Paul is the CEO of Kamua, a cloud-based SaaS that lets customers use AI to create and edit videos really quickly. The business idea came from a problem he was facing on a Netflix-for-x startup he had created before. After validating the idea through an MVP, the SaaS started growing fastly and nowadays makes +$6k in MRR.
You MUST validate your startup ideas if you want to avoid failure. In our course "Pre-Sell to Validate" we teach you an actionable framework to do it. You can get it here.
Hi Nico! My background is in sales, marketing and business transformation. I’m 44 and this is my second startup, my first was a consulting and software sales company. I have an MBA from Columbia (NY) and London Business School and I am based between London and Bucharest, Romania.
Kamua is an automation centric video repurposing studio, based in the cloud. So you just need a browser and wifi and you can use our AI to create videos really fast that would otherwise take you a long time and require you to learn advanced software that you would have to run on a powerful computer.
Kamua is best used for repurposing existing video content - for example taking a YouTube dance video and converting it for TikTok using our AI tracking. You have to see it in action to fully understand what it does, but basically if you have a video filmed in landscape mode of someone dancing, our AI tracks the dancer and can crop the video so it works really well in vertical mode like you see on TikTok and Snapchat, IG stories, etc.
I’m the CEO and one of the two founders, along with an amazing engineer, Radu Amarie, from Bucharest, Romania. My job is to make sure that I have the most talented engineers and product people I can find, make sure they are happy, and that the product we created fits the needs of our target customers. Of course there are things like fundraising, investor relations, financial planning, intellectual property and all that other stuff, but they are either impossible or unnecessary if you don’t have a product that people need and love.
Our business model is Software-as-a-Service so people pay us monthly or annual subscriptions to use Kamua online.
The idea for Kamua came from a need to provide people with the ability to auto convert their widescreen YouTube style videos to vertical mobile formats for apps.
First, we built a Netflix-for-short-film at Findie. Then, using AI we built, we automated the creation of short 10-15s trailers and we made hundreds of thousands of those for clients like L’Oréal, Nissan and Red Bull.
When it became obvious that these videos would also need to be converted into vertical versions, we built additional AI that identifies the focal point in each segment of a video and automatically crops the video to fit the desired size, like vertical or square.
Then, we created Kamua as a brand that would help us distinguish from Findie.
In the Hawaiian language, Kamua means “the first.” In Swahili it means to squeeze, compress or crush, which is kind of cool because we have tech that does all that. We just stumbled on the name and it worked out quite well. Our mascot is a chameleon (we call it “The Kamuleon” lol) which we really loved because it’s an animal that adapts to survive and we help people adapt their videos so they look like they were always intended for whatever screen they are on.
I came up with Findie, our original project, while I was doing my MBA. At the time it was obvious to me that there was no way of discovering the best curated short films out there. People were watching more and more videos online and on mobile and it just seemed logical to me that there was going to be a demand for great stories told by people outside of Hollywood who just didn’t have access to that kind of distribution. We got several thousand users but we quickly pivoted to a revenue generating B2B model.
So that particular project is kind of a success but also in a way it failed to meet its original goals. The simple reason is that you have to focus on one thing at a time, and as soon as we launched Findie, we got a bunch of requests from really great brands who wanted to have their own branded version of the site with their own content. So we quickly got pulled into building a white label version and we made a fair amount of money with that, and it led us to build Kamua because of the need to automate video thumbnails. Maybe one day Findie will blow up if some celebrity loves it or we spin it out.
I know you like to highlight failure and the role that it plays on shaping future decision making! What we did was fail in lots of small ways. We failed to understand how to build a user experience that instantly hooked people. Some people were like, well this isn’t the best video I’ve ever seen so I’m not coming back. Others wanted to watch series and not just one-off short films, so we failed at matching up brands with content with UX with user acquisition with monetization with retention. Way, way too many things to get right. You don’t start learning to juggle by juggling five samurai swords standing on a log in the middle of an alligator swamp. It would be a lot better to start with two tennis balls in the middle of your garden.
