In short
Podcast Summary: The Twenty Minute VC (20VC) - Episode with Matthieu Rouif, Co-Founder & CEO @ PhotoRoom
Episode Overview Title: Scaling to $50M ARR in 3 Years, Scaling to $20M ARR with Just $2M Invested; The Story of PhotoRoom, Is This YC's Most Capital Efficient Company Host: Harry Stebbings Guest: Matthieu Rouif Release Date: [Date not specified in transcript] Podcast Link: [The Twenty Minute VC](https://www.20vc.com)
This episode features Matthieu Rouif, the co-founder and CEO of PhotoRoom, a startup recognized as one of Y Combinator's fastest-growing companies. The discussion revolves around PhotoRoom's impressive journey to achieving $50 million in Annual Recurring Revenue (ARR) within three years, powered by capital efficiency and innovative business strategies.
Key Discussion Points
- From GoPro to PhotoRoom
- Transition to Entrepreneurship:
- Matthieu shares how his experiences at GoPro and his previous startups influenced his approach to PhotoRoom.
- He emphasizes the importance of focus and lessons learned from past mistakes, particularly the significance of narrowing product scope to enhance effectiveness.
- Key Takeaways:
- Focus on a single product or market until achieving product-market fit.
- Utilize past experiences to make informed decisions.
- Capital Efficiency
- Scaling with Limited Investment:
- PhotoRoom reached $20 million ARR with just $2 million in funding.
- Matthieu discusses how leveraging tools, a small team, and efficient resource allocation contributed to their capital efficiency.
- Critical Insights:
- Efficient use of resources and investing in quality tools can drive significant growth without overspending.
- The importance of understanding user needs and market dynamics to avoid unnecessary expenditures.
- Consumer Subscription Model in Photo Editing
- Customer Acquisition Strategy:
- Discussion covers customer acquisition costs, payback periods, and the impact of churn rates (30-40% annually).
- Matthieu explains how to maintain profitability despite high churn through effective customer engagement and satisfaction.
- Competitive Landscape:
- Comparison with companies like Canva and Adobe, highlighting differentiation in approach and target audience.
- The Future of AI and Product Development
- Innovations in AI:
- Matthieu talks about the evolving landscape of AI and its implications for startups versus incumbents.
- Emphasizes the significance of user-centered AI development and the need to adapt to new technologies.
- Key Considerations:
- Understand the balance between model size and data quality in developing AI solutions.
- The necessity of continuously iterating products based on user feedback and changing technology.
- Company Culture and Team Dynamics
- Building a Strong Culture:
- Matthieu outlines the cultural pillars at PhotoRoom, emphasizing instant learning, collaboration, and craftsmanship.
- He discusses the impact of hiring freelancers as a strategy to maintain flexibility and talent quality.
- Operational Insights:
- The importance of transparency in communication, notably the company's "no DMs" policy to foster open dialogue and feedback.
- Growth Challenges and Metrics
- Navigating Growth Plateaus:
- Matthieu addresses the challenges of sustaining growth over time, discussing the necessity of regular user engagement and feedback to identify and overcome stagnation.
- Defining Success:
- Focus on metrics such as weekly active users and user engagement to gauge the effectiveness of their product and overall success.
Conclusion Matthieu Rouif's journey with PhotoRoom provides valuable insights into capital efficiency, product development, and the importance of a strong company culture. The episode highlights how leveraging past experiences and focusing on user needs can lead to significant growth in the competitive landscape of tech startups.
For more insightful discussions on venture capital and startup growth, visit [The Twenty Minute VC](https://www.20vc.com).
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00every three, four years, like software efficiency makes it possible for a three X X more each team to take to make the same amount of revenue than a bigger team. So I think this will definitely happen like a 10 people team doing a billion dollar in revenue. And we're like pushing for that. We actually measure how much revenue per teammate we have at photo rooms. Actually, I think YC say like they never saw a company going that fast like with this blog capital. 50 million in error in just three years. they hit 20 million in error on just 2 million of invested capital. This is the immense story of Photo Room.
0:33Now with over 100 million downloads and one of YC's fastest growing and capital efficient companies, I'm so thrilled to be joined by Photo Room CEO and co -founder Matya Roof. But before we dive into the show with Matya today, did you know that every 20 VC episode you listen to is recorded with Riverside? Riverside is insanely good. Like I would pay $1 ,000 per month 4 Riverside, it's that good. Why? First off, E's. Your guests do not need an account, one click in their in the recording room with you. It is fantastic, especially for high -profile guests. Second, they record your video and audio track separately, and in the background, so they're not only higher quality, but the guest does not need to record their end, and then send after which is a total nightmare.
1:17But for me, honestly, what I love so much is how much thought they put into the product. Like, when the internet quality is low, they will disable the video for the cool, but for the recording, it works seamlessly, it records perfectly, it's so thoughtfully done and it makes such a difference. Use my coupon code 20VC, that's 2 -0 VC, and get a 15 % discount, it is the tool I could not run my business without, and speak you've game changers for businesses, arising ventures as a holding company that requires tech startups that are facing setbacks, they happen to a soul that they can be turned around and arising ventures is here to help.
1:52Arising ventures is incredible when one of their many success stories is up council, an online marketplace for businesses to find in higher legal help. Up council was facing legal challenges and a growth curve that no longer matched VC expectations. As they were about to shut to the business, arising ventures stepped in to turn things around. Within a week they crafted a winning deal, buying the assets and taking operational control. without transitioning any team members, they got up council profitable and up council was able to live on and serve millions of people a month. The arising ventures team knows what other founders care about because they aren't bankers, they're tech founders themselves and they have a team of operators ready to go should a business need a quick landing.
