Nicolas Deverge
Transcript (Translated)
I’m going to present you with a feedback on a small project I worked on in the evenings and on weekends. It’s a small side project I did on my couch. I’ll talk about Lean Startup, Fab Lab, lots of things, and all that. I’ll start by introducing myself. My name is Nicolas Deverge. I work at a Toulouse-based company, I’m from Toulouse, called Equito, where I do Lean Startup. We have a start-up accelerator, and I support entrepreneurs in their entrepreneurial approach.
By helping them be a little more lean and all that. I’ll explain a bit about that. I’m also a programmer; I continue to do programming. I also did agile coaching for a few years. And I’m a micro-entrepreneur, meaning... It’s almost become a tradition—every year, I launch a small project. Each time, it provides an opportunity for a bit of feedback to share at conferences. I gave this presentation at the Agile Tours in Montpellier, Toulouse, and Bordeaux over the last 3-4 weeks. I hope I’m a bit more polished.
So, like any project—actually, any product—for a product to work, we’re going to talk about the product here, for it to work, it needs to solve a problem. Now, that might seem a bit obvious when said like that, but it’s quite important—if you create a great product with... Amazing features, really well-crafted, super ergonomic. If it doesn’t solve a problem, no one will be interested. What we call a 'nice to have.' It’s good, it’s nice, it’s fun. People will play with it for five minutes, they’ll try it once, and then they won’t come back. So, the point is, first, you need to ask yourself: as an entrepreneur, does my product solve a real problem? I’ll come back to that. Later in the methodological approach. For me, the problem was last Christmas, 2013, during a family meal with my in-laws. My father-in-law lost his keys. He had lost his motorcycle keys; he had lost them in the street. It’s not the problem of losing your keys when you’re about to leave. Where are my keys? What did I do with them? Are they on the couch, under a cushion, or what? No, no, in this case, it’s really losing your keys in the street. So, I found myself thinking, here’s a great opportunity to find an idea, to create a product to solve this problem. And above all, a great opportunity for me to implement Lean Startup once again. I had already done it on a previous project. I do it with entrepreneurs, but here, I thought, this is it. So, the idea of the product, of the project, was also to have this methodological approach. So, I’ll give you a theoretical presentation of Lean Startup to start, and I’ll explain it through my experiences as we go along. So that’s the problem. Right away, I’m a technician—I’m an engineer, I come from the technical world, from a technical background—so I immediately think of a solution. First reflex. It’s not necessarily the best reflex. Even when entrepreneurs come to see me, the solution can wait; what’s important is something else. Of course, I don’t apply at all what I advise, but that’s normal.
So, the solution—I had a brilliant idea. A keychain with a code to contact the owner. That is, you attach to your set of keys a keychain with a small code that is unique to that keychain. And there’s a website address. The person who finds your keys in the street goes to the website and enters the code on that keychain. There’s a form where they can leave a message, and it sends an email directly to the key owner. There’s an association between the code and the email. But the person who finds the keys doesn’t know the email or anything. It’s secure. It’s reassuring. So, it’s a pretty simple solution. So, I thought, great idea, I’m going to be a huge success, it’s too easy, the solution is super simple. Anyway, when you look at simple solutions, they’re always the most effective, we know that. So, I thought it was going to be really fun, it was going to be great. So, I set a few goals for this project. First, I’m not going to quit my job to do just this. It’s really a small weekend project. And I wanted to enjoy it, meaning I didn’t want it to become a constraint, for this project to become a constraint. I wanted to be able to work on it when I wanted. It’s not something that would require me to... Get up in the morning, do this, and so on. No, it’s really when I have time, or rather when I feel like it, I work on it. That was really important. Why not make a little money with this system? I wanted a very minimalist investment, so an investment of 150 euros from my own pocket, really, not counting the time spent—I’m only counting the investment in cash, let’s say.
And so, as I said earlier, it would allow me to test the Lean Startup approach.
I’ll give you a quick and theoretical presentation of Lean Startup, just to set the context a bit. Lean Startup started in 2010 with Eric Ries, who synthesized a whole set
of ideas or methodologies that came from the work of Steve Blank, who is more of a marketing person and had written *The Four Steps to the Epiphany*, the first book, which is a bit about customer development and things like that. So, in 2010, Eric Ries started talking about Lean Startup and released a book called *The Lean Startup*, which synthesizes all of this. A startup, in fact, isn’t very complicated. It’s a bit like agility—it’s not complicated, it’s easy to understand, but it’s absolutely not what we traditionally do. Completely, right. So, it’s not complicated. The best illustration is this—it’s a bit like explaining how it works. I won’t go into the theory and all that; it’ll just be very light. So, in fact, we talk about an idea. I prefer to use the term hypothesis here. In fact, we start with a number of hypotheses.
