In this episode, Amaury and Mitchie describe the first two levels of prototyping and explain how startup founders can build an MVP relying on neither tech nor funds but actual resourcefulness to move ahead in their venture.
Pre-order a copy of Amaury’s new book, Startup Without a CTO, and get an early digital version two months before the official launch: http://startupwithoutcto.com/
Mitchie Ruiz: Welcome to Episode 11 of My CTO Friend the Podcast, where founders come to learn how to manage a tech startup. This is Mitchie Ruiz, and I'm with Amaury Khelifi. Hi, Amaury.
Amaury Khelifi: Hi Mitchie, and welcome, startupers, to today's podcast. In this episode, we are going to go through dozens of prototyping techniques to get started with your MVP, the minimum viable product. The best part is no technical skills are required. You will learn about some of the less-known options out there that you can use to create the first version of a digital startup product.
Mitchie: Now, remember that for this first series of the podcast, we are working our way down Amaury's book [Startup Without a CTO], going through the tips and real startup experiences he's gathered throughout the years. So if you're a founder without technical background, listen up, because you're in the right place.
Now, let's just go ahead and get started. Last time, we covered the roadmap definition and presentation. We've defined a roadmap, we've presented it. What's next?
Amaury: Well, usually when you create a plan, when you plan something, everything doesn't always go as you planned. That's really the case with startups. After presenting your roadmap to investors, maybe you've got some reactions, maybe the investor said, “Your idea is good, but I’m not going to invest right now. Just come back in a few months and let’s see how things go,” and you just need to move forward with or without the investment.
That's usually where the founders start to have some self-doubt. They wonder if they did things right and what they should do. They are just thinking with their own references, what they already know. But there are so many things that they don't know and they need to get to know, and that's where the prototyping the MVP phase comes into play.
Mitchie: You mentioned that the MVP stands for minimum viable product. Now, in this case, the minimum viable product would be like an app or a website. Are CEOs supposed to learn to code or to make this themselves?
Amaury: No, definitely not. It's going to take years to learn to code, and to do things correctly. Instead, I really recommend to learn tech management. And that's why we are here today, is to teach people how things work and then help them figure out what they really need to do.
Mitchie: Now let's clear something up. Is having a minimal viable product the same as prototyping?
Amaury: Well, the prototyping is to evaluate first the technical side of things, the technical assessment. Is that possible to do this or that? That's the basics of what a prototype is all about. But with startups, there is not that much technical challenge because these days, we know what we can do and what we cannot. It's more a matter of time and being sure that something is going to work on the market.
So the assessment is more on the business side of things than on the technical side of things. And that's where the MVP comes into play, which is more a kind of business prototype that is going to prove that we have a market. We build the minimum valuable product, the smallest product possible to prove that we have people that are willing to pay for what you are creating.
Mitchie: Prototyping, in a sense, is just making sure that what you're going to do is helpful for people, and that it's actually technically possible. Got it.
Now, I'm a CEO, like you mentioned. I'm filled with self-doubt right now because I presented my roadmap to multiple investors, and they were all like, “Well, thanks. Thank you, next.” What do I do now? I'm stuck in this whirlwind of self-doubt. I moved forward from that initial, let's say, impulse of doing my roadmap, and I did all of the Post-its, and I did all of the visual board, and I have all of this energy that I don't have anywhere to put it in. What do we do?
Amaury: So the next thing is just to actually build something. Either you already have maybe a PowerPoint presentation, maybe you took the time to build a simple website. And if you haven’t, that's basically the first thing you should do, is to build your first website, and then just find ways to demonstrate and start selling your solution.
You might think, “But I don't have any product.” That's why we are here. We just want to build a prototype. We just want to have something to sell, and that's what we are going to tackle. And that's where I would like to introduce you the four levels of the MVP, the four levels that I have distinguished after working with several startups and seeing what has been done at each step.