What we learned from this is: singular focus. We focus 100% on Kamua and making it the best and fastest video repurposing tool on the planet. “Will it make our customers happier and keep them coming back?” is the question we ask when we are faced with a decision. I learned this from Steve Redgrave and his gold-medal winning Olympic rowing team. “Will it make the boat go faster?” That’s how they did it.
Me, personally, I didn’t build the actual product. I elevated my CTO from Findie (Radu, one of the top 5 smartest people I have ever met) to co-founder and he built it with our core team of engineers. Sure, I was involved in most of the decisions, but my job was to set the vision and go raise the money we needed to get our MVP to market.
We started out by looking at the fully-automated tech that we had built to solve the problem of having to edit tens of thousands of video trailers. We asked ourselves a fundamental question: what happens if the customer wants to fully automate the conversion from horizontal landscape to vertical portrait and the result is anywhere from 50% accuracy to 99% accuracy? What tool do you give that customer to either fix the 50% they didn’t like (or was totally wrong) or just that 1% tweak?
So, imagine you’re the customer. You are being offered a self-driving boat ride to the island you need to get to, but the self-driving boat won’t drop you off at the pier. Maybe 5 meters away, maybe 10m or 50m or 1km. In any case you’ll have to swim a bit. Now, if you love swimming, and that’s how you were originally going to get to the island, no problem, you like manual stuff. If you don’t know how to swim, and you thought the automated boat was going to get you all the way there, major problem.
That’s why we invested so much time and effort into building an easy to use interface that works really fast and intuitively. It takes 20 minutes to learn how to use Kamua. You can spend 50 to 200 hours in desktop software learning the skills required to do what our AI automation does for you. And even if you do learn those skills, our AI is way faster than the fastest most skilled human, and most people who work in editing will tell you that it wasn’t their life’s dream to crop videos for a living. So we are careful not to destroy the love and the art of video editing. Drum machines and MIDI sequencers didn’t destroy the art of drumming, they just made it possible for millions of people to create hip hop and dance tracks without a full on drum kit. Some of the best songs in history were created with minimal musicianship. This is what technology can do: unlock creativity by removing barriers to execution.
So back to the nuts and bolts of the product building. We built everything to be used in a web browser (Chrome, Firefox & Safari) and to work in the cloud. We didn’t want to force customers to have to download and compute stuff locally. Our first customer had the exact problem we wanted to solve: they had really old and slow laptops for their social media managers, but these were the people under pressure to get videos out in a timely manner. That customer is part of the Omicom group, so they are a great first customer for us.
We then got in touch with Google and let them know what we were building. And even though they were working on something called AutoFlip which they eventually released open source, that crops videos, and we have a competing technology, they invited us into their Google Cloud for Startups program and provided us with $100,000 in cloud credits so we could deploy our service anywhere in the world and have fast processing for our customers.
After that, we did the same with Nvidia in their Inception program and with HubSpot in their startups program, so we are building partnerships with these amazing companies and getting a tailwind when we need it the most.
So now, our service is fully deployed on Google Cloud and anyone in the world can sign up and request access to it.
We’re just starting to grow this particular “prosumer” business so our marketing strategy is very much a WIP. To begin with, we are prioritizing paying customers, so people have the opportunity to become Supporters and pay to use Kamua right now. That’s a bit of exclusivity - kind of like Superhuman does - and it’s really because we want customers who are willing to invest the time and effort to learn how to use Kamua properly. Steve Jobs famously said, paraphrasing here, that the computer is a bicycle for the mind. We think Kamua is a Lime Scooter for creativity. When things get steep, you just keep pressing the lever and up you go.
We had a period of open Alpha access last year but what happened was we got a lot of people in who were purely curious. They didn’t need to edit video or work faster, they didn’t invest the time to learn how it works, so they ended up just distracting us from our mission. We don’t offer free stock footage or free audio files or free templates. If you don’t have existing video to repurpose, you’re not going to really get the most out of Kamua. So right now we’re marketing only to people who need to repurpose video for a living and we want to save them time. One of our customers went from managing a 16 hour end to end process to doing it himself in Kamua in 15 minutes. It sounds like a made-up story but it’s real. Between exchanging Excel files and screenshots and all of the stuff that 2 teams in 2 different companies used to do, the customer now grabs the videos he needs from his client’s YouTube channel, mashes together and crops the video to the social media format he needs, and sends it to them for concept approval. Fifteen minutes end to end.