Everything is a hypothesis—that is, the problem is a hypothesis, the solution is a hypothesis, the clients I’m targeting, the client segment, is a hypothesis.
In fact, when you’re a lean start-upper, you consider that all the ideas you can have,
Everything you think of is just hypotheses.
So, we’re really in a state of perpetual questioning. We’re not full of certainties, like when we often say, 'I have the idea of the century, it’s going to work, I’ll write my specifications, I’ll lock myself in my garage, I’ll work for two years,' I’ll release my product, I’m so sure it’s going to work, I release it, and nothing happens—it’s a flop. We’re not at all in that approach—completely the opposite. So, we start with hypotheses, and from these hypotheses, we build a product. When I talk about a product, it can take many forms. In Lean Startup literature, we find the term MVP, Minimal Viable Product, which is a minimum viable product. That is, it’s a product that is really minimalist in terms of... effort to build, or even in terms of functionality. It’s really the product that delivers the promised value to its customers very quickly. So, often, I make the connection here.
So, I’ll continue with the rest of the segments if needed. Agility is where we are agile, we use Scrum, Common, and all that, it's the build phase. Now, we're moving into agility. The Lean Startup aspect is really an upstream phase to agility. It's the ideation side, working on the idea and all that. The agile part is rather over here. So we build a product, and in that product, we've included a certain number of measurement probes. That's super important. That is, when we design the product, we make sure we can collect data to learn from how our customers use it. To know if the hypothesis we initially made is validated or not. Let me go over that again. We're going to build a product. It's an iterative cycle. We will conduct experiments. On which we will take measurements, observe how the product is used to learn whether it's good or not. So, it's not a survey, for example. That is, we're not going to do a survey asking, 'Would you do this, that, or the other?' No, no, we're going to put the product in the hands of customers, see how they use it, and from the actual usage in the field, we'll learn things. Because often, people say, 'Hey, my keychain, look, I'm showing you, it looks good, when are you going to sell it?' People will never buy it. That's how it is, meaning there's a gap between what people say and what they do. And it's natural, that's just how it is, we can't go against it. So the idea here is, regularly, for the different hypotheses we might have, really go into the field, actually release something, and very quickly put it in the hands of users to validate the hypothesis.
So to get started well, there's a tool called the Lean Canvas. This comes from Ash Maurya's work, who wrote a book called *Running Lean*, which, after Eric Ries, is for me the second guru on Lean Startup. So actually, the Lean Canvas is simple, it's a... So it's a canvas, for example in A4 format. I won't go into the details of each box, but it really helps synthesize your idea or product a bit. So the most important boxes are first the problem, as I was saying earlier. What are we trying to solve? After that, generally what we try to see is who we're addressing. That is, a problem, we address a certain user segment. Another important box is how to address these people? Where are they located? How am I going to find them? That is, it's a bit of the marketing side, finding your market. How do you find these people? It's important to ask yourself that question from the start, actually.
So those are the channels, then the solution, which isn't necessarily more important, and after that the unique value proposition, which is really the heart of the product, what it brings, it's good to be able to... summarize it very simply in a single sentence, and then there are revenues, costs, and measurement points. So traditionally, from my experience, I know that, by working on my little projects and supporting a number of entrepreneurs, for me the key points are here, here, And here, in fact, these are really the two most important boxes. If we're not sure about this one, it's going to be complicated to go further. Because this one is very, very linked to the... User segment. So for me, it's really this one that, today, with hindsight, these two are really the important ones. to quickly confirm in the field.
So I did the exercise for my project, I didn't fill in all the boxes because it's not necessarily useful. So for my little keychain project, the problem is losing your keys in the street, as I was explaining earlier.
It's not losing your keys in the apartment, because for losing your keys in the apartment, there are already existing solutions. Like Tiles, all those things, connected keychains with a little radar that tells you where the thing is. For me, it's not that, it's keys in the street. So the solution, I already explained it to you, it's the keychain. There, I'm telling myself, damn, with my client, anyway, it affects everyone. Yes, everyone has keys, everyone has lost their keys, so it affects everyone, too easy.
