Over the past couple of years I’ve built and run a niche job board for software engineers looking for work in Dubai and Saudi Arabia.
What started as a small side project eventually grew into something that now gets around 100,000 page views a month and ranks well for a number of fairly specific search terms.
Recently I decided to start building a second job board.
But this time I approached it quite differently.
When I built the first site, I mostly thought about the technical side of the problem. I wanted an excuse to build a full stack web application in Go, and a job board seemed like a straightforward project.
After running one for a couple of years, I realised something important:
The software is not the hard part.
Understanding the industry you’re building for is the hard part.
That insight shaped how I approached the second project.
Job Boards Are Not Really a Software Problem
When developers think about building a job board, they usually focus on engineering questions.
Things like:
- search functionality
- recommendation systems
- user accounts
- dashboards
- fancy frontends
But those things are rarely what makes or breaks a job board.
A job board really lives or dies based on two things:
- Whether it consistently has relevant jobs
- Whether it understands a specific audience
Technically, the web application can be extremely simple.
If you look at many successful niche job boards, they are not particularly complex systems. What makes them work is that they serve a very specific group of people well.
That’s what I kept in mind when thinking about my next project.
Choosing the Niche: Remote Legal Jobs
For the new site I decided to focus on helping lawyers find remote law jobs.
At first glance that might sound like an unusual niche for a developer.
But there was a simple reason behind the decision.
Over the past few years I’ve worked with several law firms on software projects. Through that work I’ve had a chance to see how legal teams operate and how hiring tends to happen in that world.
The legal profession is still quite traditional in many ways, but it has been slowly adapting.
Most legal job boards are poorly structured and tend to mix legal roles together with thousands of unrelated listings.
That makes searching inefficient.
This combination made the niche interesting enough to explore.
Why Domain Knowledge Matters
One thing I’ve learned from building niche sites is that choosing a niche you understand makes a huge difference.
If you build something for an industry you know nothing about, you eventually run into problems.
For example:
You don’t know what terminology people use.
You struggle to write useful content.
You can’t easily tell which jobs are actually relevant.
And you often misunderstand how hiring works in that field.
Those gaps become obvious over time.
If the project grows even slightly, you will eventually need to write articles, guides, or commentary related to the industry itself. Doing that convincingly is extremely difficult if you don’t actually understand the space.
Because I’ve worked with law firms before, I felt comfortable building something around that world.
Not because I’m a lawyer, but because I’ve had enough exposure to the industry to understand how people in that profession talk about work and careers.
That’s a much stronger position than simply picking a random niche from a list of job categories.
Building the Site
From a technical perspective the site itself is intentionally simple.
The stack looks like this:
Backend: Go
Database: PostgreSQL
Frontend: server-rendered HTML templates
There are also a few background processes that keep listings updated and ensure the site stays current.
Nothing about the architecture is particularly exotic.
In fact, one of the things I enjoy about these kinds of projects is how far you can get with very straightforward technology.
You don’t need a complex distributed system to build something useful.
Most of the engineering effort actually goes into:
- structuring job data cleanly
- categorising listings properly
- making the information easy to navigate
Those problems are less about complicated software engineering and more about data organisation and product thinking.
Why Simple Systems Work Best for Side Projects
Another advantage of keeping the system simple is that the project stays easy to maintain.
That matters a lot for side projects.
If a project requires constant attention or complicated infrastructure, it quickly becomes exhausting. But if the system is lightweight, you can continue improving it gradually without it consuming your life.
That’s been my experience with niche job boards.
Once the basic structure is working, improvements tend to happen in small increments:
refining job categories
improving the presentation of listings
adding useful content around the niche
None of these changes are technically complicated, but over time they make the site much more useful.
What Developers Can Learn From Niche Products
Working on projects like this has reinforced something I think many developers underestimate.
Building useful products is rarely just about the technology.
It’s about understanding a specific group of people and creating something that genuinely helps them.
For niche job boards that means understanding:
how hiring works in a specific industry
what job seekers struggle with
how professionals in that field talk about their work
The software itself is just the mechanism that delivers the value.
The real advantage comes from understanding the niche.
Final Thoughts
Starting a niche job board isn’t particularly difficult from a technical perspective.
But making one useful requires a deeper understanding of the people you’re building it for.
If you’re a developer thinking about building something similar, my advice is simple.
Choose a niche you actually understand.
Ideally it should be an industry where:
you know how people talk about their work
you understand how hiring happens
you feel comfortable writing about the space
That combination makes it much easier to build something credible.
The technology is the easy part.
Understanding the audience is the real challenge.
Top comments (0)