2:33A rising venture's quickly got up Council back on track and profitable and they could do the same for your start up. Visit arisingvanshers .com, forward slash 20VC today to learn more and connect with the team, and finally secure frame is the leading all -in -one platform for automated security and privacy compliance. Secure frame simplifies and streamlines the process of getting and staying compliant to the most rigorous global privacy and security standards. SecureFriends industry leading compliance automation platform paired with their in -house compliance experts and former auditors helps you get audit ready in weeks, not months, so you can close more deals faster.
3:13SecureFriends uses over 150 integrations, built in security training, vendor and risk management and more to make compliance uncomplicated. SecureFriends makes it fast and easy to achieve and maintain compliance so you can focus on serving your customers, automate your security and privacy compliance with Secure Frame. Schedule a demo stay at SecureFrame .com. You are now arrived at your destination. Matt, I am so excited for this. Well, people don't understand it's like for me. When I have founders of products that I love, it's like 10 times more exciting. So first, thank you so much for coming on the show's day.
3:48Thanks, Harry. We see photo room today. What was the aha moment for you that like, I'm going to dedicate many years of my life to creating his product. Yeah, that's a great question. Well, first I've been working on photo video apps for the past 15 years. Like I started that on campus at Stanford during my master. So it's been a long, long journey. That's why I got to know iPhone development was photography. And they have a moment for photo room is I was at GoPro that has acquired the started by was working at as an end of product. And well, we were doing an iPhone app quick for video editing.
4:20And we were being featured by Apple. And I don't know if you know these things, but they ask you like, can you provide the artwork for tomorrow? Oh, yesterday, oh, very fast. And while we were this small started being acquired with the US, where they were sleeping, or as I knew I was off and tell myself, well, I'm good with Photoshop, I'm good with computers, let's do it, and I opened my computer, and while I ended up spending the afternoon, literally like trying to get what I wanted, I had a clear vision in mind, and it just, it didn't work, like too many options, I didn't manage to do something I wanted, it was frustrating and at the same time, I had like on the other part of the office at the ATM and it kind of clicked for me like, okay we need something easier, programming got easier but editing didn't and it built something, there's an opportunity here.
5:03I have this strong investment, I get in trouble for this from the team, I really don't like investing in first -time founders, I just feel there's so many mistakes that you make when it's your first time, this is your third time there. What are the big fuckups that you made with the first two that you've done differently this time? I think first one is Focus. So really the my second story like we were doing too many things We tried to do two products. We were not focused and one thing that worked amazingly well for our photo Amies we fell your track from the beginning every six months We would focus more so we started with video and photo and they say like okay Let's just do photo and then we started to say okay it's focused on people selling online and every time we focus with their next number of people like signing in and being happy being like very engaged with the product.
5:49When it works you can fill the traction, you fill the pool. So I don't think you should expand or like be wider as long as you don't have product market, which you shouldn't expand anything. And so for instance, we just focused on the iPhone app in the beginning and really create an amazing experience for these users. So we have intense focus number one. What else is really different? I think starting company is like especially for curious people, you learn like it's an exploration of learnings. You get like ship faster every time you ship your learn and well I think after like 15 to 12 to 15 years in a field like you know it by heart like you're a next -barter.
6:22So I know like mobile apps, I know photo apps, it's the third time I wanted to start subscription third company you know. So all these things like you go so much faster in the beginning so can I ask do you think the speed of execution is as important as everyone says? Yeah, definitely. Actually, so we have three culture values at photo room. One of them is instant learning. Speed of execution, like, ship learning to rate is one of them. You have a finite time in life, and a number of things you can learn really depends on how many loops you can get in that. And I really want everyone working with us to think about how can you learn faster from what you do for my two ships.
6:57For us, it's critical. I think all your makeup thunder is the fastest builder I ever met. like, he ships so fast, it's amazing. How do you think about instilling learning as a process? And what I mean by that is I find a lot of people, I move fast, move fast, move fast. And they don't actually spend the time to analyze what went right, what went wrong, and learn from it. How do you think about learning from the living fast? You have to know what you're trying to learn, and that's how you reduce and you condense what you want to ship. Like, what's the B0 of what you want, and what do you want to learn from that?
7:26Before shipping, try to know what you want to learn from, otherwise, Well, you manipulate the result to say like you've learned something. It's better to like start from that but there's some validated. Do you know I spoke to a mutual friend of ours before the show Jacob at revenue can't? Nice. And I said, what should I ask Mathieu? And he said, asking about the McDonald's test. Or the McDonald's test. In the early days and we still do it like in the street, we would go with all the other, and we would go to McDonald and the Italian would go in the line and offer people like ordering things to other food.
7:58like, can we pay for it and you just test photo room? And we would run like, yeah, 10 questions about photography and then we would let them try the unboring and really give us feedback. And every time they wouldn't get to the new feature we were trying to do like, it was a new feedback, we discovered something and it was the best learning experience. It actually comes from the ZenLit team they were doing that in Paris. I did it so much that I got banned from the McDonald from next to office. So it's just like, I couldn't enter it anymore. The security guard would come to me and say, you can't go in because you're bothering your customers.