And he gets them three videos instead of just one. So if we really wanted to boast we’d say “48 hours to 15 minutes” but the reason we don’t say that is because the extra two videos the customer produces are the added value to his client, and the time savings are the added value for him and his colleagues.
That’s a force multiplier. They can A/B test now, whereas before with the time and budget constraints they could not.
Another great use case from a different person at that same customer that she just made us aware of this morning. Because of Covid-19, she is working from her parents’ place in the mountains where there is only a 3G connection. She works exclusively in Kamua and only uses files that are already available in the cloud, no uploading or downloading. Without Kamua, she would not be able to work because nobody else can take care of her parents and in the time of Covid, this is increasingly a common scenario that people are working remote on less than ideal connection speeds.
So that made us believe there are hundreds of thousands of people like that customer all over the world and we are in the process of finding them.
Revenue is just starting out. That particular B2B customer is on an annual pre-paid contract at $700/year per user. We have higher tiers and also entry-level tiers for Supporters, as low as $21/month right now with an annual subscription.
Some of our growth strategies include making tutorials that we post on YouTube and Reddit, that help people with competitor software, like Adobe. We think it’s restrictive to just post tutorials on how to use your own product when your target audience is facing problems using other legacy software, and your current software doesn’t solve all of their problems. They see the Kamua brand, they associate us with solving problems for creative people and some of them will be curious enough to see what we offer. It’s a slow-burn piece of the strategy, but it does work. We are getting hundreds of new subscribers to our various social channels every month this way, and we just started.
We want to be the go-to brand for creative people who don’t want to be constrained by software and hardware.
We definitely want to generate revenue and become profitable so we can invest back into our tech and our customer experience, but we will also balance our growth with a free tier when we are ready to support non-paying customers. If we need venture funding, we will go looking for it in the right place at the right time.
Every day we are on the lookout for talented and motivated people to join our team. There are only six of us total full time in the company, so we have built a pretty monstrous amount of tech and product with such a tiny team. But we’re a bit like special forces in that we operate better with fewer people so we’re not tripping over each other and getting bureaucratic. That doesn’t mean we don’t respect and follow process - the contrary. You should see our Github!
Our next feature is the ability to edit video by editing the transcript. We already have automated captions/subtitles, but what if you want to edit the transcript and have that automatically remove the parts of the video that you deleted from the transcript? This is going to be a game-changer for people who want to edit video but are scared by complex or goofy looking interfaces.
The number one challenge we have faced is trying to separate the signal from the noise. You get a TON of noise and it is very, very distracting.
You get paying customers who ask for new things they don’t need and don’t use. You get investors telling you that you need to do X instead of Y and they don’t really know the business or sector but are applying general things they have read or experienced. You get competitors releasing new features that scare the crap out of you but then you go have a closer look and you realize that they are completely over-hyping their capabilities. Anything you can imagine, it happens.
Money is the second biggest challenge. There is such a thing as being overfunded, but we have no idea what it actually feels like because we have always been lean and efficient. But, for sure if we had $1 million in the bank we would be doing more and better things right now, so we have to overcome the growth obstacle because there are only three main ways to get meaningful cash:
Any of these can also mean you get acquired by someone who sees either your tech or your product or your revenue or your customers as valuable. But that’s an endgame not an obstacle. Nobody has failed by realizing, “damn nobody wants to acquire us.” They failed before that happened.
There is also the personal challenge that I’ve been through. For the past five years I’ve been pretty much unable to invest in improving the quality of my life. I went from having a pretty nice monthly income to less than 10% of that income, and that affects where/when you can go on holiday, what kinds of nice things you can buy, whether you can help people in need, and even what healthcare plan you can afford for yourself and your partner. It’s pretty brutal but it’s necessary so you don’t have this false sense of security. The first $225k into the company came from my own pocket, and I know there are gurus who say don’t do that, but I think it’s helped sharpen my focus further to not have the distractions that go with more personal financial resources. Sounds totally counterintuitive, but it also means you have to sell your vision to investors, you have to find other ways of incentivizing your people, and you have to find perseverance you didn’t know you had. Cliché but nonetheless true.
