Transcript: Rani Bar Hama | Similarweb | Building Digital and Physical Products: What's Similar and What's Different?

Intro

Eyal David: [00:04] Hi, this is Eyal, you're on Product Builder. We're about to start the episode, but before that, hit follow on Spotify or subscribe on YouTube and help us reach more ears. Let's start. Hi Rani, how's it going? Hey, what's up? Great. With us is Rani, who comes from Similarweb, and we're continuing the builders series. Rani is a builder himself, so we'll touch on that, it's an interesting episode between physical and digital.

What Similarweb does and what Rani leads there

Rani Bar Hama: [00:32] Rani, so you're basically Senior Director of Product at Similarweb. In a word, can you tell us what Similarweb does? First of all, it's fun to be here, thanks for inviting me. Similarweb is a global data company that analyzes and measures the digital world, which is basically websites and apps, and provides insights to its customers to help them make smart decisions and basically lead the market they operate in. I've been there a year and two months, leading one of the product groups at Similarweb. Okay, tell us a bit about your path, how you got to Similarweb. So I'll take you a bit back to the beginning. I come from the world of product and UX, that's what I've been doing for roughly the last twenty years. I started way back at Amdocs, and even before Amdocs, I actually got to the product world because I wanted to be a journalist at first, that's what I wanted as a kid. I love to write, I love to create, and I created through writing and said I'll be a journalist, I started studying advertising. I started studying communications, and along the way I made the switch, I realized I like technology much more and connected more to those things, and I actually started way back as a webmaster for the Amit network, I built their websites, that was my first touch with technology. From there I more or less rolled into Amdocs. I was at Amdocs for almost five years in a role that's pretty unique, we were basically a group that built interfaces for the company's customers. Now it's like vibe coding, but where I do the vibe. Well, what you told me is a real builder, you were a designer, a builder, taking requirements. You get, you sit basically with that account manager who specs out for you some kind of feature they want in the system, and instead of getting back a product spec or wireframes, you simply get an already working front-end system based on HTML, JavaScript, CSS, you know, things of back then, a kind of prototype that he presents to the customer, and assuming it works and you get change comments, that's actually the solution the customer ends up getting. And from there you went on to Orange, right? From there I went on to Orange, that's basically where I shifted to the world of product management. I was at Orange for eight years, I think the eight most significant years in my career. Orange was a very very big playground for me. Think of a kid who comes into the product world at a stage where the telecom market was, the company was in a state where it started moving to what's called value-added services. Meaning, we already have the data, we have the voice, and we need to find new growth engines. What do we do? Let's build products. There are no products? Let's start seeing what to do, by trial and error. For those too young, we're talking about Partner, and basically we're talking about a period that's before the iPhone, if I understand correctly. The iPhone arrived in the middle. I was, just saying, the third person in Israel who had an iPhone, by the way, because I had to test it. But listen, you're giving away both our ages here with that question I think. On one hand yes, on the other, but this playground you called it, it's important to say, it was a period when whoever controlled the operating system was Orange at the time. Right, so Orange had the biggest asset in hand, which in the end is the device, the hardware, it had of course the cellular connectivity, and it had customer loyalty, because you couldn't switch either, you know, it was before number porting and all, you couldn't switch every second. And on that medium you could develop new value-added services. Value-added services, what did they call it back then? VAS. Okay. And where did you go after those eight years? After those eight years I moved on to the gaming world, I came to the gaming world at a company called NeoGames, I joined NeoGames as Director of Product and actually built the player experience team there. I entered gaming with no experience, I didn't really understand anything, and I got an opportunity to lead this team and basically build it. And that was the first time you enter a world that's new to you, you look at it through your own eyes, and you discover that in gaming people look at it through the eyes they've always looked through. Many times in the sense of, you know, it works so let's not break it, and I came to break things a bit, so like, sometimes that created some clashes, but slowly things aligned. I was there for a significant period of three years. I basically led the building of all the company's gaming mobile apps there. I led the building of the company's lottery product. A significant period that taught me a lot. What's the most significant thing you learned from that period? First of all, how to take over a new domain. You sit in conversations, I would just, I told my wife, I'd come home sometimes and like, in the first period I'd just sit in meetings and not understand what they were talking about. In the end I had to build the skills, so first of all taking over a new domain, which was a hell of a thing. And second, how you actually bring your view into a company that already works a certain way, into an industry that looks at things a certain way, you have to prove yourself much harder so they accept your opinion and what you want to bring. And that made me work very hard but also taught me a lot. So it sounds like build, be creative, but don't break anything along the way. You have to build, first of all you have to not, not break, because you know, gaming is an industry where you have lots and lots of concurrent users online, every second matters, and when you break, you break. So on one hand you need not to break, on the other you first need to convince that you can improve and then actually carry out the improvement. Where did you go from there? I continued in gaming, I just left Israel, we got a relocation offer to Gibraltar, we left Israel for a period of five years, me and my family, in Gibraltar. I basically spent five years at British companies, big and massive ones, in the gaming world. I started there at Coral, which later became Ladbrokes Coral, after that I was at William Hill, then at Mansion. At these three companies I actually did a fairly similar role, which is Head of Product Gaming, a kind of responsibility for the whole gaming product world in the company. Now, it's important to explain, gaming in this sense isn't the game itself but the whole wrapper around it, the player's entry into the app, the lobby, how he signs up, how he deposits. How he withdraws money, how he withdraws money. You basically need to create a very very holistic, seamless player experience, as much as you can, that will make him come in more, stay more and play more. Excellent, we have excellent episodes, we'll add that in the show notes around it, excellent. And basically after that you continued in Gibraltar before you came back to Israel, how was that? No, actually we completed five years there at those three companies and I came back to Israel, I was another year in Israel at the last company I worked at, Mansion, I was kind of on the line between Israel and Sofia, and then I decided I need to settle down for a moment. You know, it still took a bit of time to readjust, and then I found myself again in a new domain, insurtech, a world again completely new to me. I formed a very interesting connection with a very talented founder who founded a company in insurtech called Hahoy. A company that basically provides insurance policies for boats in the United States. It turns out to be a very interesting market that also makes a lot of money a year. I was VP Product there for a year, over a year, almost a year and a half. Also a really interesting product, a field that's new, again bringing your thinking into something that already exists. There I dealt with English guys, there I had to deal with navy guys, you know, there were all kinds of challenges, but also a very very educational period, amazing people. Unfortunately it didn't really succeed. And I moved on from there. I moved on to a company called Netspark, an Israeli company that develops or developed diverse AI capabilities in many fields, and on top of them it basically built a product line that I actually came to manage as VP Product. One product line is apps related to parental control in the United States, and a second product line related more to the ability of law enforcement agencies in the United States to identify pedophilia at a crime scene. Very quickly. We'll spare the details a bit here, but two products seemingly very very different, yet there's a pretty clear line through them. I managed two squads there, each squad basically handled one of the products end to end. I was there for almost two years too, and today I'm at Similarweb.