8:29What questions do you ask when you sit down with them? Two things. First, you want to understand how they approach your vertical. So far, it's great. Everyone does photo and it's a big asset. And so I would ask, are you capturing your photo, which app do you use? What do you use for using them for fun, for selling for other stuff? And give you an insight, people get comfortable. and then while it's really show don't tell, like you show them the app, they download the app, well it's very difficult because you get frustrated because you know what, you want them to tap, but all the value of it is like them telling you how they feel, what do they understand from the app, what they wish was happening now, and it's user testing, like we're understanding what's a process for them when they see that as a first experience.
9:11When do you listen to customers first, as when you go, great, but we actually have a product roadmap that we're gonna ask you against? Yeah, for first, I think you get a lot of insight from what they expect. Give you an idea, we ask people to take a selfie to show them how the app is working. One of the user testing we realize, well, some people, if you ask them to take a selfie, first they think it's create an account, we refuse, so they're going to spend like actually minutes, a lot of time, in the environment finding the right picture for them because they want to look good, maybe it's a social network, maybe something like this.
9:41I would have never guessed that, like if I hadn't seen it, someone doing it. Well, in the early days, it's also about the blockers, like what people understand, what people see, like what they understand from the product itself. So what we do with L .Y .O .T. is really, well, going back to the office, like changing the feature a bit, changing the wording, and getting back to my dongle, and testing it again. You shouldn't wait longer than like one hour to test again. That's how you iterate fast and learn faster. The sum would say like, ads, like background removal, as a service, like cool, but like it's a feature.
10:11How do you advise found, and I'm sure you've been told that, respectfully? Yeah. How do you advise founders who are told that it's a feature not a product. Yeah, I think it happens to a lot of entrepreneurs. It's a great question. What's important is you... I mean, to see the depth in the feature. And I think you become like a product and not a feature. The moment you understand, this feature is giving depth to what you can do. To give you an example, you can say, background, move all while it's like a feature or like you do a wide background. Well, first it's useful for some use cases. But being in the industry for like 10 years before that, I knew it's not only that.
10:44like first as a product you can remove background but you can also blur you can like focus on what's important you can do templates YouTube thumbnail like it's like actually you double the a lot of opportunities for other features you can build on top of that so I think seeing the step that in the product that's coming from it is really important features that are like minimal lovable product that you execute extremely well are the beginning of a lot of successful products I do believe like focusing and making people extremely happy with what you do is more important than being very wide when you approach something new.
11:19I expect many of your investors and you know, we've got many thoughts back in terms of what we should talk about. The first, which we channeled about before this, was capital efficiency. You scaled to 20 million in an era with two million of cash. Yeah. I've never had that before. I've had the opposite. Very often and that's the capital efficient company, by the way, in venture. When you reflect on that journey, 20 million an error, a 2 million cash. How did you do this? Do you think so efficiently? When you look back, what are your lessons from that efficient scaling? Yeah. I mean, I see $1 a mean, like, to tool, and you can deploy it.
11:52But we never came from with the idea of, like, I don't not spend money on, let's, like, we reinvest everything we can in the growth. So it wasn't conscious, let's save money, let's be super lean, let's save money. I think we're like the idea of being lean, and like, but we always pay for the best tools, like, so we're not lean on, like, the best tool for the team. I think we really like to hire like individual contributor and that's part of our culture We just want to move fast and we know like small team move faster I think that's one of the reason why we're very efficient and then there is Thinking like a macro.
12:20I think there is the App Store story which is a big part of it Like so Apple and Google like the app store and mobile they're giving you distribution Yes, it take like a cut of it But basically it means they offer like a worldwide distribution to five billion people and you you don't need like a big marketing sales team to do that. So it's very capital -efficient, like a very small team product team can go a very long way. And then in terms of capital efficiency, like one of the secret of mobile apps is people are willing to pay upfront. You have a payback of one month so you don't need that much capital on the outside.
12:53Payback of one month. What are the cash today? Oh, I won't give the exact number on that but you can see how much people pay or you're only subscription and you just get the money back in one month. How many people pay monthly versus annually? If you want to be efficient and you want to be able to scale, like fast like I guess, I think jumping on the YOLI by default is the best option and it's actually how you create retain users. So, most of people like going on YOLI is a really good way to scale. When you look back on those early days, is there anything that you wish you'd done that you hadn't done?
13:25We hired great people so you need to try. You don't get the right person 100 % of the time but for sure. Do you agree with high F -5? One thing we do is we work a lot with freelancers and that's a great way because they have so many opportunities so they're willing to try and we give them a lot of freedom and autonomy. You should be able to know in a few months if it's the right person or not. How do you know whether the higher a freelancer or a full time? Files, if the person is willing to learn a lot from working one week together, we love this model, people come to the office, it's amazing.
13:54It's actually less risk for them because the culture is such an important part when you're going to put like... I'm not sure you're successful. I'm okay, so they sent you. You hire them in the freelance first. They come for one or two months. See how it is. And then it's like, eh? No, a few weeks, even. Yeah. You know in a couple of days. Yeah, and they ask for, I mean, usually people are super happy to do that. They love to try and do, like, they have so much gym in and they don't want to commit to a job. And sometimes they have already, like, their status. They might be working with other companies at the same time.
