What is wrong with developers and why Simple Analytics is my first successful project
Adriaan van Rossum
The first few weeks of Simple Analytics were super exciting/scary/hectic! I launched less than a month ago so I will write about the revenue (~$300 MRR) in the next post (when I have one month worth of data) but I would love to explain why this is my first successful project. It's the first project that I ever properly launched. I've built many websites, but for Simple Analytics I tried a different approach. Usually I build the website, show my friends and try to get some organic growth through search engines. This kind of works, but does not give a great start. So this time I did things differently.
We all have read "The Lean Startup" by Eric Ries, and he talks about some great ideas. Iteratively building products and making minimal viable product, to test with your early adopters. One thing I did different, was how I wrote the code. Normally, I would grab a fancy framework that would be perfect for the user experience, I would pick a great database that could handle all the data if I would have millions of records. While working on it I would find a cool tool that is perfect to make it even better. Days later I'm setting up the frontend, database and the cool tool and I'm not satisfied with the database setup. So I need to change it and make sure all the stuff I made around it will work with the new database. My motivation would be gone by then and I wouldn't want to finish the project. Meanwhile, something new comes along. But this time, instead of building the technical masterpiece, I decided to build a working version with the tools and database I knew well so I would have the website done before my motivation is gone. This time it worked!
For the backend I used Node.js without any framework (no need to figure out black boxes), a database that I already was very familiar with: PostgreSQL (I'm so happy that I just write raw queries without any ORM), and deploying to a VPS which I already did many times. With the mindset of building something fast (and for sure not perfect) gave me so much energy. There is nothing really fancy about this project and that is why I had motivation and time to focus on marketing.
A friend who is shoulders deep into Facebook marketing suggested to make a video which could have a viral effect. Something that non-developer people would get and it should be funny. I happen to know the people from TÊTU (a film studio) who where very enthusiastic about the idea. After one day of brainstorming we got a script. The next day was film day and we had to finish it all in one day. We started in the morning at 9AM and finished by 3AM at night. A few days later the edited version was on the website of Simple Analytics.
The design of the platform got a lot of positive comments. All credits go to Michelsen Design who designed the whole website in less than 2 days. And yes, some design thingies are not implemented correctly yet, but this proves my point. It's not important to put those hours in little things of your platform at this early stage.
Launching the product. Like I said before, I never properly launched a project. I also didn't know it was very important until I became a member of the WIP.chat-group. It's a community of so-called makers where you can ask questions, have a public todo-list, participate in discussions and get motivated by other peoples successes. This group made me very aware of launching and how to do it. So I made a list of platforms where I wanted to launch Simple Analytics: Product Hunt, Hacker News and Reddit.
A few weeks ago on September 19th, 2018 I launched on Product Hunt and really hoped to get in the top 3. It was a close call, but I got into the top 3 and stayed there for the rest of the day. Got some feedback but not as much as the day after when I submitted my project to Hacker News (I will devote another post to my launch on HN later). It stayed on the homepage for 1 day and was on the first spot for 9 hours. Now I validated my idea, and it's time to improve the product and serve my early adopters.
I'm not saying everybody should follow this approach, I'm a solo developer which is a different case than with a team and I have a lot of experience building software, so maybe a framework would be faster for others. My point is, build your product with the tools you know instead of fancy other things that could do the job better. And of course I got some comments on Hacker News saying I should use better tools, but are those validating my startup? Don't think so. Build your idea fast, validate your idea and make it better or focus on something else.