The Level 0 is basically no tech. That's where we start. It's a simple website, nothing technical. It's like as easy as creating a Word document. Then you will have the Level 1, which is just tiny little tech. It's 1% of things. It's connecting dots between existing tools. And then we go on Level 2 with maybe 5% of your entire MVP that is going to become custom development. And then we go to Level 3, where 10% to 15% will become custom development, and the rest will be open-source solutions or existing solutions.
Mitchie: We are going to cover Level 2 and Level 3 on a future episode, because there's a lot of value in this. We have so much to talk about. But in a sense, this is just going to be the MVP/prototype. This is going to be the most, let's say, stripped down version of your product at the beginning, your first version, and it's still going to be much cheaper than just going forward with custom development from the get go.
Amaury: Absolutely. I always recommend to test your market and to be sure that people are, again, willing to pay for what you have to offer.
Mitchie: Wonderful. Let's just go ahead and jump right in into Level 0. Like I said—CEO, have a roadmap. I read somewhere that 1% of startups get funded, so it's clear that when I presented that to an investor, he was like, “Yeah. Thank you, next. I'm not going to put in a penny there.” So, what do I do? You said a page, yes. We start with a page.
Amaury: Yes, so you start with a simple page. There are lots of tools. We can name Wix, Strikingly, there is Squarespace as well, and there are other landing page tools that enable you to just create a simple page. It's as easy as it would be for a Google document or a Word document. You get your own domain name, so you have a real shop. It's like the fence of your shop, as we would say.
Then when you have that, the goal of this landing page is to just get people to subscribe to something. To subscribe to maybe get to know the service, to have you on the call or to get a PDF that explains something in your market, something useful for your clients. You need to attract them. It's always about trading. You provide something, you provide some value, and you get their email in return.
Mitchie: Got it. Right now, we are using simple tools: Wix, Strikingly and Google spreadsheets to organize what we have to do.
Amaury: Yes. When you have this information, you will be able to catch the email, the requests that you can get from your clients, and put them in a Google spreadsheet, for example, or an Excel spreadsheet. The goal for you is to organize your work to see what your clients’ needs are, and then you will have a better, deeper understanding on how to help them with their own challenges.
Mitchie: And this is getting your hands dirty. You're going to go ahead and do everything yourself, provide the value yourself. Everything service.
Amaury: Yes, on the Level 0, there is no tech. You do things with your hands and your keyboard, and you send an email, you get them on chat, on phone, and you just sell. It's basic service. And even if you are not profitable with this service, even if you sell something like $30 per month and that takes you an entire day, it's okay. It's just like you are going to replace yourself with a software, but it's just to prove that people are happy to get this service, and you will figure out exactly what you can industrialize in that process.
Mitchie: This is nice. You are embodying your startup. In this stage, you are your startup, you are your service, you are the help. It's nice, if you think about it.
Amaury: You know SaaS, software as a service, means services. You need to build a service first, and then you replace it with a software. That's what software as a service means. It's not like I create a software, and it's going to provide services. It doesn’t work that way. You need to know your client, you need to know how to serve them, and then replace yourself with your software.
Mitchie: Right. It doesn't work the other way around. Got it. We have it. We have our organization. We have the manual value. We're selling the services ourselves. What do we do next? Are we moving towards Level 2 already?
Amaury: Yes, Level 1. We are going to move to Level 1.
Mitchie: Sorry. Yes, my bad. Level 0, Level 1. Little tech, 1% of tech.
Amaury: So far, let's assume that you have 10 to 20 clients. It's a bit overwhelming. They are taking too much time. You’re not earning that much money. Of course, you are really testing your market. Now, it's time to industrialize things.
There are two ways to industrialize things in terms of aim, in terms of goals. First off, either you want to improve your clients' experience and provide them a better website, or maybe the beginning of what would be a kind of mobile application, or maybe it's a more responsive website. Or you can provide a very good customer service like you did before, but you just industrialize the work on your side.