14:21So we were very successful working with Philancer. A big part of the team has been freelancing at some point. I've been freelancing, you know? Have you always had good culture? We spent a lot of time, I remember sitting with Aliyah working even for our first hire, like what are going to be the three pillars of our culture. What matters to us? So I think we did the process quite early on. And with the other two pillars, you mentioned the first about the shipping, it's rising. Yeah, the first one is challenged toward progress, you know, kind of feedback. Like you should be able to challenge anything in the company if you think it's not going great, or even if it's not your field.
14:56Second one is instant learning. So really this idea of shipping and learning from that and how really we're all learners. The last part is being an artisan. We're all makers and we really love people that master the craft and the art of making. And the best that they're craft, sharing that part with others, learning from others. Yeah, being an artisan. That's the name of the entity in France. What we look at this skating as we said, 2 million in 20 million of error. at some point around there, I believe one of your investors said it was kind of a plateau in the growth. What happened with the plateau?
15:29Why does that happen? First, yeah, exceptional growth. Actually, I think YC say they never saw a company going that fast like with this low capital. So indeed, it's quite uncommon. Growth happens with its step function. So you kind of unlock new things and you just need to be passionate and talk a lot to your user, or like getting back to talking to them, understanding what are their pain points and exploring that. And these that you can't always be growing and it happens the right, I continuously. Is it challenging with team rural when you have a growth pluser? Well, yeah, I think we're very focused and driven by growth.
16:03People don't improve something they don't look at, like numbers they don't look at. So we talk about growth every week. And so people are happier when there is growth. So you have to be careful with that. But the company is growing amazing. Well, we are profitable. So it just like take a step back understand what are the kind of adjacent needs of your user and go even deeper with that. I mean, it happens all the time. So you don't want to hire people that are here just for growth. You want to hire people that are here for the mission you're solving. You said about kind of the focus on metrics there, and you know, metrics move when you pay attention to them.
16:33What defines it like successful user to you? When you think about the metrics that you focus on, what are they? Yeah, for us, it's quite simple. A user is like successful as a user, exploring x number of photos every week. Someone who is engaged and who is successful, like, the guy who has a number of exports per week, he has defined this as a successful user. Yeah, in terms of metrics, I think it's really always better to talk about people. So it's like a number of people doing that, like doing one export every week. So a weekly active user is an amazing signal for us, and the signal of someone who's active.
17:05But I really love when people think of not numbers, but people who are in the numbers. So I always have to say, just a number of people doing something, not like a number of exports per week. How do you think about photo? Is it a daily use case product? A weekly use case product monthly? It's interesting. We love our sellers and people like doing e -commerce. That's one of the best person and kind of all the early, early, e -mintry. People do list like on a weekly basis. So for us now, it's more of a weekly better than we see with the app. How helpful was it from a product marketing? So I recently did a show with Matt from Cloudflan and it originated because I said focus is most important thing, have a very, very specific ICP because it will allow you to have way tighter product marketing, much better use of customer acquisition channels.
17:50He was like, what a load of crap. It was like, is she go horizontal? You don't know who's going to use your product, you don't know how they're going to use it. And my point is, when you think about the sellers, you can so efficiently product market to them in the early days and to the 8, as you know, exactly that sellers on these three platforms make. Product marketing is much tighter, no? I think you can combine both. So we have 100 million downloads, like obviously not all of them are sellers. But I think it's really important in your product development run map when you focus to say, okay, we're going to like, to focus on this person now.
18:25It's good to be beneficial to others, but to really the way we build it is for this person. In terms of acquisition channels, how did the acquisition channels look in the early days? When you were growing really fast, where were they coming from? The early days, I mean, it's the word of mouth. So the story and it's kind of on DNA of photo omnis, we got product market fit in the early 2020. Gary V mentioned us, we got a few eBay influencers that talked about us and we got viral like a verified growth. And one month later, there was COVID. From one day to another, everyone, well, can't sell offline.
18:57So you all need to go online with your products. And you can't access a photographer because you can't be to prefer other people. So to be honest, like, looking back, I didn't realize back then but we got product market fit and like one month later COVID hit it just like double down on our growth and it was the photo and watermark so if you go on any marketplace you'll see photo and watermark on a lot of listing and that got the viral growth. Do you want people to remove the watermark? Because it's great advertising fee but you also want them to pay and so it's right. I mean in mobile subscription like there's always 10x at least more people that are never going to pay for it and they're always like at least 10x more people that are yeah that are going to be on the free plan.
19:35So the number is like it makes sense. And actually we have a lot of people that do screenshots, they do screenshots of the image in the editor and just like remove the watermark this way. And we actually reserve it. It's big. If they use it a lot, at some point, they'll pay. What's your yearly chart? Well, I won't give you the exact number, but that's the thing I would love to know if I'm all competition. If there's one number, it's not the highly -gray -al -number, do you think? Yeah, I think it's the one that you love to know. Yeah. It's really eye for, like, because we're useful. So I think there is...
20:01Because if I'm just totally honest, I'm generally more of an enterprise investor. Yeah, and when I think it consumes nutrition, I've had this out with Nico and Jason. And I do, when I'm churning 30, 35%, I'm not saying this is you, but just consume nutrition. 25 to 40 % in the customers every year. How do we make that business work? Yeah, the dynamic is, I think you need to live with the idea of, mobile software is tenix bigger than anything you'll do on desktop anyway. We are addressing 5 billion people. My kids, the first app they use is photography, they use photo room. It's like one of the first software you're using in your life is photo editing and probably the last one you'll use to.