Rani's path: from Amdocs to product (and how he started as a builder)

Orange as a "playground" and value-added services

Entering gaming: learning a new domain with no experience

Relocation and British gaming: end-to-end player experience

Insurtech and AI: new domains and what you take from them

Eyal David: [09:05] Okay, so to wrap up for whoever has stayed with us until now, a magnificent career, lots of fields. And I'll ask you actually a question I always ask, not always, but I'll ask you, what in your eyes are the most important product skills in product people?

The three most important product skills: user, glue, curiosity and focus

Rani Bar Hama: [09:20] Okay, so here too I need some kind of MVP, because you know, a product manager needs lots and lots of skills, but if I need to focus on three, three skills, then actually I'd go first of all with the world where you see the user. Meaning, for me the product manager is the eyes of, he's the user's representative in the whole thing, okay? He needs to know that today he's building something for an end user so that the user uses it as easily as possible, creates value for himself, and keeps using it. That's point one. Point two is the whole world of, I like to call it a kind of glue, you need to be the glue between lots and lots of parts in the organization to reach the goal. Usually product departments sit somewhere in the middle, between the world of business stakeholders, the folks running the P&L and so on, and the world of development, and you need to create the connection between them, to make sure the business strategy is translated correctly into product strategy and then also executed. That basically includes the ability to activate all these parts of the organization and glue them together, often, or most of the time, even when you don't really have authority over them. Lots of communication, it starts and ends with communication, which is a very very important part. And the third skill, I think I'd go to the world of curiosity. Routine eventually beats us, and to get to the next floor you always need to look a bit from above and look at things you don't look at day to day, and in that context also know how to filter out noise, meaning know where to focus.

Eyal David: [10:59] Focus on the more important things and put the less important things aside. An overall package. Excellent. Okay, so that brings us into the episode itself, the main topic. So as we said, we didn't call it builder for nothing, because you build both physically and digitally. So let's actually start from the side, everything you've built so far we understood, we heard at least in the brief. But let's talk for a second about what you build with AI, okay. You brought some example of something you built, right? So first of all AI, building with AI on the side