Those are two approaches. For both, you will be able to use the same type of tools. For me, the best one is Zapier. Zapier allows you to connect the dots between a website like WordPress—like a simple Web page or a landing page, your automation-making tool—any type of tool, and your Google spreadsheet if you want to add a new line every time a new customer fills in the survey, you can do so as well. And that will enable you to create some automations, maybe first in Google Drive if you want to learn macro, that kind of works too. But you can move forward and automate things one little piece at a time.
Mitchie: Just to pedal back a little bit and to have a little bit of reference to our previous episode, on Episode 7 of this podcast, we did mention the founder’s basic toolbox. We did mention Zapier, and in this case, it's just the way that we're starting to automate the actions that we're already doing every day, providing the value that we are manually from Level 0.
Right now we haven't differentiated. We have two focuses, like you mentioned. One is improving the experience for the customer, and the other one is improving your experience yourself as a service provider. Everything that we're going to talk about right now is focused on one or the other, but we're using the same tools. Am I right?
Amaury: Yeah, we use the same tools to organize your work. And basically, Zapier is one that is quite central even if there are other tools like Automate.io, Workato, Flow from Microsoft if I'm right, IFTTT as well. They are basically doing quite the same. But let's focus on your customers, because I like focusing on customers.
The customers will need a better experience. They will need a website. They will need something that looks good and that makes them trust the company. You can balance, either they are going to trust in you as a person or as a brand. And if you move to a website, you will have to have just a proper website that’s well-designed. So the first approach would be to build this better website, and often the idea that comes to mind is to use WordPress.
WordPress is a CMS. It can be WordPress or another CMS actually, it’s quite the same, but WordPress is quite well known for this. It's a CMS, a content management system, which means that you can build a website with it. You can edit the content, add content as well as create a blog. You can host a podcast. You can put some videos. Any type of page can be created on WordPress.
You also can level this up by searching for plugins in WordPress. There are thousands and hundreds of thousands of plugins for WordPress, and there are plugins to create Airbnb-like application. There are some to create like an Uber application or, let's say, booking. Almost any type of application has their own plugin, if you want to create the same on WordPress.
First the pros, it's just very cheap. You can buy this plugin for maybe $50, and have an Airbnb-like website. The real cons, and the real drawbacks of these, is that it's going to use WordPress database. It's going to use a database that has been designed to create websites and not that much, which means that it's going to work for maybe 1,000 or 10,000 users, but you will reach the limits of the system sooner or later.
It's okay. It's totally fine. The goal for you is to prove that you have a market, to prove that you have your first customers. If the investor that you met when you were showing your roadmap said, “Come back in three months and show me that you have some paying clients,” fine. I build a WordPress, it costs almost nothing. I sell something, I get a little bit of money, and I have a proof. Now maybe he would be more interested to see what I achieved, and then inject a bit more money in the project.
Mitchie: Just to be clear, we're on Level 1 of the MVP construction. In most of the MVP construction, we are focused on doing something that doesn't have to necessarily be scalable, but that you can actually prove that it works. Prove that, again, we go back to the definition of prototyping, that there is a market for it. That people are willing to pay for it, that it’s going to help people in some way. And two, that it will actually work, that it technically can work. This is what we're doing with this approach.
Now, WordPress is building a website, and building something behind the website. We're mucking up an app, in a way. We're pretending that it's an actual working thing, or is it an actual working thing?
Amaury: It is. It is, and it works. You can serve clients. Like I said, 1,000 or 10,000 clients, it works. It will slow down. Definitely, it will slow down, because it has not been built for that. It will be complex for you to connect with other information systems. Your developer, if you want to develop something very specific, will have to create maybe another database or another structure, other tables, and at some point, it will be a bit complex to handle down the road. But that's fine. Just consider this, if you go with WordPress, just know that you will drop it sooner or later. At some point, it will not scale, and that's fine.
That's exactly the same if you want to build an e-commerce website. There are dozens of tools like Shopify, Magento, WooCommerce, which will allow you to create a website. WooCommerce is built over WordPress, by the way. And that's fine. Build an e-commerce website, sell services, goods, whatever you have to sell on these type of techniques. It doesn't cost that much. You can just buy your templates to have a well-designed website that looks good with your own clothes, and you just replace the logo and that's done. You have an MVP that works and you can start getting revenue and receiving payments with it.
