have managed it faster, there's at least one peer person in the audience here who hits exactly the right moment and exactly the right kit but in most cases you'll end up like me earning a few hundred quid a month, something like that. That's probably a few more hundred quid year rows a month and having a little thing to put on your tax bill and be able to pay for your oscilloscopes and toys. So, it's worth doing but don't plan on it taking off immediately. If you have an idea for something that you've made and you've got it on your breadboard and you think this will make a kit, the first thing you have to do is look at the rest of the market, think what are other people doing because it's very easy to think that you've made something awesome without realizing that other people have made something that does the same thing, only more awesome or the market is saturated. I'm going to use as an example the Raspberry Pi. The Raspberry Pi has umpteen little boards with LEDs that are flashed by GPIOs and every month brings news of a new one and it's shaped like a snowman or a Christmas tree or a melon or a banana or something and people keep bringing out Raspberry Pi LED boards for some unknown reason. And I look at those and I think surely a little bit of market research would tell you that the Raspberry Pi LED boards market is saturated but no, people keep doing it. Personally, I had much more of a niche product so when I looked, is there a little radio receiver or a little radio transmitter for the Raspberry Pi, the answer was there's almost nothing. But had there been an established player doing awesome little radios, I would have probably moved on and pursued some of my other ideas. So always check your marketplace. It's market research in any business and particularly in this business. Then you have to think what makes a good kit and how can I make a kit that I would want to build because making a kit is different from building something yourself from scratch in that most of the people who build your kit, they have construction skills but they're not elite engineers and if you give them a kit that they can't build, they're going to ask for their money back, they're going to complain about everything and it's not going to go well. So think about the kits that you made that really did it for you. In my case I always think of, if there are any Brits here about my age you'll probably remember CM House Amateur Radio kits from the 1980s and 1990s and they were brilliant kits, they worked. You just stuck them together, they had instructions that worked, all the components were right, everything. Or older Heath Kit kits from the 50s, 60s and 70s. Fantastic, some of the best kits ever made. I'm not talking about modern Heath Kit, whoever's bought the brand now, I don't know what crack they're smoking but really. The kit pair in contrast with a typical Chinese kit. You can buy a little radio or other Chinese kit on eBay for about two or three quid, five euros, whatever. And you get it, it's got all the bits, it's got a nice printed circuit board but the instructions are in Chinese. And then you find the Chinese speaker in your hackspace who translates it for you and it still makes no sense because the person who wrote it had no idea how to write it. I actually know somebody in the UK in the amateur radio world who's made a nice little living buying these Chinese kits, figuring out some decent instructions for them, repackaging them with decent instructions and selling them on. There's a business idea for you in the kit market. So you've got past those first steps, you think there's a market, I can make a kit, great, I'm going to do it. So you've got your little project on the bench. Of course it works for you because you made it, you designed it. You've got to make sure it works every time for the person who's going to build it. You have to build exactly what the customers will get. It's not a case of, oh, I need that component, I've got one in my junk box. It's a case of you have to buy exactly the components you're going to supply to the customers. Get exactly the printed circuit board made, absolutely what the customer's going to make. And you have to probably build several prototypes. My little radio went through about three or four revisions of the printed circuit boards before I had something that I could sell. I thought it worked every time, and of course it didn't. So you have to ensure that you have a reproducible design. Then you have to think, how am I going to sell this? What's it going to look like? If it's just a sort of nasty little bag of lots of components, then it's not going to be attractive. You have to think, how am I going to present this to the world? In my case, I went for a click seal bag, but I relied on nice instructions inside to present it. You have to think, how am I going to make something that when a person sees it, they're going to want it. It's going to be robust enough that I can bung it in the post and send it to the USA, and it'll arrive in one piece. It won't burst and lose all the components in the mail. And then you have to think about the