How Rani builds with AI and what it changed for him

Rani Bar Hama: [11:33] opened a window of creativity for me that somewhere, you know, was already on its way to closing. You know, in the end you work somewhere and your ability there to deal with actual creation, at management levels, shrinks over time, and suddenly I found myself sitting at home with the ability to use AI as I wish. What did you use? So I'm a Base freak, I really really love Base. I actually started with Lovable, moved to Base, but today I'm already at a place where I'm going a bit broader with it, you know, to get into the worlds of agents and things like that. But at the core I started with Base, and I stick to it. So let's dwell there for a moment and continue. No problem. Tell us a bit about Base, why you prefer Base, maybe you have some winning tip for Base. I, so, so in Base, I'll touch in a moment afterwards on creation in the physical world too, but I approached building with AI in a way that wasn't right. Meaning, I approached building with AI the way I approach building in my physical world. I threw all the tools on the table, started thinking about what I'm doing, or in other words, I wrote one big, magnificent prompt and hoped for the best, and then I started telling it, fix this for me here, fix that for me there, and very quickly I got to spaghetti, and that's where it, that's where it stopped, it became much more complicated. Until eventually I understood, I read, I researched, and I basically started working with a kind of prompt strategy. Meaning, first of all, if you're building something really significant, then sit aside for a moment, spec out for yourself what you want, write some kind of PRD of what you want, convert that PRD into a prompt strategy, which means a set of prompts that will get you the expected result. This set is also already built in some logical order that isn't supposed to break one another, under each prompt you basically get what you asked for, what you're going to get, and what your acceptance criteria are for what you got, and that enables building that's much easier and calmer, and Base was great with that. And that actually connects to the question of what I build. So you know, I started with things just to get my head straight, all kinds of small silly things. I built an app where you give it a quote and it brings you the relevant Seinfeld episode with the mention, or not even just a quote, write "car" and it'll bring you the episode and tell you what it was about. I built a small app that lets me track everything I watch on TV and get reminders and when and where it's airing and so on. I built a very interesting agent that lets me locate parts for antique watches, which I deal with as another hobby. Things like that, but really really for personal use. And the first bigger thing was, as I started to say, my wife is actually a Hebrew teacher and she teaches reading. So she teaches reading at the Ministry of Education, of course that's the day job, and we talk a lot about how she does it and what she does, and she told me, listen, basically I'm probably doing what I do well, I have a waiting list, but I don't have time, the resource I need to handle for scale is time, and that's it, what will I do, there isn't any. And we started thinking about what to do, and basically I told her, listen, I have some idea of how to take the world where you need to sit one-on-one with a student, using a method you apply, and translate it into a world where the student talks in front of the computer, and that way, in the same unit of time, you can handle more students. And she, like, specced it out, really, like, I asked, to tell you that. No, no, I told her don't flatter me, you're not just coming like that, write a proper spec, say what this is and then this and then that, and she really sat down and specced it out. I took that spec and basically built a PRD from it. I approved the PRD with my main stakeholder, which is her, and then I built a prompt strategy around it, and it took me a relatively long time, but I built her a system that lets her do two things. One, to orchestrate the whole world of these private lessons in one place. Starting with how you set a schedule, and two, how you manage payments and things like that, and those are the simple things. And the heart of the story is the ability for her today to record herself reading and let the student sit in front of a screen that defines a task for him and lets him follow the reading, and it also knows how to measure how much he read and in how much time. And when he finishes, at the click of a button it gives him a dashboard that shows him his pace and how much he's progressed, and we also added gamification that gives him a coin after every lesson, and then she knows how to make a claim for what he should get after a number of lessons. In short, very very nice things that really really improved her life and helped her a lot to advance this thing, and me too. How long did it take you to build it? It took me around three months, which, forget it, I don't work on it every day, I work on it mainly on weekends and at night.

Why Base44 and how to avoid prompt spaghetti

Tips for working with Base44

Examples of small side projects he built

The system he built for his wife: scaling reading practice, dashboard and gamification

Eyal David: [16:13] Forget three months, and tell me about the prompt strategy you mentioned, do you have something that collects it, some good link worth reading around it?

Rani Bar Hama: [16:21] One of my problems with AI is that you're exposed to so much information that in the end, like, so I have a method, I just have a WhatsApp group with myself where I send interesting links. There are a few folks I can recommend who create excellent content, I can give that. Excellent, we'll add it to the show notes. Excellent. So Dalit was happy. Very. And this thing is running now, I understand. So this thing is running now on her personal side. Suddenly, you know, we started thinking about it a bit, maybe a bit bigger, but let's say, first of all the proof of concept on the first user.

Eyal David: [16:53] Works, doesn't work, and from here we'll move forward. Excellent. So okay, that's really an example of what you do. Now, wait, let's move to the second thing you do. Yes.