Mitchie: Got it. Now, are we done with the customer focus? Is there something else that we should focus there?
Amaury: No, not so far. If we are really focused, usually the prototyping approach, if you talk to people that do design sprints, they will focus on not even building an actual MVP, but just building screens, and that's what some designers call prototyping. They just use a tool like Figma, like Adobe XD, like Marvel. They create some fake screens. It's just a Photoshop screen, and they make it clickable.
But you can also do that with a PowerPoint. You just create a frame or a rectangle and you say, “When I click here, you go on to Slide No. 5.” That's quite the same. It can enable you to show a fake application or record a screencast video of your application. Fine, but I would prefer a real approach with maybe, yes, a WordPress behind or something that can make things real and really serve people through the first two.
Mitchie: So you're saying an actual CMS?
Amaury: Yes, that's definitely an option. And the only way or the only situation where you would not consider a CMS is, first, if you have the money and if you are really sure that your market is going to use your tool. If you were working in these markets, doing services for 10 years, and you know exactly what people need, in that case, maybe your approach will be more on building stacks, building software that will industrialize the process of your clients. And then the second approach, the more backend approach, would be more interesting to have.
Mitchie: That's where we move to the second approach. Let's go back a little bit. If you need to prove your market first, you focus on the customer for a Level 1 MVP building. And for Level 1 MVP building, if you are completely sure, you've done the market research, you've worked in the industry for 10 years, you know what's going on, you should focus your efforts of improving your product in the backend, so what has to do with you as a founder actually providing the services.
Amaury: Yeah, the second one is more scalable. If you focus more on building the backend, it's like building the right foundation, and that's something that can be more scalable down the road.
Maybe first we should explain for a bit what the main components of a software are. A software is usually built with a front end, and that's what you can see on the website. That's what you can see on a mobile application. This front end is going to communicate with what we call a backend.
A backend is more an organization of software that is on the server side or on the cloud, where we store the customer information. And inside the backend, we can even identify two components: we have the database or the storage, where we store information files or user information; and there is what we call the middleware, which is basically the pipe between the database and the front-end application. This middleware is not a new pipe, it's going to organize data, generating some automations, sending email when it's required, sending SMS, analyzing an image if required, computing things.
So we need to have these three levels of a software in mind to understand the approach. Even if we are talking about the prototyping, as a CEO, you need to understand where your efforts will be focused. So far, when we focus on the customer, we are focusing on the front end. And now that we are focusing more on building a foundation, a solid foundation, we are focusing on the backend. So, to get started with the backend, on Level 0, your backend is a Google spreadsheet, if you’re following what we were talking about.
Mitchie: The backend would be Google spreadsheet as your database, and then the middleware would be yourself?
Amaury: Yes, absolutely. Good analogy.
Mitchie: You would be the one putting in the information in the database and communicating it to the customer, so you would be the one communicating with the front end.
Amaury: Absolutely. If your customer doesn't have a real interface, you are doing the middleware, taking data from the Wix web pages or landing page, and putting it in the database in the proper structure. The goal, in that case, is to automate the middleware.
If you want to automate the middleware, you have several approaches. The first one would be to use and to improve the way you use your Google spreadsheets, and connecting Zapier, for example, and creating some automations. That's the first level.
If you want to do things a bit more professional, you have another tool, which is called Airtable. Airtable is kind of a database, like an Excel spreadsheet on the Web. But you can organize things with different forms, with different views. It's really better than Google spreadsheets, and it can be used as a database for your application.
That's where we are moving slightly to what we call the headless CMS. A headless CMS is a kind of content management system, but without a website. It's a database with a middleware and a backend interface to administrate your information for you as a company.
And that's great, because with these type of technologies—I already tested two, Strapi and Ghost, they are both amazing. You can just design your application, you design your database, how you want to store your information, your customer information, your car's information, your room's information, whatever the business you are in, and you make relationships between tables, and you are designing basically a real database.