20:38So when you're addressing 5 billion people, it's much bigger than anyone else for sure in your SaaS business. But it's really the top of funerous use and you shouldn't embrace that for mobile subscribers. I'm just going for this because I am too intrigued. I see everyone's kind of coming into each other's space. Something you've got, Kamma moving into other people's place, Adobe moving into other people's place, you moving up. And I'm looking at it going, I don't know how this plays out, because everyone's trying to just eat all of the pie. How do you think it plays out when you look at your photo, your camera, your Adobe, your Mojo, what happens?
21:10What we really try for is making studio quality photo accessible to any professional or any entrepreneur in the world. And I think for that we focus on a different use case. Then Canva, which is more for design on your desktop, or Adobe, which is for the experts. Adobe is for the tens of millions of people that are spending the day in front of their computer and their experts. We are providing the service for people that come up for that. So we're making it more accessible, 10x more people because they're on their phone, and it's wider origins. But I actually don't think it's the same product that you can develop.
21:45You need to choose the one. and we'll all make like a more than a billion in revenue. I mean, the adobe is already doing it and can work close to that. Can we do it while ever? Yeah, so can we do it while we're on the same track and we'll be doing one billion for different social agents? Do you need to move into enterprise? Well, we already do need to move into desktop. Our customers are like small businesses or B2B. And the price we already, like that was our, we started with the photo -home API. So we have like iOS, Android, Web, API. We can render the same thing, which is quite unique. And we were very successful with the API.
22:18There were, you sense a strong pull. And so last May, you get the Barbie movie. And they love it a lot. But they're more like Robby. And they did like a big part of the marketing was using photo room to create like a Barbie poster for anyone in their origin. So they use our tech and anyone like not a pro, could create pro quality posters. I mean, the Cartesian, Rihanna, they use that tens of millions of users. And that's like the API is taking off with that. And that's one example. So that's what we've been doing for the faster. We're just delivering to more people with the enterprise part and the APA.
Read the full transcript
22:50What was your most when you look at the future of the space when you see everyone concerned? As you say, like the field is moving very fast and AI is making a lot of product that kind of obsolete. So you have to think of like to make sure you're not building like a all product with new tech. I think that's a very important part to give you an example. The first company I made I was trying to make like postcards and print them from my smartphone. Well, looking back, the obvious answer is like, that's not what you want to do. You want with the new smartphone, you want to create Instagram, like you want to share an image of where you are and share it with your friends.
23:22That's like the new product with the new tech. And I think it's very important for photo room, like when the tech is moving that fast to, okay, what's the new way of doing that, not like the own way with this tech, and so you need to destroy the old product all the time. Do you think founders say, are using new tech being AI? To solve old problems? I think what's happening with the eyes, you have like all past products that are building tech to serve their existing customers. So if you take an example, like this is the idea of the co -pilot, right? If you're an incumbent, you're a car maker, you're going to build like the navigation board, you're going to build like some security features.
23:58But it's just on top of what's on the existing. If you're the newcomer, if you're the funner, you're building the Tesla, you're building the autopilot part where the input is like, I want to go to this place, I want it to be like a scenic drive and it gets it done. I do want to dive into the depths of AI. If we just start from the top, one thing that I'm always questioning is say, what's more important, is it model size or is it data size? I think on data, what matters is quality more than size and on the model size, it's a trade -off. So it depends on what you want to do. Like bigger is the eye equality, but it's also a lot slower.
24:35And what FUTUREM does is like being 10x faster than everything on market, because we are focusing on what use case. So you can take smaller model because you have like a specific use case and that's very key. So it's kind of a product trade -off when you talk to your customer. And again, that's one thing we do is we talk a lot to the users and understand what are the trade -off they want to have with the iPod. Can I ask you when you look at what you've done with your own models given the size of your team? Is that worrying for the future of AI in terms of commoditization? I think we'll benefit the most from, first, like the leader in the space on the full stack.
25:08So when you look at chat GPT, they do the product and the API. And so I don't think you need to own the full stack if you want to deliver an IQuality product. Like the tech informs a product and the product informs the tech and you build this like a feedback loop. Do you think the foundational models will be commoditized? Would you be a buyer of OpenAI? For what we're not specializing like techs, that's what we do all the time. We start from like something that's open source, commodity ties to understand the product value. But if you want to build a mode, you have to understand what to train and iterate for your model.
25:37Also, I do think Commodityization brings value to the innovation for like open source, but then the best player will be able to like retrain something on top of it and start to really commoditize it. There's value in having the tech and understanding the training in that side. How do you think benefits more today? Is it income transfers at startups? Yeah, I think both benefits from it. Like it's a reshuffling, but I mean, I told you the exception story, but another part of it is like I had been working on AI. So we were the first to ship like a caramel, before caramel exists. So the iPhone framework, an app with AI to do video editing at GoPro.
26:13And one of the frustration for me is like, it's very difficult when you have an existing audience and an existing user base and existing product to just kill it or destroy it and build something new. So I think in long term, like, we've been with the most is all the entrepreneurs and what I'm like paranoid about is more like, what is this new team of, well, Matt and Elio are going to build like in the 2023 that is going to be something we haven't seen in the space. How do you feel about the co -pilot strategy? I had someone on the show say it's the last dying bracketing components which I quite like.
26:43I'm quite aligned with that. It's quite brutal for them and I'm sure some will survive but I think the problem is where I see that being very true is if you have a nogence today, you have programmers, you have designers, you don't want to alienate them. You want to give them extra value. So you're kind of thinking, what can I build for them that makes their life easier, but you don't want to have something that's kind of a notapilot to replace them. It's very difficult for the incubin to say, okay, we're going to kill the product you've been using, make it totally different and make you useless.