Physical building: functional art from "junk" and 109inside

Rani Bar Hama: [17:01] So the second thing is the physical world, it's creation in the physical world. Like, we've been talking for a few minutes already and the words creation, creating and building came up here, I don't know, what, five or six times, so it's a very very strong motif. At the core I really really love building things, I love creating, I love building, it's part of who I am, I look at the world that way and that's what I love to do. So luckily I work in a job that allows it, meaning I basically build things, then we expanded it to the AI world, there too I build things. But it exists for me in the physical world too. Since childhood I know how to work with my hands, meaning, I really love working with my hands and I love building things. And over the years I developed a hobby where I simply create functional art from junk. What is junk? Junk can be something I see next to some trash can and pick up, it can be something I bought at a flea market or something like that. And I have some vision in my head, I bring it home, I have a studio at home with all the tools I need, and I just sit there and start some creative process whose end is some kind of unique item, one of a kind, usually functional, it can be a light fixture, it can be a table, it can be a cabinet, it can be lots of things, but made of junk or of something whose purpose you changed literally by 180 degrees. And that's something that in the end also meets a user, it has a price. Yes, it's something that... So again, I also have a brand that collects this thing, called 109inside. Not a brand because I'm some kind of, you know, now some big brand selling thousands, I just really felt like gathering everything under one title, so I simply built a site called 109inside, and I also have an Instagram account that supports it, and all my works just go up there. And if someone wants to order, we'll put a link to that too, totally, very beautiful things, really. And in the end you mentioned it meets a user, so yes, because in the end I really love building things for people, meaning people dear to my heart, and I need to give them a present for a birthday or an event, so I don't, I build them a present. I don't give them a present. And in the end, you know, you build for a user at the end, you need him to love what you made, you need him to connect to it, to also find the value but also understand why you made exactly this for me,

Eyal David: [19:19] appreciate it and use it. Listen, that's lovely, we'll put a link, these are really gorgeous things. Thanks. You also need a big house I think, to have a workshop and also, where do you live? I live in Shoham, I don't live in some estate. But I imagine you moving with your workshop through this whole life journey you described. Just before we continue, click the link to the Product Builder WhatsApp group in the episode description and join the community, where you can continue the discussion, ask questions and be the first to get updates on new Tachles episodes. We're waiting for you there. And now back to the episode. Okay, so now that all the background is clear, we want to talk about the difference between a physical and a digital product, and let's even take the digital one to an AI product so it's interesting, okay. So there are all kinds of aspects we can look at. We can start right from the beginning of planning, right? In physical, for example, you talked about how, well, I find something, I have an idea and I run with it, right, there's no meticulous spec or anything like that, versus what you said about an AI product, where you actually have a strategy and you tailor the whole thing.

Physical vs. digital vs. AI products: planning, scale and versions

Rani Bar Hama: [20:28] Tell us a bit about that. So it's interesting, because let's take it for a moment before AI, so at work obviously when you build something you don't do it haphazardly, you plan and you check data and you check who your segment is and who your persona is and what the business need is at all, and you try to reach market fit as fast as possible. Meaning there's lots and lots of background work so the process eventually succeeds. And... and in the physical world, I'm putting AI aside for a moment, for me at least it works pretty much the opposite. Because if you put me today in front of a manual and tell me, listen, here's a thing that's so cool to build but you need to follow the steps, you've lost me, I, I, I, I switch off, because it interests me less. I love the... I love the challenges, I love the... finding the problems, suddenly I have something and I don't really have a connector like that, so I need two connectors, and that's what I love, that's what does it for me. And in the world of, in the world of AI, at first I thought I'd found a duplicate of that, because I said what? Just like I throw all the tools on the table and start juggling with them, I'll throw in all the prompts without much planning and let's go. And you discover it's not like that, that you really, you can do it, but the result you'll get is that after two or three prompts it will start to tangle itself up, and getting out of there is a whole story. So you do discover you need to combine some element of planning, some element of preparation, which is very very different from the physical world, where chaos is also a kind of fertilization, because you simply.

Eyal David: [22:00] I think it's also very related to the level of complexity, right, a bit earlier I started from the first products you built, the Seinfeld app, that's why I compared it, and then we can get to Dalit, so there you didn't do any discovery or anything like that, that's why I jumped straight. But come on, let's continue. So basically, do you also plan in physical? You talked about doing a one-piece, right, right, right. So, so usually I don't, I don't have

Rani Bar Hama: [22:25] any production line, my resources are me actually, and usually something I build is a one-time thing. It can also be because that piece I found simply exists once, so I work only based on it and make only it. So in the end you get to a state where you make one product for one user, which is basically pretty easy, you need to know who the user is, you probably also know him well, and the product simply delivers what you need. And in the real world there's what's called scale, you can start from one thing and then you need to start growing it, because otherwise it will always stay small and unprofitable or whatever else you want it to be. And that's something that's really really different, it requires a different strategy, you always need to look much bigger, you always need to look in multiples, you always need to say okay. Even if today I reached an MVP that includes three features, I know that in the big picture I need ten to create that scale, and I need to know how to plan that already. And that these features are already sitting on the shelf waiting and they're specced, and then I need to start playing with them like some kind of Lego, when I'll put this in and when I'll put that in. And overall it also takes into account, think how many variables or how many information sources we'll need around it to make these decisions.

Eyal David: [23:38] At my table I'm alone. Right, I have to ask you, before we even get to marketing and sales, how much does this hurt the creative process? When you're free now to take, you know, something and throw it on the table and start flying with it, versus now going into planning?

Does planning hurt creativity or strengthen it?