If you connect this middleware, the headless CMS like Strapi, to a dedicated database like MySQL, which is a specific database format, PostgreSQL, MongoDB—it also works with NoSQL format—you will figure out that on the database, you will have the real data and they are usable through the API—that's what the headless CMS provides, so that means that you can build your future application based on this API, or you can, by tomorrow, build your own backend software as well as do some requests on this API.
So that's really great because it puts all the well-known standards, which are REST API, a common database format, and you also have an interface to administrate your data. That's really an option to consider, and it’s not that complex to then connect this type of information system to future development.
Also, you can connect them to Zapier. Zapier is able to do some REST API requests. It's a bit tricky, it might require a little bit of help of a developer, but within half an hour, an hour of his time, he will be able to connect the headless CMS to any type of tool.
Mitchie: That's where the 1% of custom development would get in?
Amaury: Yeah.
Mitchie: Now, let's review that for a little bit, because that was a lot. When we are focusing on the backend, we're still at Level 1. We have the first choice, which is the one you should do or focus on when you're just coming out of Level 0, and you decide that your approach does have a market. And that would be using Airtable as a database, and using tools like Zapier and connecting Airtable with other tools. If you move a little bit forward and you decide to go all in, you would use a headless CMS.
Amaury: Yeah. You can start with a headless CMS if you are already at that stage, and if you're really confident. I’ve worked with some startups who based their backend on headless CMS without passing through Airtable. Airtable is more the do-it-yourself approach. For headless CMS, you will need a bit of help from someone who is a bit techy. Not that much time. It took me less than an hour to build a backend on Strapi, for example, but still.
Mitchie: Got it. Now, on this aspect, we are focusing on the backend still, getting someone with a tech background to help you out with whatever you have to do here. You said it was an hour for you, would that be the case with a regular developer? How would people know who to get and what headless CMS to use?
Amaury: Well, the point is that headless CMS technique is not that known so far. I mean, it's quite recent. You can either find a specific developer that knows Ghost, or that knows Strapi, and ask him to build your backend for you. But the difficulties would be for him to really understand your need. Or you can try to find someone who is a backend developer, and let him know that you want to build your first backend based on this.
So basically, if you want to go with a Strapi, which I think is built on Node.js, you can find a backend developer with a Node.js background, and ask him to be build the backend on this application. Even a front-end developer would do the job. It's so easy. It's just a small tutorial to follow, that if you have a front-end developer who is used to developing in JavaScript, he will be able to understand how this tiny backend works and directly design the backend with his mouse, and the easy interface that Strapi provides, and then build the application. Actually, that's a use case I have with one of my clients.
Mitchie: Got it. Now, in this scenario of headless CMS, you do have the front end separate? So we would be building off the landing page, or we would be…
Amaury: You will have to develop something, or you will have to build something for the front end for your customer. What the headless CMS is going to provide you is an administrator in the front end, which is an interface to administrate the content of your database. But that's usually not an interface that has been built to be provided to your end client. So you will have to go through developing your mobile application using existing front-end development and connect them to your rest API.
Mitchie: It just translates whatever's going to be coded into the application, it translates for you, because, hello, our listeners are non-technical startup founders. So, you want something to translate it to you?
Amaury: Yeah.
Mitchie: Got it. Alright, that was a lot. Are we missing anything?
Amaury: No, I hope it was not that techy and that you, as a CEO, were able to understand the value. It's not really understanding each technology or each name. It's more the approach. It's knowing that you can build something without having to code, or just by finding the right strategy. And definitely, going with the CMS is easy, a landing page easier, and even easier using tools. Don't be shy by using a Google spreadsheet. That's fine. You are a startup; you need to do things out of the box. And then later, you will be able to make it in a more professional way. That's totally fine.
Mitchie: Yeah. The point of this is to recognize that outsourcing your development, or hiring a developer, or hiring a Web agency is not your only option. If you don't have the funds for it, you can do things yourself and do things with minimum tech, with minimum money, with minimum external input, so you can actually have something to build off of. That's the main thing. And if you do have something to build off of, you can actually show that to investors, to incubators, and that will be something that will inspire them to participate.