I'm going to be a huge hit, it's too... So I don't think too much. Anyway, to reach customers, AdWords, you read about it everywhere on blogs. You put up AdWords, you get lots of people coming, they buy and all that. $10,000 a month, too easy.
So, let's go. So that was really at the beginning. That took me five minutes.
But still, it's good to... So yeah, the Lean Canvas, as I showed you, is something we iterate on, meaning it's not fixed in time. That is, at first, I filled in my boxes, and then as we conduct experiments, each element within the Lean Canvas is a hypothesis that we will test, we keep this Lean Canvas alive over time. So I'll come back to that, regarding where I am today.
So the first hypothesis to test is the unique value proposition: your lost keys come back to you, roughly speaking, I'm going to test the solution. That is, before looking at the problem, everything I told you earlier, in the end, I went straight to the solution. Which, well, very good, but at the same time, it made me laugh. So I went to test the performance of the solution. Why not?
Because one morning, I woke up and said, 'This is going to be so fun to do.' So I did it.
Yeah. Oh, you're ruining it for me.
So I see the experimentation part. So I'm going to scatter keys in the street with my famous keychain. So I'm going around the... Locksmiths who make duplicate keys, I get sets of keys from here and there, also from colleagues, friends, and all that.
To see if it actually works. So the measurement is easy. How am I going to measure that? Before starting, I think about my measurement points. So it's the number of keys that come back to me compared to the number I scattered in the street. It's quite simple. And there, you also need to set a goal before starting. Even if it's very hard to set a goal because we don't really know, it's still good to set one. That is, there, I told myself, if less than half come back, still, it means that...
It's worth continuing.
So, I see myself in the build phase, where I'm building my first MVP, my first product. So there I go on Amazon and I buy these, keychains where you can slip a little tag. There, I have some here.
It's these things. So that allows me to... I make little tags with a code. On the other side, it says, 'If you find these keys, go to the website and enter the code.' And so, indeed, my MVP isn't super nice. The product isn't necessarily super pretty or super well finished. It looks a bit cheap with the paper and all. But it allows me what we call a low-fidelity MVP. That is, quickly, I only spent 5 euros on Amazon. First product. So here I am, going on Amazon and buying these, keychains where you can slip in a small label. Here, I have some right here. These are the things here. So this allows me to... I make small labels with a code. On the other side, it says, 'If you find these keys, go to the website and enter the code.' "
And so, indeed, my MVP isn't super great. The product isn't necessarily very pretty or well-finished. It looks a bit cheap with the paper and all. But it allows me what's called a low-fidelity MVP. That is, quickly, I only spent 5 euros on Amazon.
Three days later, I received my things, and I was ready to start. I didn't go into a phase of industrialization, contacting keychain suppliers or manufacturers of who knows what. No, here, I wanted to test this. So, you have to be lean. So, I found a way to test it fairly quickly. So, I estimated that with this, it at least allowed me to do this first experiment, which is really about the solution, ultimately. Not about the product itself, but about the solution.
So, in Toulouse, I scattered 17 keys in the street, in more or less busy streets, at different times, in very different places. So every time I placed a key in the street, I noted the time, the street, and the... The foot traffic, to know if it's busy or not. I wasn't very comfortable. I felt like everyone was watching me when I dropped a key. I tried to be discreet, but it was tiring. Because, well, you feel like it, but actually, no. But anyway, each time, it's always the same. So out of 17 keys, how many do you think came back? There's a keychain to win. The one that's all of it.
So, 3, 2, 4, 10, none, 0, 5, 8. I'm still not there. 20, no, because there are only 17.
17, oh, 1, no, still not, 0. 12, who said 12? Over there, here, you won a keychain.
So you'll come see me at the end, you'll get the version 2 of the keychain anyway. So indeed, 12 keys out of 17, that's still 70%. It works, you can applaud.
So, still, well done, hypothesis validated, the solution works. I mean, it's still incredible. I didn't believe it myself.
Yes. So there are some that I didn't get back. Maybe I'll get them one day. Because I have no idea where they are. Afterwards, I went back through the streets, they weren't there anymore. Maybe it was the street sweeper, I don't know.
So what I know is that the fastest key took 45 minutes to come back. That is, between the time I lost it and when I received the email saying someone found it, it was 45 minutes. The longest was three days.