Rani Bar Hama: [23:52] I think many times, it's funny, many times when you're inside the planning your creative process can be even stronger. Why? Because basically your challenges are stronger, your challenges are more significant. In a solo creative process it's my problems with myself and with the object in front of me on the table. And it's, you know, it's challenging and all, but in the end it's... No, but there are limitations here too. Right. This part doesn't fit, this part... Right, but in the end these are really localized problems, they're usually tied to one or two variables, you know, say the electricity doesn't connect, it can be either because of this or because of that, like, there's a limit. And there's another thing, that over the years you also learn, meaning, if someone runs into some kind of problem, I already know how to solve it in a way... Well, that's maybe also true. Yes, but in the real world, like at work, when you actually need to build something, you're one of many, meaning, you're one of many and you're one function within the process, there are lots and lots of other functions. I can talk about the spec of the feature I want and rely on that data, but I need the data to work on it, I basically need to work with an analyst who also has his input, and I want the input, I embrace it, bring it. I have user experience that comes and gives me the layer of how this thing will really look and function, and in the end I also have the development world that comes and tells me, listen, it's better we start this way and not that way, and also take into account 1, 2, 3, 4, which are insights I really love to get. By the way, there are a few types of product managers, there are product managers who are very centralist, meaning, you, you have your opinion and you want people to start falling in line. And I think that actually in places where you work on big things it's really convenient and good to convene at the start some kind of thing that we call a kind of initiative, of the relevant people around that problem, and each one basically pours their 2 cents in, and you reach a whole that's much bigger, much more mature, with much better chances of success.

Eyal David: [25:46] Which is true also, I think, in physical creation too, right? If you put four people who are experts in their field to think together, it will probably avoid many more pitfalls, but maybe it will hurt the creative process.

Rani Bar Hama: [25:57] Yes, but it may be, but for me it's because the creative process is very private, it's very personal to me, I'm also the one who, like, doesn't need to take other opinions into account. Meaning, I don't find myself, if today I'm just painting some component and I'm debating with myself whether it's this green or that green, in the end I made the decision and that's it. And if I suddenly need to bring into the mix, wait, what will he think, then it takes me a bit out of the side of it, or out of the way I see it, in the end it's mine, like.

Eyal David: [26:27] There's another thing that the question raises for us, that basically, again, we'll get back to marketing and what you do with the thing, but if you even really define some version, there aren't versions here either, as you said. But if I define some version, then I'll support a few use cases, and in a digital product I'll support them. The physical product will always be there. Always, always they'll see it, and you can't switch it off, right, right. So there's also an element here, I think, of aesthetics in the space, I don't know, you tell me.

Rani Bar Hama: [26:56] So first of all it depends on what kind of product it is, you know, not all products are products that sit in the space and you see them, but let's say most of the things I make are like that. And then you really need to take into account the transition, who the user is, everything, in what space it's going to sit, is the space big, small, functional, not functional, practical, not practical, things like that. It can also depend on the materials you use. I'll give you an example, I built a table from a bicycle. And when I built the table, so, so, my grandfather's bicycle. How are they, there's no horizontal support, or do you sit on the chair? My grandfather rode a bicycle his whole life. And I was very attached to my grandfather, and when he passed away it was important to me that this bicycle be preserved, and we were abroad then, and I told my mother, listen, do whatever, don't throw this out, it's mine. And when we came back I basically took it and said I want to build something from it that I'll see on a daily basis. And at first I thought about a bookshelf and that worked out less for me, and then I had the idea of a table. And I simply took the bicycle and cut it where needed and created some kind of base for it, and it stands in my study. And apropos what you asked earlier, my first intuition was simply to put a wooden board on it. But then you realize you don't see the thing at all unless you bend down. So you say okay, not wood, glass. You put the glass, that's your breakthrough. And that really connects to what you're saying, to how this thing sits in the space and things you need to pay attention to, and these are also often things that, you know, are hard to change on the move. Meaning, you already made something, and changing it is a whole other story. Compared to changing a version of some app that I just shipped and it's stuttering a bit for me.

Eyal David: [28:34] There's also a user experience here that needs to be timeless. I don't know about timeless, but there's a different look at user experience here. Practically too, you mentioned that, which is interesting, but also like

Rani Bar Hama: [28:43] expand on that a bit. Listen, achieving something timeless in design is the very very very edge of art, there aren't many like that. Meaning, there are many things that, you know, also, you'll see, put some very beautiful picture at home, there's an element where you just get tired of it after Samson and David or something. You can get tired of it after a while, like. So here, beyond functionality, yes, it needs to be something you'll love, that you'll always be excited by and love, so you won't want to throw it out after two months. But here the element of your connection to the object also comes in. Meaning, many times when I want to build things for people, I can tell them, listen, you can bring me something that speaks to you, that you want to preserve, that's important to you, and let's make something else out of it, give it some twist. And then at the core, at the core, it's something they're very attached to and want, and so they won't. What does the process look like for you, do you get requirements beyond that? So again, I don't really have customers, meaning no, but I mean, say now you got, it's very different. There's, it can be like, wow, make me a lamp and I won't tell you how, but make me one, and then I say okay, the person is like this and like that, I'll make it for him. But most of the time it's like, wow, I'd really love you to create something for me, to make me something, so I, like, understand what the person loves, what he wants, and according to that, and according to that, in the end I really build it.

