How to Design Better Products Than Anyone Else (3 Step Process)

If you're working on a product right now and you want to have a strong, consistent design direction set, you want to already know some of the features that you're going to build for the product, and you want everybody working on it to be aligned, this is the exact process for you. My name is Charles Postiaux. I'm the founder and designer at Baobab Studio. I've worked with companies worth over $160 billion. And in this video, I'm going to walk you through my exact 3-step, very simple 3-step process that I've used for a company worth over $25 billion that I've just finished working on, on the design sprint with them. And it's super simple. So the first step is kind of what you can see here is going to be to define the theme of the design, of the scope of the sprint, if it's a sprint or of the product itself. And so for example, here the theme was kind of like this AI agent theme that is pretty common nowadays with startups and companies. And so that was kind of like the goal of this sprint and this project. And so then I also created some other keywords that were related to that project. And that would help me kind of figure out, okay, what kind of designs, what kind of direction do we want to explore? Should I find out? And so, for example, we wanted something that was ambient, that was always here with the user on the screen. We wanted something that was quiet, so not super big, not visible all the time or bothering the user. It was something that was easy to be aware of, that was always ready on the go, no matter what happened, it was there. And so you can see that we wanted something that was both customizable, powerful, and intuitive, but at the same time, very low screen footprint, very quiet, very ambient matching, kind of like your background, always there whenever you want it. And so from there, I basically just explored as many products, as many design and visuals as I could find. So for example, here is like some of the work that Open Purpose did on one of the projects. Here is like V7. You also have some of the recent or new iOS kind of glass texture. You have some like dot apps here. You have some granola design. You have sky. The goal here for me is just to brain dump as many ideas, as many references as I can. Based on the theme, so obviously the core theme is the company, right? The client, which I cannot say who it is because of legal reasons. That was already a big theme, and then it was also again that AI agent always there for you kind of thing. And so these are like the big major theme, the kind of umbrella that was covering everything else coming from there. And so then the next step is okay, what kind of design, what kind of direction and visuals and style? I want to get for this specific project and for those specific themes. And so just brain dumping everything. So you can find design on like Cosmos, Pinterest, maybe even Instagram or Twitter, X. You can probably also find some of them on mobbin.com. Go to your favorite design resources and try to find those designs. It doesn't have to be huge. Like again, for those, for example, from Apple, I just went on like the Apple iOS or new OS design page that they have something like that. And it was just basically showing the like glass design that they work on. And it can just be like this little design here, this little widget. It could just be like this little gradient color, right? It doesn't have to be massive. Or even if it's just a screen, like for example this screen here, maybe it's not the whole screen. Maybe you really like that comment or search bar right at the top. So that's what you're looking for. And the goal, what I'm trying to do is define, okay, we want directions that are different enough that we can pinpoint easily with the team, with the client, what direction do we prefer out of all of those options. And so for example, like what probably is going to happen for some designers is they're going to like maybe this kind of color gradient and then they will also like this and they will set those 2 as different directions, but they should probably be the same one because then the range, the variety, the spectrum of directions and styles is going to be very short, very small. So what we want to do is explore like this very extreme, this wide range, this wide spectrum of directions so that it's a lot easier to be granular on which do we specifically want to go with, or which options, or like which range is good enough for us and for the client. And so that's what we want to do. The name is mostly more like secondary. So once you've figured out kind of like the ideas, like design references that you want, then you're going to set direction. So for here, it was like very organic, very colorful. This one was more like a floating overlay, something that was again, like very low footprint on the screen, very easy to see, but also doesn't take a lot of space. Something that's pretty ambient as well. Another idea was like to be keyboard-only summoned. So you do a shortcut and something appears. It could also just be like low-profile notch. So like this, for example, just on the sides. So that was kind of like the direction and ideas that I had. And again, the goal is to be, to have that very wide range of options so that we can easily pinpoint which ones are interesting or good enough to go with and to explore. We don't want all the directions to be valid, right? The point is to only identify and to select a few of them, a few winners that we want to go with. But ideally, if you were to set directions, I would ideally try to find between 3 to 6 directions that are different so that it's easy for you to explain the rationale behind each option. It's easy for people analyzing it, reviewing it, and deciding ultimately which one they want to go with. And so this is a pretty intensive work because you usually want to do that in a very short period of time, right? Like you should take probably 1 day, maybe 2 days max, but 1 day is usually like a good timeline for that. And again, you want to choose something specific. Like for example, on this screen, maybe I really like this chat interface. And so I would just take the image here and then have this frame and then just, you know, place it in there and specifically focus. Like, if it's just a gradient, just specifically focus on that gradient. And so that's how I approach it. And so once we have those directions, we can already kind of set the idea of like, okay, actually maybe colorful is not quite right. Maybe floating overlay makes sense. That would be really good to explore. But maybe we could combine that with, I don't know, the, the The low profile notch, like maybe the low profile notch could, you know, when you open it or when you get close to it, maybe it could open up like this. And so you can, you see, like you kind of brainstorm, start brainstorming some ideas, some features, some experiences. And that's where these kind of feature ideas come from. Because then once you have those directions, you're like, okay, what kind of features are we talking about? Because direction is good, but it's not really applicable. You need to apply the direction to a context. How is it going to be applied on the product, on the features that it has? Or the future features that we want to build. So then the features, how do I brainstorm those ideas? Again, I retake some of the screens I have because that's the goal, right? Like when you have this massive brain dump of ideas, of direction resources, you