weight. Of course, you'd take the post office and stick it on the scales and suddenly find it's five quid more expensive than you expected to send it. So all these things you have to think about when you're working out a kit. And then, finally, pricing. What I do is I create a spreadsheet. I list my bill of materials, and I don't just mean all my components. I mean literally everything down to components, packaging, sticking, printing, customs when you order printed circuit boards from China. It's great that I love the people at 30 PCBs. I order 200 quids worth of PCBs, and it comes with $5 gift. But one day, the customs man is going to find me, and I have to be really certain that when he does, I have budgeted in that I'm going to pay the customs charge. I can't remember how much it is in the UK, but it's all on the web, and that's on my spreadsheet. So when I price stuff, I do several different prices for a run of 25, a run of 50, or a run of 100. I don't know what other people do for kits, but I made the mistake with my first kit of making 100, and I've still got about 80. If anybody wants a Raspberry Pi RF prototyping board, I can sell you one, I've got 80 of them. So I make, I price for 25, 50, and 100, and my first run that I see myself doing to sell is 25. Sometimes I end up with 20 left over, other times the 25 go all in one, but at least I haven't lost too much cash if I end up sitting on 20 kits. Now, I'm a wordsmith, my background's working for the dictionary, words are my play thing, literacy is my toy, so I see the instructions as the most important part for kits. They're actually secondary to the kit itself because they're what draws in the customer, they guide the customer through building it, and they're what tells the customer how to use it. You can have the best kit in the world with those dodgy Chinese instructions, and the customer isn't going to like it. We need to have bloody good Chinese instructions, but never mind. You have to keep your instructions simple. I try to pitch it so that a novice can build my kits. Now, all my kits are surface mount, so they have to have some ability, but I assume they've never sold a surface mount before. I have a section on how to sold a surface mount, and the great thing is everybody who reads it says, no, that's the wrong way, I'll do it differently, but at least I'm trying to tell people how to sold your first surface mount kit. Actually, all my surface mounts, they're all 1206s, quite big surface mount, but if you know radio amateurs, a lot of them aren't young. I divide these into sections. My instructions, they start with a friendly intro, sort of welcome to your Raspberry Pi Amp to radio kit. This kit, when built, will allow you to listen, blah, blah, blah, blah, blah, blah, blah. It's just like the introduction I'd write for a Hackaday piece. It sets the scene. It doesn't give you anything useful, but it draws you in. I have a nice color picture as well at the top. I then describe the tools and techniques. You will need tweezers, a soldering iron, a magnifying glass, et cetera. At that point, I also have my how to solder surface mount devices. Probably most of the people in this audience wouldn't need that, but it's a necessary section. Then I have a step-by-step guide. I go to town on the step-by-step guide. I have pictures, diagrams, screen captures from my PCB program with big red circles round. First you solder R1, then R2, then R3. I normally break it down by three components at a time, step-by-step. It doesn't matter if it's two or three pages of that, if they can go through it bit by bit. That's what people always say makes the 1950s Heathgates, that you had this wonderful step-by-step instructions that had everything laid out. That's what I've tried to emulate. I've had plenty of people complain about different things, but I've never had anybody complain about the instructions yet, so I hope I'm getting it right. Then I have a setup and troubleshooting. In my case, that's normally plug it in. Is there smoke? How many amps does it take? Okay, it probably works. Plug it into your Raspberry Pi. But, for instance, here you'd have all the, you know, you expect this frequency here, or this current here, or tweak that coil for a dip here, or whatever you do. You have to yet again write this in an accessible format. Try and imagine that somebody's never done this before. Then finally I have appendices. Here's the circuit diagrams, the technical data, and stuff like that. And finally I have a declaration of CE conformance. This is a very useful thing to know. If you are making a kit of parts for amateur radio, there is an exemption on the EMC. If you ever dealt with EMC for commercial electronics, you know, it's paying the arse, and you have to spend thousands of dollars on a lab. If you're making kit of parts for amateur radio, then you're completely exempt from all that. And I'm clinging to that exemption. So I have a CE conformance. A lot of kit makers don't bother with that, but I figure that since I do some of it for a living, I ought to do it with my own kit. And then finally, give it to some friends. Give