27:14So that's very difficult where I think what's very interesting for the autopilot analogy is you address people that couldn't afford that before. That's what photo -miss doing on your. You have a small business, you can't afford a freelancer to LPU. You don't have the skills, you don't have the time. Yeah, if you take the copilot analogy, you don't have the time to learn to drive. The copilot just tells you where to go, but you still need a driver, right? So what we give them is more like an autopilot where they come up with their restaurant, where it is, and we just serve that to them. We serve the visual, we automate that part.
27:47So what do you think of the biggest misconceptions people have around AI? two things. The first one, it's only the beginning. Like product so not optimized. There's so much thing people are going to build on top of what's happening right now. So just one person don't like it's opening so much thing you can build on top of. Like it's easier to build new ways to build new ways to consume content. So that's like people that kind of judge like this is not quite the I quite enough. Like people spend like a few months on that. For three years old. So it's only the beginning. A more practical thing I think is a second misconception is people don't like to write prompts.
28:22It's kind of a common line interface and for others it doesn't make sense to input text. So people are fascinating about it, people are nice and I'm totally agreety. So what do you think that can a common line interface of the features? Yeah, I think it's like we say for human images worth a thousand words. While it's true for machines to like from one photo you get like more than 10 ,000 words of 1 ,000 words of things you can build on top of. So I think the input will depend on what's the job you want to get done. And we'll have a lot of changes in the way, kind of the atomic part of UX. Like maybe keyboard doesn't make as much sense on a smartphone in six months or stuff like.
29:01Cheating EUI becomes less or more important in a world where AI is first. Again, I guess on the show recently, that EUI total BS in a world of AI first. You'll speak to your phone more, you'll have less of a physical relationship with your phone because you're not inputting things. UI will change dramatically. The basic input will be different. I don't think the keyboard, for instance, if you can dictate, is such a pain on the smartphone to pull out the screen with a keyboard. This should go away. Voice is going to be more affordable. A lot of instructor data from today, images, voice, are going to be much more understandable by people and by software.
29:38Sorry, and I think it does change like mobile change a lot like with touch screen Well, I do think AI is going to change a lot of that Does a business model of the future change in AI always still have per seat? We moved a lot of people say well, we've to a consumption based model Yeah, I think the more in the business model you can align for you if your customer is the better going back to first principle you want to have a System of alignment of interest and the best means that model like success that can ask you mentioned text versus video there and and also images as well. When we think about cost of query, for your business model, it's much more expensive, I imagine.
30:14The cost per query given images versus a text. How does that change when comparing text versus image? So indeed, like GPUs are a bit more expensive. It's more as a genie. I think it's a margin, I mean, the cost of GPUs, but I think it will go down. But one thing that is coming is we are making like, again, is day one or month one of genie. So someone, we came up with an improvement, an open source image generation is too exfaster. So I think it's just like a question of time and a few months the cost of generation for text and image will go down a lot. Do you think VC is overexcited about generative AI?
30:50Is the hype reaching hype bubble? Well, I think there is a reaction, it's kind of amplified too much. I do think it's one of the biggest opportunity of the decade at least for entrepreneurs. So starting from that, but maybe they'll override that part even more than it's it. You just... Do you think 99 % of VC dollars today invested in generative air will go to zero? And maybe they need to for development to happen. Yeah. But still go to zero. Well, what's the success rate for a startup is like 5 % right? So you're just saying... I didn't know about that. I didn't know the actual success rate on getting a 1x back.
31:24But like a lot of startup, like an insight, what is success? Yeah. Yeah, billion dollar companies, I'm half a percent, a quarter percent, whatever it is. But actually companies that lose all your money, quite rare. Yeah, all right. Do you see what I mean? Yeah, yeah. People are very ambitious. A lot of companies are doing research -based AI. I think that's quite unique about the photo -room is we're really doing like user -centric AI. And when you do research, I think the failure is higher than like when you're building a product. So what's he's a centric AI? We use a feedback from the users to tune, learn with the model.
31:57And what we do is we learn from how they use the product. I told you about not writing problems, but we really developed the app for the user not to have to write pumps in the app We realized on the mobile phone like people don't want to wait so we make image generation like less than one second And that's like a killer feature because no one does it and so that's kind of you make architecture choice of your model There are based on the user feedback and learning them like fail in McDonald using your app kind of We speak a little bit about your kids beforehand your children will be generative AI first generation children What does that mean for how they think about the world, culture, identity?
32:33How do you think about that? For product development, it's always been super interesting to see. Like a good app is an app that a five -year -old can use it. We actually use that internally. We have this Myoney model of ideal customer. And one part of that is a five -year -old should be able to use photo -room in terms of accessibility. And you see them, it's amazing. I'll quickly understand the UX. Many people said to me, we're going to have multi -billion dollar companies with 10 people. Reims see differences in terms of capital efficiency to build massive business. Do you agree with that? Or we just find new jobs for them to do?
33:04I agree, on the trend. Every three, four years, like software efficiency makes it possible for a three X X more regime to make the same amount of revenue than a bigger team. And that's actually a big opportunity for entrepreneurs and investment. So I think this will definitely happen like a 10 people team doing a billion dollar in revenue and we're like pushing for that on impact like we actually measure our Much revenue per teammate we have at first rooms. Harding measure is the impact of a teammate and that's the good way of Looking at it. You don't want to try or someone you need because of that obviously and we are like we double the team in like the past six months So we are like growing very fast The espyre to Jan Lecun before the show and he said you just got to ask him mate.