Eyal David: [30:02] And in terms of marketing, so now we'll touch on it, so really there's no big marketing here, like you finished, or you show the building process or something like that, share it and then sell, what does it look like?

Rani Bar Hama: [30:10] No, I do it, I photograph the process a bit, often more as a before-and-after to show the story, I have the Instagram account of the 109 that I upload to, and after that it simply goes up on the site as another item in the portfolio, like. And my intention, you know, is that people come in and see, so it can give them ideas and they derive from that. Whereas in a digital product you wouldn't be looking for any inspiration now in order to sell or something like that, you'd start from there already. Right, because there I'd need to make sure, before I make anything at all, that I even have a chance to sell it, that anyone wants it at all. And how much they need it, and for how much, and what the business model is, and how much time, and ROI, and race, you know.

Eyal David: [30:52] Let's talk about the work tools you work with. Okay. Tell us about it a bit. You talked about how you basically lay everything out and you start working.

Personal tool stack: chat, Docs, Claude and screen layouts

Rani Bar Hama: [30:59] What does it look like with a digital product? Your work tools are, if you work with AI, cool, but it's pretty similar, no? So today, because I deal more with... let's call it management and less hands-on, so you know, my work tools are mainly my mouth and my ability to sit in meetings.

Eyal David: [31:19] Yes, but put yourself in the app with Dalit. Yes, but there, so again, so there my work tools basically, you get into the whole spec stage that you do in a kind of, let's call it, user ideas, what kind of tool you use, then how you actually

Rani Bar Hama: [31:35] get what comes out of the idea into some kind of document. So you sit and write it in Google Docs and then you go with it to Claude and ask Claude to make a more PRD version of it for you, and then you start refining. And then afterwards you do another stage that again depends on which tool you're working with. In this case I worked with Claude and built the prompt strategy through it. Why Claude specifically? Because I found that Claude, at least for me, in the whole world of connecting to the product world and the code world, works better than GPT's take, it kind of treats it in a way that seemed to me more meticulous, and it just flowed for me. How do you even compare, do you run it in both and check what's what? Honestly that's a really good question. Usually I read the opinion of someone I appreciate and I kind of go with it, and only if it really doesn't work for me in a really serious way do I change. But sometimes there are cases where I'm building, say, an interface and I'm really curious to see how two tools treat it, and I'll throw it into Lovable and I'll throw it into Base and sometimes even into something else, and I'll see what the output is, and it's really really interesting to see how these things come out, and often it didn't change my choice, but often it gave me maybe another idea that one tool gave and the other didn't, didn't cover, another way to approach

Eyal David: [32:46] a certain design or a way to lay out a screen. And how should you measure, I'll ask this now from the end, so say now a physical product, you finished building it, you do some QA on it, right? If it's, as you said, electricity, I'll see a bit, but during the building process you're already basically doing it.

Rani Bar Hama: [33:02] Yes, like, there's also the building process, you know, it's divided into parts. You're building a light fixture, there's a chance that the electrical cables, which you thread now, after I advance another stage in building I won't be able to pull them out anymore, so it's important to test the electricity at this stage and not at the end. So it's, as said, and that's also the critical path. Yes, exactly, exactly. But if I go for a second now to a product like that, to a digital product, AI,

Testing, regressions and freezing parts that work

Eyal David: [33:25] actually here testing is a whole event. Testing is a whole event, right. How do you deal with it? Think of a system split into ten features,

Rani Bar Hama: [33:33] and each feature is currently specced in a separate prompt. And what you do with the AI is basically discover the right way to build these features. What's right to build first? Is it right for me to first build this component, and then that one, or something else? And then once you've cracked that, and you basically have a prompt for each component, it's not just that there's the prompt and I throw it in and that's it, I also get a set of things, I define, I tell it, listen, I want you to build this and this and this, and then here tell me, like, what tests I need to do to see that this thing really works. And does that really prevent regression for you? Unequivocally. Gorgeous? Unequivocally, because that was a very big change that I learned the hard way, like, you know, you added a feature that now adds a record to a table, so it tells you, you added a record, check, you added, it was added, you deleted, it was deleted, you passed. Another thing by the way that Base has, and I assume other tools already have it too, is that you have the option to freeze parts. If you already have something that already works, I simply freeze it, and then I know it won't be harmed and I continue with my strategy.

Eyal David: [34:33] Tell me, and in terms of measuring success in the end. In physical you can measure success pretty simply, right? You get feedback.