it to people who know you, who've seen you making the kit and stuff, who know the kit. Make sure that they can follow the instructions too. This is just your review thing, just so that you've got some feedback, so that you know it works. You'll end up with a whole load of little radios, or your friends will end up with a whole load of little radios, but at that point that's your first making sure that it's not just you thinking your kit is wonderful. If other people can make your kit, then you're probably on the right lines. Now, when I say it's all in the presentation, I mean, coming back to how the kit looks. When they empty the kit on the table, what do they see? If they just see a pile of bits, they're going to have problems. So I label everything. So little pieces of tape with three surface mount resistors. I'll be there with a magnifying glass and a magic marker, just carefully writing 10K a hundred times on tape before I cut it out. I actually always give an extra component. So if I need two resistors, I will give them a little tape of three, because with surface mount components, they're going to lose them anyway. I bag up some components, because some manufacturers, if they supply it on black tape, black marker doesn't show up. So I then have to use a tiny little click seal bag with a printed label on it, which is a royal pain in the backside. But when you have, say, surface mount inductors come into black tape, that's what you've got to do. And then, when you're kitting up, I do it on a production line basis. I will do first R1, then R2, then R3, and I will carefully prepare the right number of each component. I'll have a pile of empty bags, fill pile of full bags, and bit by bit, I'll build up the kit. But you have to be very careful to make sure that every kit has the part. Now, that might seem like an obvious thing, but it is astoundingly easy to end up with a kit going out that hasn't got a resistor or something. And of course, the customer, they haven't got a 1206 0.1 microfarad capacitor. Some of you lot might, but your average radio ammeter certainly hasn't. And at that point, they can't make the kit. So this is a vital part of quality control. And then, finally, coming back to how it looks. I know I'm harping on about this, but this is how you flog your kit. Design the label. I have a color printed sticky label on the front. It's got my company logo at the top, and then it's got big letters, the name of the kit, and then underneath a little CE mark, and this is not for children. Children will eat these resistors, just the standard stuff, but I'm trying to make it very clearly. This is what this kit is. And behind that, I fold my instructions. So the color photo is on the front, and I always stick the printed circuit board in the front so they see a picture of the finished device and then a nice green printed circuit board underneath it. I don't want them to see a great mess of bits of SMD tape. I try to get those behind. I want them, when they pull it out of the envelope, to think, wow, this is a really nice kit. Later they can find out what it really is, but I want that first impression. Oh, yes. When you've got a pile of kits, the first thing to do is pull a random one out. Pull several random ones out. Count the components. Make sure that you've got enough. Yet again, it's something I'm harping on about, but you never want a customer to receive a component, a kit that's short of a component. Then find a friend who's never, ever seen it. Even know you're making a radio kit or whatever it is you're making. Give them one and just say, build this and come back to me. Don't give them any help. They are a simulation of a real customer. They're not the hackspace mates earlier. We've seen you through all the starting of building and designing the thing. They have never heard of it. Find somebody like that and if they can make it, then the chances are your customers can. Fortunately, I always... I never failed at that point. Every time I've given a random friend a kit, they've always come back with a working kit, but one day I'm sure one of them will come back and I'll have to go right back to the beginning. Then your pile of kits, these are your stock. These are the things you've got to flog to your customers. You've got to take care of them. Keep them tidily. Keep them away from dust. Don't let them get crumpled. If a customer gets a kit that looks like it's been through the wars, they're going to think, hey, this is a customer return. I don't want this. These are the things that earn you money. You've got to take care of them. Then we get to you've got a pile of kits. How do you sell them? First way is to sell them direct. Go around hacker camps like this one. Sit at the table in the hardware village and try and flog them over the counter. Make affairs, radio rallies, hack spaces. For that, you need all the accoutrements of a stand. You need a table. You need a chair. You need a cash box. You need vast amounts of energy. You probably need a larger state car and you need to be able to exist without sleep. Personally, I don't do that because I can't keep