And the geographic dispersion is between about 30 meters and 2 kilometers. That is, between the place where I lost it and the place where I found it again. Well, these aren't metrics that are very useful, but afterwards, it's fun.
So, I really think it's a great idea, it works too well, it's too easy.
So, I haven't tried it in other cities. No, I haven't tried it much in the countryside and all. I should do that, because we talk about it regularly. But well, I have...
No, not that either. But afterwards, there are plenty of other hypotheses to test. Afterwards, we make a backlog of hypotheses, there are a number of things, we prioritize and all. Why not?
So, second hypothesis to experiment with, here I think I have a brilliant idea. What if the people who found my keys became customers of the solution? Those people, when I lost the keys and when I was in contact with them, I really pretended to be the guy who lost his keys, not at all the project leader, to avoid bias as well.
So I had people on the phone where I met them in person, yeah, this thing is great, it's true that it's good, without me bringing up the subject, meaning I never brought up the subject of the keychain, I was really like, great you found my keys, and the people, well, they said, okay. So, second experiment to conduct, because I think, if that works, the marketing is done, I scatter keys in the street, that's it.
Too easy. There you go.
Another brilliant idea, right? So what do I do? In fact, when people fill out the form and all, they leave their email to be able to talk to the guy who lost his keys. So I send an email to the people who found the keys inviting them to buy a keychain for 5 euros.
So in fact, it's an email, relatively nice let's say, a bit promotional, with a button inside to go to... the website, on the site, I have the thing in the photo, I have a buy button, and when you click on the buy button, it just says 'there are no more in stock, come back later.' Because I didn't want to set up a payment system and all that, I'm lean, I stay lean, so I'm not going to get into something a bit complicated again. So I just have a button, a call to action, on which I have a measurement point, that is, I know who clicked on the thing. So in fact, with the tool, today I'm able to know how many emails I sent, who opened the email, who clicked on the link, who clicked on the link inside the email, and who clicked on the buy button. So this is what we call a funnel, we have the whole workflow. It allows us to measure all that.
The measurement is simple. It's the number of clicks on the buy button compared to the number of emails sent.
So here, I tell myself, if I have 20%, that's good. That means it's a good hypothesis.
So, out of 7 emails, how many? I sent 7.
Because 7, because I was so eager to do it that I didn't even wait for other keys to come back. That is, as soon as I had 7, I sent it. So? Two. Three.
More than seven.
More than 7, it could have been, but no.
5, no, still not.
2, 6. Ah, zero.
Who said zero? I don't know. The first one? There, in front? Yes. Okay, you can come later. So indeed, zero. Still, what a bummer. I mean, these are people who had still lived the experience.
They had helped out, they had gone through the thing saying, 'I’m glad I found the keys, glad I returned them to someone, it’s something they could share in the evening, it’s still a bit funny, a bit rewarding.' Still zero.
So there, the hypothesis is invalidated; it’s not a good method. It’s not a good marketing method at all. The hypothesis isn’t good. Meaning it won’t work afterward. I couldn’t, at that time, a priori, it wasn’t a good marketing method.
So there, a bit of questioning everything. And then I tell myself, I still need to focus on the customer segment because that’s what’s important. I have a good product, a priori, a great product that’s efficient, that works. At least it’s efficient, afterward.
And I tell myself, but really, who could be interested in this? So there, back to basics, I return to the fundamentals. And so, the fundamentals are 'get out of the building.' It’s really about leaving your place and going to meet your customers and users. Meaning really...
Go talk with them, meet them, there you go. So there, in fact, I’m going back a little to the theoretical part, I’m actually going back to the problem, meaning is the problem I’m trying to solve really a problem that exists and is it painful enough for the people who encounter it that these people have already tried to solve it themselves? And would they be willing to pay to solve this problem?
So that’s what we call the problem interview, meaning there I’m trying to reach people who, in fact, there I had targeted my customer segment a bit more, it was people who had already lost their keys. What interested me was people who had already lost their keys and to know if once they had lost their keys, whether they would look for solutions to not lose them again and all that.
And that’s the story of the channels, precisely. And there, I was a bit bothered. So, I have answers, but I’ll give them later, as we go along. So there, what I do, the simple way, a tweet. There you go, Twitter, if you’ve lost your keys, can you answer this questionnaire? So the questionnaire, inside, it says, I don’t remember how much it cost you, the guy who came to open your door, if you made a spare and all. There are lots of questions that don’t interest me that much in the end, but it draws people in. And at the end, there was a little field, if you have a quarter of an hour to spare for me by phone or Skype or Hangout, put your email, we’ll get in touch and there you go, I’m interested. What I wanted was to have the people, to talk with the people. It’s really important. So this tweet had almost zero success, it didn’t work at all, just two retweets. There you go, unfortunately. I don’t know, it wasn’t the right time, the right place, or people don’t lose their keys. It might also be an indicator.