Measuring success: a user's smile vs. tracking and events in digital

Rani Bar Hama: [34:43] Or you're pleased with it if you built it for yourself. I do, I need more, I'm pretty pedantic about these things, meaning there are things that even if I do them quickly, quickly, quickly, I'll want, for myself, to meet some standard that's important to me. So that's basically the first criterion. Measuring success is indeed easy, you know, it starts with the smile of the person receiving it, how pleased he is, then you see he shares it, you see that... It's already happened to me a few times, you know, that a group of people came into my home, and say there was someone who had already been to the house and someone who hadn't, so he told him, come, come see, I have to show you something. So you understand from that that it's really successful, that's in the physical world. In the digital world, you know, it's different. I think in the digital world measuring success is also much much more Sisyphean, meaning, it's not something you launch and forget and move to the next thing, you always need to have some chapter. Okay, let's talk about launch here, right? Yes, so you know, you launched, lots of pressure is released, tension and all, but here a new stage begins in this whole product management life cycle, which basically includes, what's called tracking. Meaning, I need to come and, one, have tracking capabilities at all, have the right events that I put in place, and have the right dashboards, and then I need the mindset. I discovered there are people who don't have the mindset for it, meaning, they're done, they move on. And you basically need to start tracking, sometimes in a Sisyphean way, the things you shipped, to see if they function as you thought, to see if their pace of progress is right,

Eyal David: [36:15] easy because it will jump up right away, but it continues. Right, I was actually thinking about pivots when you talked. So in a physical product a pivot will happen only when there's no choice, when the material stops you, right, there's something you didn't plan.

Rani Bar Hama: [36:28] Or you suddenly feel like a new direction. Yes, does that happen to you? Yes, you know, or you suddenly discover a new need, like, I took the example of the bicycle, so I did a pivot from wood to glass. Or suddenly, or you discover that something, you can't, no matter what, you can't, or a new idea suddenly pops up and you say, wow, this is much prettier, much more elegant, I'll do it.

Eyal David: [36:48] And tell me, when you set out now on a physical build, do you plan and know how long, do you already know how to estimate how long it will take, how does it work, until it's ready, like?

Rani Bar Hama: [36:58] First, it's hard for me to estimate because I don't know how much time I'll have to put into it. Meaning, I, you know, as I said, at night it's harder to build because it's, you know, noise and so on, but it's mainly on weekends. And honestly, I also need a muse for it, meaning, there are periods when I approach it less, I'm less in the mood. There are periods when I approach it more, even, like, I don't want to say as a therapeutic tool, but it's a kind of outlet. Meaning, many times I want to forget for a moment, to feel a bit that I'm letting go, I go in and build. So estimating how long is hard, it's also, you know, because it's also not a process that's an orderly process according to a plan, where you can come and say it'll take me between X and Y, you just roll with it, that's also part of the fun in my eyes. Well, maybe now that AI has taken over the world, it's good you have this hobby,

Eyal David: [37:44] you'll be able to work in a more orderly way, and so basically, tell me, do you use AI, we talked about it and I didn't remember the answer, do you use AI to plan?

Rani Bar Hama: [37:52] I use it here mainly, like, in this world, mainly for research purposes. What, like there used to be Pinterest, so you basically go in, and... Also, but in other places, for example, I just want to follow some trend in a certain world, just, going back again to the world of watches, I'm interested in some specific trend, so I, I have, I mentioned earlier, I have some kind of agent that knows how to run after release notes, releases, like, of things like that, and simply bring me a summary of it. Many times I want to focus on something specific, so I just switch, I just go into research mode in one of the tools and research it more deeply to understand. And that's also true for the real world. In the real world we use tools like these a lot to make our work easier, especially, you know, in places that involve going into or examining lots and lots and lots of information, which would take a very long time to analyze today. I'll take as an example, user reviews in app stores, we built tools that let us simply run over them automatically, get lots and lots of insights and plan accordingly. So yes, like, in general, my approach, like, everywhere, this is what I ask myself, like, almost everything I do, I ask myself whether AI can do it for me, and whether it's easy or hard. And usually I try to challenge myself and indeed go to those places. I think it's important, I think it's also a very important key to staying relevant as time goes on, and yes, that's why I believe in it. And I thought about something else, also, like, getting into the thing, getting into the field,

Building is easy today. The hard part is choosing what to build

Eyal David: [39:23] what's harder? From zero to... you know. Getting into the digital field and the field of... digital, forget it, getting into being a product person now or a product network, that's complicated. Let's go for a second to starting to actually build products and being a builder.

Rani Bar Hama: [39:38] I think it's pretty easy, because in the end, again, think, if you had, if you had asked me a different question, if you'd asked me five years ago, tell me, how hard is it to open a business that builds apps or builds something, I'd have told you, listen, I need this and I need to hire programmers and I need this and I need that. Today you don't need anything, you need 50 dollars a month for your credits, and honestly you need to have some kind of love or passion for the thing, and in the end the time to sit and invest, and then you find yourself sitting in your room in Shoham at three in the morning. You know what, I have to interrupt you, not exactly, I'd actually say

Eyal David: [40:16] discovery, or the business process that precedes the thing, that's the muscle I think. Like you said, I can build lots of things, but something to stick, and that I need to sell it.