Amaury: Yeah. And the point is that there is not that much headless CMS agencies or developers because they don't have that much work to do. If it's something that they can build in a few hours, that's not really profitable for them to sell these services. That's why they are more focused on real, hands-on code or projects that are going to take days and months because it's more profitable for them.
For you, as a startup founder, finding these workarounds and these tips, and knowing that we can do things in a much faster way is just key. And then, if you get a developer, he will be able to dedicate the time on what is really important for your business, which is maybe building a real custom interface.
Mitchie: Got it. From a headless CMS, you can build off of, since the front end is separate. What the customer's going to see is separate, you can just go ahead and do whatever you want on that side, and still be connected. That's a very good option. It's more scalable than the other ones. But remember, you do have to go to it when you are completely sure that what you're doing is going to work, when you have that market understanding.
Let's just go ahead and conclude. We've gone through different tools, we've gone through different techniques. The point here would be to recognize which tools are going to help you at the beginning: Wix, Strikingly, for your Level 0. And then on Level 1, Zapier, Airtable. Level 0 would also be Google spreadsheets and so on and so forth. CMS Level 1, headless CMS Level 1, databases—we cover a bunch of them and you can look at them on our notes as well.
The idea here, Level 0 and Level 1, is to create the necessary connections as you go along, to make sure you're providing the actual value and service you want to provide. It has to be in line with your startup, even if it's the first thing that you're going to provide. They have to be that core value that you defined on the previous episode—Episode 9 would be defining your roadmap. That core value has to be there. You are just figuring out what LEGO blocks you're using to get there.
Amaury: Absolutely.
Mitchie: Now, next we're going to cover the next two levels, right?
Amaury: Yeah. Level 2 and Level 3 is where you have your automation, you have your tools, you connect the dots, but sometimes, Zapier is not that reliable. Or maybe you want to do something that does not exist, so you will have to do a bit of development. Basically, we are going to inject a bit more custom development for two purposes: again, to automate things at a higher level for you, and to provide a better customer experience.
Then, on Level 3 is where you are going to have all that ecosystem of tools, and you will say, “This one is not useful anymore. This one, I can replace it with my core software.” Then you are going to move from external tools to internal ones, keeping the control of your real development and building the core of your real application, and that's the transition to a real product.
Mitchie: Wonderful. We're actually moving towards what we want, even if we don't have that much investment.
Amaury: Yes, absolutely.
Mitchie: We are not touching the savings fun yet.
Amaury: No. That's the goal. The goal is to prove that you can do something with a low budget, because that also means that you will be always able to do the same and not waste money just because you got some investments.
Mitchie: Got it. Next time, that's it, Level 2 and Level 3 of building your MVP. This time, we're talking about scaling your MVP with almost no money. Now, we've covered a lot. I hope it was clear. If you have any questions, please go ahead and leave us a comment on myctofriend.co/podcast, and if you want to look at the links and notes for today's episode, you can go to myctofriend.co/11, which is the number of today's episode. Anything else they should check out, Amaury?
Amaury: Yeah, they will find on this URL, myctofriend.co/11, all the links to all the tools that we mentioned today, which will make it easier for them. And remember, we are writing my book right now. If you want to pre-order a print copy of that book, go on startupwithoutcto.com and you will get an early digital version, two months before the officially lunch.
Mitchie: Wonderful.
Amaury: That's it for today. Thanks Mitchie, it was a great time again.
Mitchie: Thank you, Amaury, I had a good time. I learned a lot, and I hope you guys did as well. Again, I hope I hear you soon next time on Episode 12.
Amaury: Yeah. Until then, go on iTunes and put a little review. We would really appreciate it.
Mitchie: That will help us with the ratings, that's true. Thank you very much.
Amaury: Thank you for your time, everyone. Bye, Mitchie.
Mitchie: Bye.