Normally, 1:30 PM is not bad. From experience, 1:30 PM, yeah. I don’t know, do you have any advice on that? I’m not Parisian, me. I’m not Parisian.
In Toulouse? Yes, but the metro doesn’t matter.
So, in the end, I had few responses to the questionnaire and all that. I still had three people I could talk to, which is already not bad. From a previous project, I had, or even the theory, from the moment we manage to... talk with 15-20 people, we already have a good idea, it answers a certain number of questions and things, and we have a good vision of the thing. Beyond 20 people, things start to overlap a bit and all. So there I only had 3, it allowed me to start a little. So I did what’s called the problem interview, at the same time I mixed in the solution interview, normally it’s done in two successive phases, but since I already had the solution,
I did things a bit backward.
The idea was that. Have you ever lost your keys? Have you looked for a solution? These interview stories are open-ended questions. It’s not yes, no, stuff like that. We have a certain number of objectives. We set objectives to cover during the discussion, but we let people talk, we’re in active listening mode. It’s not at all... Because we’re here to learn. So I learned a number of things.
So, I’ll put you back a little in context. Me, I’m in my project. I’m all in. It had worked well, the solution and all. And it’s the typical case of the entrepreneur. We’re all in on our project. We have blinders on. We have our head in it. And we don’t take any distance from it. So, what I’m going to tell you now seems like commonplaces. But I didn’t see them at the time, I didn’t see them. And it’s good that it comes from the outside. That way, it’s... Reassuring. And even if we have it in our circle, the circle is completely biased, so it doesn’t work. So it’s good to deal with... strangers. The three people I had were practically strangers.
So, what I learned is that people weren’t necessarily looking for a solution. Once they’ve lost their keys, it’s fate. And I make a spare or I change my door lock. That’s how it is, but I don’t look for a solution. So, what does that mean? Meaning my AdWords won’t work.
Because people won’t go search on Google for how not to lose their keys. They won’t do that. So it’s up to me to go to them. I need to push messages to them, so I need to do marketing, targeted ads, things like that. On blogs or whatever, or mailing or things like that. So there, my AdWords hypothesis is swept away. What I learned afterward, it’s just three people still, it’s still interesting.
I learned other things that are a bit less impactful. There, I learned a great trick. This interests parents who entrust their set of keys to their children for the first time. Their teenager in middle school, they give them a set of keys, the house keys for the first time. Kids don't pay much attention to their things. And so, this reassures parents to have this system. So, I'm not the parent of a teenager, so I would never have thought of that. And well, there, a customer segment is emerging. I hadn't thought about it at all. And this came from a discussion with three people. Out of the three, two of them still told me that. So, So, this shows once again in Lean Startup, there's a very important aspect, which is that... When you have an idea, you share it, you talk about it all the time. That is, we're not into NDAs, all that, filing patents, all those things. No, you talk about it and lots of things will happen to us. That is, lots of things are going to happen. People always want to help, to lend a hand, and especially to give their opinion. People always give their opinion, that's how it is. Not always interesting, but most of the time, they want to help. And there you go. So there, I tell myself, hey, that's interesting. So I note it down somewhere, I put it in my backlog of hypotheses, and I tell myself, I'll have to experiment with this, do a startup loop later.
And there, I learn something that knocks me off my feet. The keychain has to be pretty. I thought that was it. My thing was perfect. I had my keychain. It was good. I had it on Amazon. It was easy. In fact, I hadn't realized that my thing was ugly. I was too focused on the solution, the service, the use. But the keychain itself has to be pretty. So that was more from the women. Wait, on my keys, I'm not going to put something like that. Keys are important, we always have them with us.
So there, a bit of a letdown. There, I tell myself, a pretty thing. While I thought I had the solution. It was efficient. And yeah, pretty. You see, that was really the thing. So there, I'm back to searching the Internet to try to find something customizable. Because for me, it had to be customizable, I need to put a unique code on... Each keychain, it's a bit complicated. So there you go, you can find anything on the Internet. Some things really not great, we can say that. So that, that was more from the women's side. Wait, on my keys, I'm not going to put something like that anyway. Keys are important; we always have them with us.