Rani Bar Hama: [40:26] So, so I don't look at it as the business process, I look at the idea process, okay? Like, what's your idea, basically, like, businesswise, no, I can also make something that has no business potential, but in terms of the idea, for me it's cool, it will give me, as Rani, value, and so I'll do it, and that's enough for me. So I think it's not for nothing that they say, you know, wherever the market is heading today, in the end the people who start getting the most value are people who can bring the ideas from behind, who can drive these wheels. Listen, it's not far-fetched to think that in X more time one person will sit and orchestrate a system that does everything A to Z, like planning, building, going up to the stores and everything, but in the end you need someone to drive it, you need the head that thinks, the head that brings the creativity around this thing. Totally, I'm just saying you can build, but everything, as you say, is really easy. Yes. So actually what to build is super significant, and that's the muscle that maybe takes more time today. Right. Assuming you're building it for purposes that are completely business. Ah, now I'm now making Dalit's app again. Yes. Business, granted, but for her personal use, yes.

Eyal David: [41:31] You knew how to define the need really well, who will use it, and you've probably also already thought, as you hinted, that maybe it could already be some SaaS product, for that matter. Interesting, like, otherwise you could have built something that doesn't hold water. I don't think without this muscle it would have come out like this. I agree, I think yes, you need this muscle too. So let's go back, so what seems more complicated to you to get into.

Rani Bar Hama: [41:52] This muscle, as you define it, requires abilities that are sometimes much harder to acquire. Meaning, abilities of thinking about what will really work better, and then you need experience to understand what will work better, and that already requires, and that basically requires thinking that's a bit deeper about the persona at the end. AI also lets you close gaps in lots and lots of things you don't know and understand less. You can actually sit and consult with it and it will be your coach and it will be everything. You can close lots and lots and lots of gaps that used to be very very hard. You simply need to be smart about operating it and know how to operate it and what to ask and how to say it and what to say. And from here I think the gaps are very very much closable, much more easily than they used to be. Excellent, yes, taking it somewhere. So now, if someone wants to get into product and they have no experience, and now it's really hard in the market.

How to start building a side project to get into product

Eyal David: [42:40] Finding products, in my opinion, is a great way, especially as side projects, even, you find time for it in your busy schedule. How would you suggest starting with it, again, you know, in order to get into the role, which is what people. I think today AI, or all of vibe coding,

Rani Bar Hama: [42:54] basically lets you create a kind of environment that closely simulates what you're supposed to do as a product manager, even at home, in your room. And it's an excellent way to start, because as you said, you know, we just talked about how you prepare for such a thing and what research you do and how you built the initial PRD and then how you started building a prompt strategy, that's lots and lots and lots of abilities that, if I translate them for a second, wow, that's prioritization, that's gathering insights from stakeholders, that's data, that's this and this and this, and then in the end coming and building the thing, which in the end, I say in quotes, the orchestrator who builds this thing, it's something that gives something. I think also many times when you come with some kind of asset like that and you say, wow, listen, I built this and it does this, and maybe you even make a few shekels from it because you brought something that's cool, it's something that can open for you, listen, basically it can get you jobs, honestly. But yes, I think it's an excellent way to do something on the side, understand whether it suits you at all, whether you connect to it, make the leap from it into the real world, and in the end use what you have as, here guys, look, I also did this. And that also often, you know, breaks the ice, gives a smoother entry on the interpersonal level, like it's less. Looking casual, I also think. We'll talk less now about numbers and formulas and so on, like, let's open up for you, I'll show you everything, good.

Eyal David: [44:21] Come on, so if Similarweb is hiring, then do it, go to Rani. Rani, what fun it was, it was interesting, I hope we touched on all the differences, I think we did, we touched on everything, okay, and even if not they'll tell us, but it was lovely in my eyes. We'll put links to all the things. Tell me, if people want to get in touch with you, what's the best way? So the best way, like, I'm on LinkedIn obviously, and there's the site we'll add, which is 109inside dot com, both the site and its Instagram account, and that's it, talk to me, I'd be happy.

Wrap-up

Rani Bar Hama: [44:52] Lovely. What do you wish for yourself a year from now? I wish for myself that I continue to get excited by creating, that creating continues to drive me. Beyond that, you know, health, family, everything else is nice bonuses, but that's in the end. Lovely how you framed yourself with AI resilience and you're still, that's lovely in my eyes.

Eyal David: [45:16] I, you know, part of the abilities I bring to the table. Gorgeous, thank you very much. Thank you, it was really fun. Come on, bye for now, bye, see you. Hey friends, thanks for listening. If you found this podcast valuable, you can subscribe and follow us, of course, for more episodes on Spotify, Apple Podcasts or any other app. Of course, if you didn't find us on some app, I'd be happy if you write to us. We'd really appreciate five stars on every platform and for you to follow us, so more listeners can discover us and find the podcast. You can also find the previous episodes on any app or on the YouTube channel, we have links in the description. So until next time, come on, be effective, and bye bye.