ZlapPrace.pl - post-mortem
On Friday, May 20th, after only 8 months I have finally decided to shut down zlapprace.pl; a Polish IT jobs aggregator I wanted to create ever since I started working in the industry back in 2016. And I’d like to talk a bit about the longest-running, personal, project I have ever worked on in my entire life.
I vaguely remember that this is the type of project I wanted to create since joining the IT industry five years ago. I don’t remember why exactly, but it was always on my mind.
Generally speaking, I have a soft spot for data scraping projects. Also, five years ago JustJoin.IT was just an ugly website with very few job offers. There was plenty of room to grow. I just didn’t have enough skill to get it done. That’s how I was looking at it last year when the project finally went live.
I did a small MVP over a year ago called Joinly. It even gained some traction at that time with top Polish job board owners re-posting it (including Piotr Nowosielski, the owner of JustJoin.It). But the code was bad, the traffic wasn’t great and I simply didn’t know how to make it into an actual product. I shut it down a few weeks later. The code is still publicly available in my GitHub repository.
A few months after that I finally had an epiphany and I thought I know how to make it better.
Say hello to “ZlapPrace”
It took me around a week to re-write the old code. I wanted to create an application that gathers data from ALL possible job sources (i.e. IT companies’ websites) — not only popular job boards.
I started writing custom scrapers and soon had over 15,000 available jobs updated daily.
I went to Facebook Groups to see whether it resonated with people. And it did! Which gave me a huge confidence boost. I really thought I had something.
But the reality hit me again. The number of changes people wanted me to make was too big. My technical skills are still limited and there were issues I couldn’t easily overcome. Re-designing the layout or improving search capabilities were among the two most important improvements I wasn’t able to implement myself.
However, this time I wasn’t fresh on the IT market. I’ve been here for almost 5 years now and I managed to meet amazing and extremely skilled professionals over those years. I am lucky to call some of them “friends.”
The re-design was done by a designer I met in my first IT job. Mareta is a talented artist with a keen eye for detail. We agreed on the general concept of how I’d like this to look and after a few weeks, we were ready. I learned CSS/SCSS in the meantime and implemented layouts once they were approved.
The back-end was a different kind of beast. Luckily, I know a great Senior Engineer, Boguś, who implemented the requested features for me. And I am not going to lie, without his help, I wouldn’t be able to continue with the project.
I did everything I could (or so I thought) to set up my project for success. Released the new version and expected a “wow” effect. But it simply didn’t happen.
I even quit my job to work on the project for a month or two without any distractions, but after just 2 weeks I realized it’s not going to happen for me this time. I let it live for the next 3 months without really paying attention to what’s happening on the site. And then I killed it.
How much did it cost me in total?
Heroku + the domain + design + small one-time expenses = ~3000 PLN (~USD 814)
IMPORTANT: I am not including the cost of NOT WORKING. Which is something a lot of people don’t take into consideration when discussing total costs of a project. If we included those 2–3 months of being “unemployed” then the cost would be much higher.
There are plenty of things I learned about myself in the process. For example, I now know I am not a businessman. I don’t have that “hustle” mentality needed to build a successful, thriving business.
I like building stuff, but selling it is much harder. I now understand why so many engineers have dozens of unfinished projects in their drawers. Maybe if I had someone who could take care of this part of the business and I’d focus on building the product, zlapprace would be in a different position.
I learned that if you want your business to grow at an early stage, you’re going to spend 15 hours every day trying to make this happen. And that wasn’t for me.
Go-to-market timing is critical. Five years ago that could’ve been a different case study. Today, there are more and more aggregators widely available. Mine didn’t have any outstanding USPs.
Not having a business plan was a bad idea. I should’ve thought about it even before I started building the project. Not knowing how to make money won’t help you deliver the value others look for. One of my friends (hey Michał!) asked me about it at an early stage of the project, but I found a lame excuse for not having it.
Not writing tests from the get-go was a terrible idea. Well, not writing tests at the right time was a terrible idea. At first, with only 10 scrapers and simple front-end writing tests would be a time wasted. With 100+ scrapers bugs emerged one after another. Having a test suite would’ve saved me dozens of hours on debugging.
But the story of zlapprace is not a sad one! I learned that I can really hack stuff if I want to. In a way, I followed the Lean Startup framework and I think I’m actually good at it (understanding that building an MVP is better than building the final product from the start). Knowing when to cut corners is an extremely powerful skill to have.
I learned how to code! I could do something before, but this was the first project that I built from the ground up. Not all of it of course, but it was enough to get an offer as a junior RoR developer at one company! I could’ve become a developer…
I am not afraid of speaking to people, getting their feedback, and acting on it. Maybe a Product Management path is something I could explore in the future.
I found a dream job! Because I had enough money saved and a lot of time on my hands I could pick a job I wanted, not a job I needed. I got an opportunity of a lifetime and that wouldn’t be possible without my side-project and everything that lead to its “failure.”
At the moment, I don’t have another brilliant idea I could commit to, and frankly, my day job and a new hobby (stocks) keep me occupied. But if something appears on the horizon, I’ll be more than happy to take another leap of faith.