So, a bit of a fail there. At that point, I thought, something pretty. While I thought I had the solution. It was efficient.
And yeah, pretty.
You see, that was really the thing. So now, here I am scouring the Internet to try to find something customizable. Because for me, it had to be customizable. I need to put a unique code on each keychain. So it's a bit complicated. So, you find everything on the Internet. Some things that are really not great, we can say that.
This one is a festival of sailor fields. You could put a little of what you wanted, but well, it's not much prettier than what I had before.
Yes, it floats. Yes, that's true, if you lose them while fishing. I hadn't thought of that.
So, what bothers me the most, actually, is that I have to spend money because to make a series of things and all that, you need a certain number. Otherwise, the keychain costs a lot, a lot. So, that's the big disappointment. So, it's the emotional rollercoaster of the entrepreneur. I was on top, meh, meh, meh, and then everything collapses. Really, I thought, well, too bad, I'll drop the thing. I'll drop the thing for a bit. For two months, nothing happened with the thing. And then, finally, what happens? The revelation is Britney Spears. For those who wouldn't have recognized her. Oh my God! I discover the Fab Lab and laser cutting. And that changed my life. So, the Fab Lab, for those who don't know what it is, is first a place managed by an association where you can find all kinds of industrial equipment that you don't traditionally have access to, or there, you can go use it. So, Fab Labs are well known; they're talked about on TV for 3D printers. But there are really a lot of other things. So, this is the Fab Lab in Toulouse, one of the largest in France.
There are 3D printers, spaces for electronics with oscilloscopes, soldering irons, all that stuff. Things we wouldn't necessarily buy personally. So, there's the laser cutting machine. The laser cutting machine allows you to cut certain types of metals, plexiglass, cardboard, paper.
Wood, well, certain types of wood. And so, it allows you to cut and also engrave. I'll show you later.
There are also fridges because there are people who do biology, who grow things, I don't really know.
And so, It's an associative place; you join for the year, and then you have access to that equipment. And it's very interesting; you meet all kinds of people. I was saying earlier, it's known for 3D printers, but that's really not what's used the most because the uses for 3D printers are complicated. However, laser cutting is heavily used. So, here I go, I'm so happy. At that point, I thought, there we go.
With this laser cutting, I'm going to be able to make amazing things. I'll make my keychains by hand myself.
I take a free two-hour training every Tuesday evening at this Fab Lab. There's a two-hour laser cutting training. They explain the tools, the fire extinguisher is there, the blower, you have to turn it on, otherwise you get cancer. A whole bunch of things that are still quite important. Because afterward, you should know that when you arrive at the Fab Lab, there's a person on duty to open the door and help a little, but they're not really there to do the work for you. That is, it's really the associative and self-organized side, meaning you arrive, there are people, you agree with the people who are there about who uses the machine when. That is, there's no reservation of the tools and all. It's really left to self-organization, which also encourages exchanges and everything. I've seen artistic projects or others where people do really nice things. So, I encourage you to go check out the Fab Lab near you; I think there are some in Paris, there are several. It's always interesting. In terms of curiosity, there are always people doing quite funny things.
So, the laser cutting training, then I buy plexiglass, so actually at the Fab Lab there was a small display with things that had been made with laser cutting, in particular
a small box of colored plexiglass samples, there were several colors, it was sold by Inventables, it's a website based in the United States.
So, I started looking, and actually, it would have cost me more in shipping, over 70 euros to have them sent from the United States, than for the raw material itself. So, I was a bit bothered. But luckily, actually, well, luckily, yes, let's say that, I was going to New York for my honeymoon at that time. And so, I had the plexiglass delivered to my hotel and brought it back in my suitcase. I brought back Converse and plexiglass from New York. My wife was happy; it was great.
And yes, you have to be inventive. Lean startup, you know. Lean, you know, especially.
Yeah, but well, I'm being filmed, I mean that wasn't the main reason I went there.
So, engraving and cutting, you don't really see what the dust is. So, actually, there's a plexiglass sheet, and here, it's engraving. There you go, you'll see the result. Actually, roughly, this is a plexiglass sheet that I brought back, which was cut.
So, I only cut half of it. I'm keeping the rest for... I'm not stocking up; I'm lean.
And in case there are requested changes, I have the material left for later.