up with it. But there are other people who do it. You'll see them at events like this and they make a very successful business out of it. Certainly the easiest way from the point of view of just going out there and doing it, but it's probably the most difficult way in that it's bloody tiring. Then online. At this point, I have to do some responsible disclosure. I'm contributing editor for Hackaday.com, as has been mentioned. The sister company of Hackaday.com is called Tindi. I'm going to mention Tindi because I think it's a good service, but it's not just some sort of paid shill because I work for the sister company. Tindi is best described as Etsy for makers. It's an online marketplace. I don't use it for various reasons, but it's successful for those who do use it. There's at least one Tindi seller in this room who is happy with it, I know. But as I say, responsible disclosure, I would promote Tindi, of course, because I work for Hackaday. Alternatively, you can set up your own online shop. That's what I did. I used Shopify. Total turnkey shop. I've done online shops in the past from scratch, things like Magento, and I will never do it again. If you're tempted to do your own online shop, take a reality check, take a sanity check, go with a hosted service, have somebody else who knows what they're doing, take the strain. Unless you really are an elite guru in online shops, put the idea room back into a real slow. I find Shopify to be great for me. It's got complete integration with stationery, royal mail, mailing services, reminders, the works, and it only cost me $20 a month. But your choice of online shop is obviously up to you. It's a personal choice. I'm happy with Shopify. And then crowdfunding. Everybody's come across Kickstarter. I actually tried to crowdfund my first product. It was the Raspberry Pi RF prototyping board. And it failed. I made a mistake. I thought I could make 200 of these. I can sell 200 of these. And I had 160 backers. But there's a lot to be said for failing in a Kickstarter, because, of course, you don't have to ship the product. And in fact, most of my backers came and bought one anyway. A good chunk of my backers came and bought one. If you are tempted to go down the crowdfunding route, Google for crowdfunding horror stories. There are many. The chief one is people woefully underestimate how much everything will cost in a massively out-of-pocket. I don't think I did that. I think if I had succeeded in my crowdfunding, I would have been able to ship my 200 boards. But I would say beware if you go down the crowdfunding route and read up very carefully on it, because there's been some spectacular cockups in that market. And then, finally, you've flogged a kit. Fantastic. You've sold a kit. You've sold it to the customer. You want it to arrive at the other end in perfect condition. And, of course, it's going into random postal services all over the world, and you have no idea what will happen to it. It could be run over by a truck, anything. You've got to make sure that you have the best packaging you can afford that will fit within the postal class that you want it to be. And you need to have the best courier that you can find. It's a classic thing with large online stores that they think will cut costs. Instead of using Royal Mail or something like that of DHL, we'll use some white van man company. And, of course, the customer, they don't see the wonderful shop. They don't see the wonderful logistics, whatever. They just see the bloke who delivered it late or bunged it in the neighbor's bin or whatever. The delivery person, the postman, is the personal face of your company. You are not the personal face of your company. You can have the best kits in the world. And if the postman screws it up, then as far as the customer is concerned, it's your fault. So, I always use Royal Mail because in the UK, Royal Mail are good. I don't know if your mail system may vary wherever you are. I use puffy envelopes, jiffy bags because they protect my kits. Some people use those card boxes, roughly the size of a video cassette kind of boxes. I would say always do some tests, send yourself some packages, send some friends on the other side of the world a package with a kit in it, see how it gets there, and then make your choice based on that. Yes, that's the other thing I do. I make tracked and signed a feature. If you just send it, there's a chance it might arrive, but you have no idea. I charge a little bit more for postage and packing, but all my customers know that they will get a tracking number and they will be able to see where their package is and they'll be able to see who signed for it. I consider that to be worth the extra for you, Quid, and Touchwood, so do my customers. When it comes to actually packing a kit, I have a checklist, and it's very important to have a checklist because you can miss things. My checklist is capture payment, get the money from the credit card people, pick the kits, choose the right kits, and you have to check this against the order that you get. Pack it, put it in the envelope, stick on the address sticker. I use a combination