How I built a $1M/year company (5 lessons I learned along the way)

It all started with an idea. An idea to change the world. It starts that way quite often for many entrepreneurs. I was one of them. Three years ago, me and my cofounder, Ramon, sat down for some tacos and beer a Friday evening and the idea of Crema was born. Help freelancers get paid faster, internationally. As former freelancers, we knew the struggles, the pain points, the market, and the opportunity ahead. It was simply too complex and too expensive to get paid when you work with clients from around the world. So, we decided to solve that. The lesson here is more inline to deeply understand the market you are going after when you are thinking of starting something. We of course knew the market very well. We knew the pains, we knew the competitors, but most importantly, we knew our few initial customers. And I think this is key if you are going to start something on your own. You need to know those three key things before you even start. If you do, then you can move on to how you can actually start building out an initial version of your product to solve the problem you are going after.

After the idea, comes the building of it. This is a crucial step, because it can make or break your company. Because if you scope out a too large version of your product, it can take too long to launch it, and you will have less time to iterate and see if you are actually solving the problem. Build it too small, then you won’t give enough value to your customers for them to actually pay for it, which in turn forces you to build more, take more time, more money, to actually get useful feedback. It’s a fine line. In recent years the concept of the MVP has gone way out of control and has lost its meaning. The way I think about it is, build out enough value for you to get paid customers, but not absolutely everything they need. Then you can iterate on your product and build it out little by little until you are solving everything the customer needs. Initially you just need enough to get paid and get usage. The way we did it with Crema was to build out sufficient technology to get our freelancers paid faster, at a cheaper rate, but do most of the work behind the scenes. No need to automize operations at first. That comes later. It started out with a simple link to get you paid from clients wherever they were, and be reasonably priced compared to the competition, all while making sure it was faster. Later on we realized that the faster part was not that important for our users. But making it super easy for their clients (to optimize for them to actually pay) was much more important. But, this was an insight that came with launching the product fast and getting feedback. I generally encourage this approach to anyone that asks me.

Product is ready, now I need people to use it. I mentioned earlier that a lot of companies are made (or die) during the initial product creation fase. And that is true in the sense that you can either spend little money, or a lot of money (or time) on the initial product. I do however, think that getting the product in the hands of users is equally important. Because if no one uses your product, then it doesn’t matter either way how good (or bad) your product is. Product distribution is super important and not enough product makers put sufficient importance to it. It is generally where you see a company either stay in the $1k-$10k/month range or get to the $100k + range depending on how good they are at distributing their product. If you know the problem you are solving, and you are thinking about distribution in the ideation of the product, this part becomes easier. It is never easy, but it can become easier if you think of it before building the product. Now, when I think of a business idea, it carries equal importance as the idea itself. I always think about how can I get my first 100 users with this and how would I do it. If I don’t have an answer, I simply don’t do it. At Crema, we knew a lot of freelancers and we knew that we could count on them to at least try the product and receive feedback. We knew that we could quite easily get 100 people to try our product and that was really important. This is lesson number 2: know how to get your first users before you start building. After that we started worrying about how to get to the next 1000 users.

The third lesson is optional. But part of the lessons I carry with me. It is related to capital. It is not always needed. It depends on the idea. In our case, we could have built out Crema to be bootstrapped, but we decided early on that we wanted external capital to scale faster. We also knew that the idea had potential to be a venture backed company and that with additional capital we could grow it faster. In hindsight, I think it was the correct approach, although I would have liked to see how it would have played out if we didn’t take anyones money. The price for that is that it comes with additional pressure. The pressure of returning that capital to the ones that believed in you. Sometimes those people are your family or friends (if you’re lucky enough to count on that), sometimes it means capital from more sophisticated investors (venture capital). Either way, it comes with the pressure of making sure this actually works out. Because 9 times out of 10, it doesn’t. And those are the bleak odds your are facing when you are playing this game. It is what you sign up for. My hope, is that you learn something from these lessons and maybe avoid some of the mistakes I made along the way. My lesson here is: only take money if you are shooting for the stars. We did. But, for my next onces I would be very careful about doing it again. It probably has shortened a couple of years of my life expectancy due to stress. If you can, try to do it without external capital. The good thing is that it is so much more available now. With AI you can practically create an tech company without hiring anyone.

Which perfectly segways into my last lesson. Team. This is the part where I think I struggled the most. In my honest opinion we made several mistakes here. We hired too fast and fired too slow. We didn’t hire enough A players and that hurt us a lot. And onces we realized they weren’t A players we didn’t fire them fast enough and it cost us money and shipping speed. They aptly say, A players attract A players, but B players attract C players, and I think that is very true. I also think that we hired too many people. We got caught up in the ZIRP era of seeing headcount as a metric. Thinking many employees = cool. This is a terrible way of thinking of it. I now think of it as, more people = slow. Now we only hire when we absolutely have to and we have stressed the entire team into falling apart. And when that happens, we take a long time to hire the right person. For example, we have been looking for a senior back-end engineer for 6 months now, and we still haven’t found the right candidate. And I don’t mind. I only want to hire the best now, and that takes time, and most of the time, convincing. The right person will raise the standards of the entire company, the wrong person will lower them. Spend as much time as you have to to make sure that it is the former option you land on.

These were my 5 lessons for building Crema to a $1M/year business. I have many others, but in generally I can always boil it down to: Picking the right idea, in your zone of genius, where you know the problem, and how to get your first 100 customers, then choosing what to build and ensuring you are providing value to your users so that they pay you, and not take too long to get there, and once you are there focus on distribution, iteration, finding capital (if you need it) and hire only A players, and if you can’t find them, don’t hire.

Cheers,