Our greatest disadvantage is our ambition. We know we are taking on the world’s giants like Adobe, Apple, Vimeo (part of the massive IAC group) and many other well-funded companies and startups. We envisage an ideal world where people from anywhere and any financial capability can express their creativity. Why is that a disadvantage? Because we are constantly biting off more than we can chew and trying to do more and more, introduce newer and more amazing technology, and getting further and further ahead of the tech curve. That tends to put more of a focus on building and less of a focus on growth, and because we rely on web browser technology to deliver our service, we are always pushing the boundaries of what those browsers were designed and built to do with video. It’s one thing if you just want to play back video in a browser; it’s another thing entirely to get a millisecond-specific cut in a video based on the exact timing of a sound. So to counteract that, you have to continuously try to bring innovation to your customers but resist the temptation to build “just one more feature.” We’ve built a massive service with so many moving parts and we did that while trying to just make it easier and faster for people to convert their videos from one size to another. Never underestimate complexity, but also don’t let your ambition drive you to increase that complexity beyond its minimum required level.
We should have released the fully automated trailer/preview/video thumbnail generator two years ago when we had the first working version. Sure, that would have resulted in a lot of frustration for customers who would not have been able to change anything in the result, but some people would have been happy with the fully automated results, and we missed an opportunity to make those people’s lives easier. We decided instead to build the editing system because we looked at cloud computing costs and it would have cost a lot of money to produce a lot of discarded results, and we anticipated the frustration of people not being able to tweak something that was automated and almost perfect - but wasn’t. So me personally, I didn’t want us to be thought of as a brand and company that promised customers they would be able to tweak things “soon” knowing that browser based manual editing is a very complex thing to build. “Fake it ‘til you make it” only works if you are providing something with a lot of promise or you have really gullible investors. A lot of the faking leads to dead companies but we get biased when we see the unicorns who used to fake it.
All that said, we would have had paying, or at least non-paying but happy, customers and kept the frustrated people in a mailing list that would be much bigger than the one we have right now. So that’s definitely something we would have done differently if we could go back in time 24 months: communicate really clearly what the tech can do and people who were willing to sift through a bunch of unwanted results would still have found value when our tech got the results right.
A piece of advice I would give to anyone just starting out: never underestimate the power of consistent persistence. Success compounds. Drill that into your brain.
You have to master the balance between questioning what you are building and having doubts about what you are building. Remember that if you have a solution to a pain point, it might be an aspirin (nice but not essential) or it could be morphine (need it right now). I know a startup in the medical video call space that in February this year was two weeks away from shutting down due to low demand and a month later were taking VC money at 10x their previous valuation. Don’t underestimate the role that luck and timing play in your success, but also be prepared to pivot when you know that you are not even close to achieving product-market-sales-fit. Musical.ly - which became TikTok - was building a distance learning video app and pivoted with their last $10k. Slow growth is OK. Pivoting to an epiphany concept is also OK. Persistence is only truly valuable when you are persisting thoughtfully.
For learning, I want to give a shout-out to Bram Kanstein, a super supportive guy who embodies a lot of the principles of true entrepreneurship. He created Startup Stash which might still be the most-upvoted product ever on Product Hunt. He also has a course called NoCodeMVP and I think that’s something any entrepreneur starting out should subscribe to and benefit from. It’s on AppSumo right now for $100 off.
There are so many distractions for entrepreneurs that your reading list could end up being 1,000 items in just a month. I’m just going to pick my top 10 across all resources.
Podcast: Making Sense with Sam Harris (not tech related but makes you appreciate critical thinking)
Newsletter: Benedict Evans (mobile and tech focus)
We have a blog at Kamua on which we post not only tech stuff but also video tips and soon we’ll be posting business and entrepreneurship articles as well.
We are just getting our direct-to-customer launch up and running so our social media machine hasn’t quite kicked into gear yet, but it will soon.
Your readers are welcome to reach out and connect to me on LinkedIn and they should mention this article in their intro message.