stationery where it prints all in one go with the address sticker on it, and the number of times early on that I folded up the invoice, stuck it in the envelope, and then sealed the envelope with the sticker still inside it. That taught me the hard way. So yes, pack, stick, then seal. Well, okay, pack, stick, check everything, then seal, and then send. And then once I've sent it, I get a receipt from the post office. I do it manually, I'm like going down to my local village post office because it's a hub of my community. And then I get the tracking code and I bung it in the Shopify website and the customer gets an email, job done. Then I probably never hear from the customer, and that's a success because the customers you never hear from are the ones that have gone away, and are happy. And probably about 99% of my customers I never hear from, and I hope their radios work. But obviously you want people to find you and sell stuff, so you have to become a marketing genius. I approach this by going on social media and forums. Now, I didn't approach it by spamming my kits. If you know about me, you probably don't know I sell kits because I don't spam about kits. I took the approach of trying to be a helpful expert. If somebody was talking about using the Raspberry Pi as a radio thing or a transmitter, I would come in and try and talk about it with them, try and be a helpful person knowing what it could do and stuff. And I know I picked up quite a few sales like that because every time it wasn't that I said, by the way, I sell kits. If I was on a forum there, it's my signature with Raspberry Pi, my radio kit or whatever. Or my Twitter bio has maker of Raspberry Pi radio kits. And it is a bit organic. It wasn't quick. But I like to think I didn't piss anybody off along the way by spamming them. And then finally, you've flogged the kit. You've got to provide support. I have a mailing list. It's got hardly any people got on it because most of them mail me direct, say, my kit doesn't work, or this resistor is missing. But you've got to be 110% helpful. You've got to bend over backwards and do anything they want. And if that means send them a new kit, so be it. I've had one customer where he had a kit he couldn't build. I think it was because he was an old guy and he couldn't solder very well. I sent him another kit and he had problems with it. So I actually built one for it and sent it. I don't normally build kits for people because of that CE thing about selling kits. I don't sell ready-made units. I sell kits and that keeps me within the CE law. But in his case, it was a nice bloke and he was really trying. So I actually sold it up the kit and sent him. So I lost money on that one. But the great thing is he'll go to his amateur radio club and he'll tell everybody that Jenny's kits are awesome. She's got really good support. So I thought that was a good investment. And he was a nice bloke as well. He wasn't nasty about it. I think I've had a lot of nasty customers just refund them. That's what I do. I've even had the old nasty customer who's obviously spoiling for a fight and I say, how fine does we refund you? And then it's like, I still want to argue with you. Sorry, mate, I've refunded you. You've got to be prepared. Sometimes you lose money. It's only probably less than 1% of my customers do this. But if you get a customer like that, you just have to be nice to them, refund them, move on. So how have I done with my kit business? I've been selling kits for about three years. In my first year, I certainly didn't make enough from kits to survive. I was doing various other things. Second year, it was starting to pick up and I reckon I could have got somewhere with it. And then of course I got Hackaday. And Hackaday became a full-time job and makes me enough money to live on. And so in a way my kit business has never reached its full potential. I sell the kits, I send them out, but I'm not doing all this sort of hardcore marketing stuff that I like to. So I think it's possible with this kind of business to make a business and I think by now I will be making a living. Probably not as good a living as I'm making from Hackaday. But I probably would have just about given up my day job by now. But after three years you've got to be prepared to basically live on ramen for three years. Hang on, you're at a hacker camp, you're prepared to live on ramen for three years, aren't you? But anyway, that's Project Kit. I hope you found it interesting. If you have any questions, then I'm here. If any of you want a Hackaday sticker, I've got some Hackaday stickers, otherwise I've been bunging them all over the table in the info tent. You can find the stuff I write at this URL, just hackaday.com, and then find my author thing. Or on Twitter, I'm Jenny Unscore Alto. So, does anybody have any questions? Thank you very much, Jenny. Very interesting talk. I enjoyed it a lot. I'm sure there are questions. Go ahead. Hello. I'm Axel. I have a suggestion and a question. I have a suggestion. If you get S&T components in a black tape, use a white pen. You have very small white pens, like fine liners. Funnily