33:44Why is so much happening in Paris? Well, I think Jan is a good responsible for that. Like, there are the best PhD talents that are in the big tech labs and are still in Paris. There is a very good undergrad system for engineering. We do a lot of math, which is good for AI. Plus, the labs creates a very good environment. People did stay after that. And then, amazing companies, amazing startups that are now building from Paris. You have the pool side team that move from California to Paris. You have the dust team, the mist wall team. So I think the ecosystem on AI reached escape velocity and it's been in the building for the past 10 years And now it's like it's the best AI ecosystem in Europe.
34:22So that's and it's escape velocity So people just are attracted to it now final one and then we will do a quick fire But we sort of like kids earlier if you could call yourself up the night before your first you have three now and say Yeah, Matt you shouldn't be this what would you tell yourself? I think it's very difficult to know what it is Before I hand it can't make you a better leader You have to think how you make them grow, you have to be an example for them. And I believe strongly that the best way to lead is like showing a strong example of the way you should do it, like being very hardworking, transparent about you doing, like being an example and I think that's the best way to lead.
34:58And do you believe in balance? Not much, I think I like to do things 100%. But I do have the kids that I see them, but there's some balance to find in them. Is that tough? Yeah, it's tough. And that part. Do you see the kids much? Yeah, I spent another time with them. Your wife's gonna be listening to this, like, No, I'm shouting more, I'm shouting, yeah, I know, I did have that, sure. With more time. Yeah, exactly. Okay, we're gonna do a quick fight. So I say sure, they'll be me or me, they'll be at thoughts. That's not okay. Yeah. Open AI, long or short? I mean, I'm longer on the, and the Jimmy's Sharp, they have a, like, a brand now, and yeah, Sam is super sharp, so I'm longer on them.
35:34What's the biggest misconception, like, the single biggest on AI? I think it's just people judge the result, but it's moving so fast. It's like people don't understand how five people team came up with an algorithm or like a 10 people team. So like now there's optimization, there's product building. So it's just the beginning of building super useful, exceptional product on top of it. You can have dinner with anyone in the world that are alive. Who do you have dinner with? I'm going to go with Steve Jobs. He's making me a Kishie while Disney maybe. Which brand do you most respect? Am I? I was gonna say night, it enables people, like this idea of pushing people to do more, like just do it entrepreneurs and I love that part.
36:14I'm at that. What's the strongest belief you had which turned out to be wrong? I had a very strong belief that you could only innovate and be creative in person. And while we started the photo movement earlier and COVID hit and we just like, okay, it's time to build a remote company and be hybrid and all with thriving. We are very international. So it really changed my mind on that. So now you're hybrid. Yeah, we meet once a month in Paris. I grew up in a beautiful HQ, but the team is all over Europe and that brings like so much diversity to the team and to the product we build that. And it's such a pleasure to work with like different nationalities from all over the world.
36:51Are you in the office every day? Yeah, I am. Yeah. But the team let me to be. No, like a 30 % or a third of the team is in the office there. Yeah, every day. And then once a month they come and... Yeah, it's the amazing, it was the amazing time. Like we did like a Barbie, it's him day about that. Like we, yeah, we were very creative. I think creativity is a muscle. So we just like find new activities all the time. And this is kind of the arm of the team. What's better and what's worse about being a hybrid? Better is, it's more diverse. I like kind of the glass ceiling of, at GoPro of being like 30 French dudes.
37:22And you don't account for all the needs of everyone. And now we have a more inclusive diverse way of developing product and you learn so much from people from different fields and different countries. The worst part is there are a few fine details on building crafting products that are not always as easy to get. I think tools will come for that, but that's would be the worst part. I always say things like, share a voice is better when it's done remotely. People don't feel as intimidated to speak up in a lot of cases. That's right. I think if you do, especially we have a very transparent culture, we have for no DMs culture.
37:58So like, you don't write DMs, I agree. So no, but it's like, is it open space? And we just do thread like people whispering, you know, that we try to be organized on that. But I think it brings a lot of people can speak up and a very like eye transparency culture on that. So I have to ask about that, no DMs. So if you are with Elliot and you're like, how are you, Harry's not performing to your group with me. That would be on a DM, no? Yeah, no DM as a policy except like on like salaries or stuff. or performance management letting people go. If there is something you want to challenge someone towards, you should be public, especially, so we try as much as possible to do it.
38:33And it's a good way to start from. The default should be D .M. I don't know, you don't want to take some kind of screen estate in the channel and don't say something you think people don't care, but brings this transparency, this openness to just picking the open. How impactful was YC? It's been one of the best decisions for the company. I think it puts us under 1 billion, and a trajectory when billion revenue hit. Why? Just like, what did they do? I'm just intrigued. As a product builder, you come with a lot of insights, a lot of things you want to build. And you're not like, I made a lot of progress.
39:04And for instance, storytelling and really being efficient at scaling. And I mean, that's kind of biased. Like, you don't do what I see if you don't want to scale and do like $1 billion in revenue. So I think it's like a nurturing bias phenomenon. And we still talk to them. Like, we still talk to our partners all the time. And they are like really helpful. Where we have good stuff and the Eric. And then final one, if we think about Federer in 10 years time, which is 2033, where are we then then? Yeah, I think, well, we are in the dictionary. Your photo and something. You know, our vision is, I talked about Nike, but really what we want to give is every entrepreneur, every professional, the poor to create Nike -level visuals.