want to figure out, okay, which one would be features ideas, which one would be brand directions. They can be related. They can be closely related. It's like the theme, then you have the direction, and then you have the features. And what's interesting and what could help you figure those out is The features could belong to any of those directions. Like, that's the goal. All of those features can be translated into like organic colorful, for example, or every feature could be translated to like keyboard-only summon, right? And so that's how I kind of think about it. And so then features is like, one of them would be to create and update patterns. And so we want this kind of granular graph representation of maybe data or rules or actions, history, something like that. To help the user have a more transparent view of what's happening. What is the AI or the agent doing or learning or taking into account? So that was a very interesting idea that I think is really relevant to the project. Another one was that transparent kind of status mode, something that's— I think that was like a big theme or kind of like sub-theme of the project. It was to be very time relevant. Like you don't just spam users with notifications, with messages, with stuff like that. It was really. Just at this specific moment, that's when this specific notification is the most relevant to the user. That's when we want to show that. And so that's kind of like the goal with transparent status mode is that it shows, for example, when it's listening, when it's learning, if it's digesting new information that you've just shared, a new document, if it's analyzed. And also just like notifications, how do we show notifications in a way that's not like spammy or that doesn't again take the whole screen? And so that was like an interesting feature as well to have. Another one here, because it's Not only about you yourself, but also how do I explore that kind of like agent level experience with the whole team, not just with yourself, but you can share it. You can use it with multiple people in an organization. So it was super fun to work on that and to think through it. And so for example, here, like the idea was like, okay, how do I make it so that everybody can have access to that? So maybe it could be groups, maybe it could be able to show like activity, like here, for example, and show me like specifically which user has access or has done which task or which agent has been connected to which user. And so that was again, a more granular, in-depth feature than something like a theme. And another one which I've talked about with transparency was context aware. So again, like the right timing to send something specific, to show something specific for the user, or based on which action do you show something. We want this to be almost like an ecosystem so the features can interact and can work with each other. They're not specifically separated. The goal is to make a product that works and that interacts with other features and it creates something that is so important, so essential that the user just cannot not use it. That's the goal of every product, right? Like you make features that work with each other, not against each other. And that's also why you don't build so many features all the time. Like the goal is not to just come up with 20 feature ideas. It's going to be really confusing because the user at the end of the day, a user might only have 1 or 2 things like 2 goals, 2 features that they're going to use most of the time, and the rest is not going to be useful. So that's what we want to be mindful of. And that's why I'm not trying to like push for so many different feature ideas. And so when you think about it, like all of those features could actually just work together. And we don't have to say, oh, I don't want feature 1, but we can keep feature 2 and 3. Like you don't have to do the same process as the directions where the directions kind of set the roadmap and the path that the team has to follow. Features are more like, which one do we prioritize? Which one are relevant to the goal of this sprint to the goal of the product, to the story we want to tell to the client. And ultimately, when you work on this, when you do this 3-step process, this is going to really set the foundations extremely well because it's going to serve not only as a reference point for you and the team throughout the whole project, but also it's an amazing way to communicate and to express visual ideas. Because when you just speak to the, to the team, to designers, to managers, to clients, just saying stuff with words is a lot more difficult, or there are more gray areas that can be interpreted differently based on the person. And so having those references and visuals is really helpful to agree together on the design direction we want to take, and also just to talk about it. Like, your goal as a designer is to design the best possible products and experiences and features, not only for the client, but also for the users. It's not for you. You don't design for yourself. You design for other people. And so even if nobody likes your directions, if nobody likes the features that you've done, you've still done something to move everybody to the right direction because it's a lot harder to define what you don't want than what you want. And so if actually you've helped the team to say, okay, we don't want any of those, that's amazing because you've just deleted like 8 possible ways to do the product. And so that's how I want to try to approach it. It's a great way to again, set the foundation right, but it's also just a great way to discuss the project. and to get some ideas going and to build that momentum. Because like, it's incredible. As soon as you start figuring out those directions, as well as you start figuring out those features, it just keeps flowing. And when you ping pong those ideas with another designer, 2 other designers, a manager coming in, you talk with a client about it and you get their feedback. Like the amount of potential and ideas, how much faster you move toward that goal, that common goal, it's incredible. And so I would recommend anybody working on a product, working on the sprint, To not necessarily copy that approach and that process because everybody has their own way of doing things and there is no one way to design a product. But just take that into account. Think about, okay, first of all, we need theme with like some keywords that are relevant for the project and that is going to help us figure out, okay, what design, what references, what visuals do we need to bring up in there? And then you're going to start kind of like grouping them together, figure out, okay, which ones are matching together, which ones have a relevant or related connection and direction. And then you just group them in, in again like those. They just make 2 squares or 2 rectangles and just put those designs in a way that makes sense and that highlights specifically what you're talking about. It doesn't matter, like here you can see, it doesn't matter if it doesn't take the whole space. So you don't have to be perfect about it as long as it clearly outlines, it clearly shows what directions it is about. And then you can just come up with feature ideas. This is an amazing way to do it because Not only do you show initiative, but you also show to the team that you are already thinking about the future, thinking about the flow, thinking about the product on a much deeper level than most designers would do at this stage. And that's it.