Afterward, product photo session, and all that. So, this is with a friend. I made a lightbox, also do-it-yourself, with an Ikea lamp, everything. The camera. And so, the friend who knows about photography, because I know nothing about it.
And the result, drumroll please. Ta-da! You can applaud.
Thank you. And I can tell you that I'm quite proud of my thing because, after all, I'm a computer scientist, I'm in the virtual world, meaning I make websites, information systems, and whatnot, and here, I built something real. I mean, it's not just anything. I have some, of course.
I can show you.
You can take a look. There you go, I'm happy with my thing.
Pass them around, I have different colors.
What I'm showing you are keys that I threw into the street. These ones have small defects. Afterwards, the real ones don't have any.
So, here's the thing, what also happened is that at one point, my goals had changed. When I discovered laser cutting, ultimately the product was secondary. What I wanted to do was laser cutting. I didn't really care about the product.
Because it's still, as a geek, a self-proclaimed geek, it was still super important. So, I had done the laser cutting. I mean, no matter what happens afterward, it doesn't matter. The project was a success. So I could stop there, but well, I still have other things to tell. Yeah, I need to speed up then. So, the question of price comes up. It's always delicate, meaning how much I'm going to sell this product for. So, I mean, how much does it cost? So I use the Van V... method. There you go.
Weston Dorp. So the method is four questions. At what price do you consider that this product cannot be of good quality? Meaning, from how much? The price is so low that it's not possible for it to be of good quality. An iPhone for 200 euros, that's not possible. There's a problem somewhere.
At what price do you consider that this product is really a good deal? So that means it's not expensive enough, roughly speaking. At what price do you consider that this product starts to be expensive? So that's the opposite of the previous one. And the last question, at what price do you consider that this product is so expensive? You don't buy it. So, I did this with about ten people, and it gives a median price. So the price, how much? How much would you pay? So, before the first one speaks, because it's the first one who influences everyone,
How much does it cost to have a set of keys remade? I'm not talking about mailbox keys, an apartment key, a car key. If you have a locksmith come to your place, there you go. I still have a 70% key return rate.
No, but it's important. So, how much would you be willing to pay to buy one of these keychains? How much? 15? 1 euro, sorry. I understood 15. 15 too, yes, okay.
5 euros, yes. 5 euros. Okay, yeah. So I asked this question to about ten people and I arrived...
No, but I got everything, from 50 cents to 50 euros.
50 euros.
I didn't sell it at that price, of course, I'm honest. But I didn't sell it. And so, in fact, the median price was 10 euros. Well, 10 euros, yes. And it's not lies. It's really the thing. So, even on a Saturday afternoon when I was doing my laser engraving at the Fab Lab and all that, there were people who came to visit the Fab Lab because they had probably seen a report on TV about the 3D printer. People aged 45-50 who didn't really know much about all that, who were curious to see the 3D printer, who asked me what I was doing and all. I explain the concept to them. And then... So of course, I ask them the price and all. Because every time someone asks me the price, I turn the question around.
And well, there was one person who bought one from me. It was my first sale at the Fab Lab.
In hindsight, I think he wanted more to bring back a souvenir from the Fab Lab than...
Than the thing itself, but well. At the time, that came to me a little later. That's also what I owe.
So then, the question of packaging and shipping comes up, meaning I can't go sell hand to hand, it's still online sales, as we see everywhere. And especially, I wanted to send one for free to the three people I had interviewed, to thank them for giving me a little of their time. So, packaging and shipping, I buy envelopes. Padded ones, stuff and all. I put my thing in, I say to myself, there you go, too easy, I'm handling it, and in fact, no, big fail.
The thing arrived broken, well, out of the three I sent, two were broken. In fact, I had attached the ring to the keychain, and then in the postal machines, it... Broke them. Even though the thing is pretty sturdy. I don't know if you've seen, but you have to go for it. I've had mine in my pocket for 4 months, it hasn't moved. It's really... So, I learned again. Same, I could have launched my thing full steam ahead.
I was with an article on TechCrunch and if I had sold 300, I would have sent the 300 and then I would have had 200 broken ones. So in terms of images and all, no, I had done it small. Anyway, I didn't have a choice.