39:41This kind of visual agency that's powered by AI that does the job and automate that iterates. And design is a reaction from, so it's evolving fast. It really increases the quality of the content that's errated, segmented for all the... If you're a small entrepreneur, the trend is like there are more and more small businesses, very small businesses selling on Shopify and Amazon and Marketplaces. So we're going towards a more diverse world in terms of e -commerce. We're just helping people to be... to thrive in that, to be proud of what they do, like good visual. I mean, good visual for your show, or like your product yourself, and you have the show.
40:16like they call TikTok content is amazing. So we want them to be proud of what they do. Like they start their chef. You start a restaurant. It's great to be proud of what you do and to be unique, like generating something unique for them. This is such a pleasure to you, Matt. Thank you so much for joining me. Thanks, Harry. Great questions. I absolutely love that. It's such a natural discussion with Matt there. And I always so much prefer to put the schedule aside and really just have a normal conversation. If you want to see more from us behind the scenes, of course you can on YouTube by searching for 20VC.
40:43That's 20VC. but before we leave you today, did you know that every 20 VC episode you listen to is recorded with Riverside? Riverside is insanely good, like I would pay $1 ,000 per month for Riverside, it's that good. Why? Well first off, E's. Your guests do not need an account, one click in that in the recording room with you. It is fantastic, especially for high profile guests. Second, they record your video and audio chat separately and in the background, so they're not only higher quality, but the guest does not need to record their end and then send after which is a total nightmare. But for me honestly what I love so much is how much thought they put into the product.
41:22Like when the internet quality is low, they will disable the video for the cool, but for the recording it works seamlessly, it records perfectly, it's so thoughtfully done and it makes such a difference. Use my coupon code 20VC, that's 2 -0VC and get a 15 % discount, it is a tool I could not run my business without, and speekive game changes for businesses arising ventures as a holding company that requires tech startups that are facing setbacks. They happen to a tool that can be turned around, and arising ventures is here to help. A rising ventures is incredible when one of their many success stories is upcounsel, an online marketplace for businesses to find in higher legal help.
42:01Upcounsel was facing legal challenges and a growth curve that no longer matched VC expectations. As they were about to shut to the business, arising ventures stepped in to turn things around. Within a week, they crafted a winning deal, buying the assets and taking operational control. Without transitioning any team members, they got up council profitable and up council was able to live on and serve millions of people a month. The arising ventures team knows what other founders care about, because they aren't bankers, they're tech founders themselves, and they have a team of operators ready to go should a business need a quick landing.
42:34A rising venture's quickly got up council back on track and profitable and they could do the same for your start up. Visit arisingvanshers .com forward slash 20VC today to learn more and connect with the team and finally secure from is the leading all in one platform for automated security and privacy compliance. Secure from simplifies and streamlines the process of getting and staying compliant to the most rigorous global privacy and security standards. SecureFriends industry -leading compliance automation platform paired with their in -house compliance experts and former auditors helps you get audit ready in weeks, not months, so you can close more deals faster.
43:14SecureFriends uses over 150 integrations, built -in security training, vendor and risk management and more to make compliance uncomplicated. it. Secure frame makes it fast and easy to achieve and maintain compliance so you can focus on serving your customers, automate your security and privacy compliance with Secure Frame. Schedule a demo stay at SecureFrame .com. This nice sir appreciate all your support, stay tuned for an incredible 20 growth episode coming this Wednesday and the one and only Andre Igodola coming on Friday.
From the publisher
Matthieu Rouif is the Co-Founder and CEO @ PhotoRoom, one of the fastest-growing YC companies having scaled to an astonishing $50M in ARR in just 3 years. Their capital efficiency is immense having scaled to $20M in ARR on just $2M of invested capital. Prior to founding PhotoRoom, Matthieu founded several start-ups, including an app for ski resorts, HeyCrowd, and Replay, a video editor which was ultimately acquired by GoPro. Whilst at GoPro, Mattheiu led all image editing products.
In Today's Episode with Matthieu Rouif We Discuss:
- From GoPro to One of YC's Fastest Growing Companies:
- How did Matthieu make the move from GoPro to founding PhotoRoom?
- What are the big mistakes Matthieu made on prior companies that he did differently with PhotoRoom?
- What does Matthieu know now that he wishes he had known when he started PhotoRoom?
2. Scaling to $20M in ARR with $2M of Cash:
- What allowed Matthieu and PhotoRoom to be so capital-efficient in their scaling?
- What are the biggest mistakes founders make when it comes to resource allocation and capital efficiency?
- On reflection, what did Matthieu not spend money on that he wishes they had spent money on?
3. Consumer Subscription + Photo Editing: Is it a Good Business:
- What are the customer acquisition costs by channel for PhotoRoom?
- What are their payback periods on a per-customer basis?
- How can it be a good business when the churn rate annually is 30-40%?
- How does this space play out with Canva, Adobe, Veed, Kapwing? Who wins?
4. The Future of AI:
- Who wins; incumbents or startups?
- What matters more; data size or model size?
- Will UI be more or less important in an AI-first world?
- Why does Matthieu believe that everyone hates command line prompts?
- Will we see $BN revenue companies created with just 10 people?