enough, I looked into that. White pens are really expensive. I don't know what this is in euros or dollars, but they're about 25 quid. I just thought, no, it's actually cheaper to put them in little bags. But yes, good point, yes. I thought for CE certification or approval, I thought you needed a complete project, like product, because otherwise, how can they test it? That it's certifiable, RF, et cetera? I think the point with the exemption for kit of parts for radio amateurs is that in doing the radio amateurs exam, the radio amateurs is deemed to be an expert in radio who, when they make it, can make sure that it doesn't cause the radio interference. The rest of the CE marking, I self-certify, which means that I go through, get all the documents for all my parts, all the data sheets, and I rely on the fact that they're CE certified, and if I ever get audited, I can turn around and say, look, here's all the data that says all the bits I use are CE certified, therefore the whole is CE certified. But yes, that exemption, kits of parts for amateur radio, is a godsend when it comes to making something like this. I couldn't make, for instance, the product I'd like to make for the Raspberry Pi but can't is a low-pass filter to make a transmitter for the FM broadcast band, not legal at all, but aside from the illegality of that, I mean, I should sell one because of course these people who just stick a bit of wire on it are radiating harmonics everywhere, but even if I could sell one, I couldn't sell it because it's not for radio amateurs because of the CE marking. Anyway, sorry, go on. Good morning. Thank you for your talk. How do your customers find you? Do you know that? A lot of them find me by Google. A lot of them find me from forum posts, things like qrz.com and things like that, radio amateur stuff. A lot of them find me by word of mouth, friends, Twitter. I, as I say, having sort of rather sort of put this business slightly in the background since I started writing for Hackaday, probably I could have done more to get people from more places. But it helps that mine is quite a narrow niche. There aren't many people selling radio kits for the Raspberry Pi. It makes it easier for me to be found. If I was selling LED boards for the Raspberry Pi, I suspect I would have sunk without trace because if I was selling those, I'd really have to work at it. But it's the tiny niche that helps me there, I think. I don't know if I missed it, but where can we find your kits? Good point. Okay, my company is called Language Spy Limited. The reason for that is because having worked for the dictionary, I thought I could make it in the world of computational linguistics and there's a really nice corpus analysis system which may be no money whatsoever. So shop.languagespy.com Thanks. I was wondering, do you use genuine components or do you buy them from China on Aliexpress and sites like that? I use genuine components. That comes down to you want the customer when they build the product for it to work. So I order my bits from Mauser or Digikey or Farnel, sometimes even from RS when I'm really in a hurry. Yes, it's a bit more expensive, but I know that my customer, when they sold that part on, it will work. Thank you. Anybody else? More questions, maybe? Do you sell just as one kit or do you sell a lot of kits? Because I sell one really niche kit, that's definitely not making any money from me at all. So it's just curious if you sell a lot of kits or just this one thing? I sell several kits. I have a couple of little modules. I have the Raspberry Pi development board. I have a little direct conversion radio receiver for the Raspberry Pi. Basically, anything that uses the Raspberry Pi's clock generator is an oscillator. And I also have a range of low-pass filters for low-power transmitters for the Raspberry Pi. So if you use the Raspberry Pi as a whisper beacon, you might use one of my little boards. Ah, my shill. No, I actually just thought of this question. So there's a thing you didn't talk about, which is deciding what kits to make. What about that? Well, I suppose it comes back to my market research as well. This is a good idea, but to be honest, I didn't put the huge amount of work into that that you might think because I thought, I've made this, this is really cool, I can make a kit. So you're right, it's something to really think about. I probably didn't think enough about it. It comes back to the things we write for Hackaday. We write stuff for Hackaday because it interests us. And similarly, I made kits because I wanted one and it interested me. So that's probably a bad answer, but I'm afraid it's what I did. If you were doing another kit now, would you use the same method or would you do something more? It would depend. I think I would probably still go with something that I wanted. Based on the idea that if I'm enthusiastic about it, then hopefully people like me will also be enthusiastic about it. I appreciate that isn't a solution that will work for every endeavour, but it has worked for me. Well, at least it proves that somebody likes it. Yeah. Well, questions? Thank you.