So, I still have customers to find. So I'm launching a small A-B test, meaning testing different messages. To see which is the most effective. I'm speeding up a bit because I don't have much time left. So, I'm going back to a hypothesis I had set aside. It was the parents of teenagers, I mentioned it earlier, who might be interested compared to anyone else, meaning a more generic message. To do this at the experimental level, what I do is run targeted ads on Facebook. On Facebook, you can run ads and target, meaning here I was addressing people over 45 who have at least one teenager over 10 years old. You can do that on Facebook. So I have two versions of the ad; what changes is just the text here. Here it says a pretty and clever keychain so you no longer lose your keys. So this is the generic version of the thing. And here, no longer be afraid to entrust your keys to your children. So, in addition to the fear and all that, I think it will work well. After that, the rest is the same. And so, there is a 'Learn more' button that goes to the website. So I do this, I run it over two different weeks, one week this one and one week that one. So what I'm going to measure is which one leads to a purchase. I'm talking about a funnel, meaning there are a certain number of steps to complete. And I set the same budget, so 15 euros for each ad, which means I have roughly the same number of clicks, the cost per click is similar. So that makes it comparable. So in your opinion, between the two, which one works best for parents?
For teens, yeah. So I'll show you the result. The good news, or rather the bad news, is that I sold zero, no matter which one. So in fact, what is represented here are the steps. So here, in the ad, they clicked on 'Learn more.' So I have 70 and 68. Since I had the same budget, the cost per click is roughly similar, so I have roughly the same number of people entering. After that, they arrive on the site and add one or more keychains to the cart. So here, I'm more successful with parents of teens. Here, I had 12, and there, I had 4. So already, this one works better. That means I need to go more in that direction. And then, one person entered their details, and after that, I lost them. In the end, there's no one left. So what this also means is that... Well, here, I'm not too bad in terms of conversion, but it means that here, I have a problem between these two steps. So I need to try to optimize these steps. That is, here, it might be a UX issue or something—I don’t know—but it’s not working, and I need to work on it. I had set all these measurement points; again, it's important to measure. It's useless for me to optimize credit card entry and all that because, in any case, I have zero. So you always have to try to optimize the upstream phases to ensure the funnel is as wide as possible at every step.
I also did iterations of lab testing on the price, where I reduced it to 5 euros. In the end, I feel that the price doesn’t really impact the performance of the thing. So today, where I stand, my overall conclusion—well, this is well summarized—I feel like I’m onto something. I’m not far off, but it’s kind of a big fail, actually.
In fact, it doesn’t work. I didn’t sell any online. When I say none online, none from people I don’t know or whom I haven’t talked to or sold the thing to. People who arrived cold on the site and went through the process of buying the product. So I don’t have something scalable. I sold some to my circle or people I knew or people I met to whom I explained the concept. Because they find it fun to lose keys in the street, to have the experiment and all that.
So that’s it, I didn’t manage to go beyond that. At the same time, regarding my goals, I still achieved a number of objectives, which are: I had fun, it was still fun to do this thing. I continue to have fun because it’s nice to go talk about it at conferences and all, to exchange on the subject. No, but it’s a feedback experience.
Make a little money, so at least I broke even. So I didn’t do the calculations too much, but there you go, my 70 euros of... Plexiglas have been amortized because I must have sold a good twenty.
So I’m within my costs; I didn’t make any money. So my investment of less than 50 euros, that I achieved.
And I was able to test the startup approach. So what should be remembered, what’s interesting, is that for less than 150 euros, I still managed to test a market, a business. Of course, I could go into B2B with Mr. Minute, to get accounts. No, I don’t want to. Those are constraints; I don’t want to do that. And I have even less desire to do it because... I don’t feel that people are really interested in the subject.
If I had done that from the start, I might have been able to do it, and then ultimately, no one would have bought any at the Mr. Mint counter. Here, with ultimately few resources, being a little advanced, staying very lean, I managed to test a market that doesn’t exist.
No, but that’s it, I realized that this market doesn’t exist with few resources. So what I explain about Lean Startup, we talk a lot about it in the startup world, but it also works in the context of larger companies or for existing products to test new features. We no longer call it MVP, but MVF, which is a Minimum Viable Feature. That is, this approach can be done in Lean mode, by saying, hey, I’m going to add a button I don’t know what in my UI, and instead of it really doing something, I’ll first see if people will click on it. So this entire approach is repeatable in other contexts. So that’s it, I’m done. I just want to tell you that it’s almost Christmas. Ah.
Well, it’s almost Christmas. So it’s a good gift idea if you know people who lose their keys. Don’t hesitate. There you go. I have some with me too.