The Project That Taught Us Everything

In 2013, Matthias and Jan were doing race timing themselves for the Flemish Triathlon and Duathlon Federation. Athletes crossing finish lines. Split times for swim, bike, run. Rankings calculated instantly. Results online before the last runner finishes.

They needed something robust. This was the pre-5G era—flaky mobile connections, first-generation iPhones, venues with no reliable internet. The software had to work offline, sync when it could, and never lose a single timestamp.

So they built it. VTDL is still using it today.

That system has been running for 12+ years. 99.9% uptime. Hundreds of races. Thousands of athletes. Not a rewrite—the same system, maintained and evolved.

That project—long before glue.blue existed—taught us more than any other about what makes software last.

Lesson 1: Boring Technology Wins

The sexiest framework in 2013 is a legacy burden in 2026. We’ve watched trends come and go:

  • The JavaScript framework of the month
  • The database that was going to change everything
  • The architecture pattern everyone had to adopt

The VTDL system runs on boring technology: PHP, Bootstrap, MariaDB. Reliable. Well-documented. Easy to maintain. When something breaks at 7 AM on race day, we’re not debugging some abandoned library’s weird edge case.

Pick technologies with a future, not a hype cycle.

Lesson 2: The Client Relationship Outlasts The Code

Twelve years with VTDL means twelve years of trust. They know we’ll pick up the phone. They know we’ll fix it fast. They know we’re not going anywhere.

That relationship is worth more than any contract. It’s led to referrals, expanded scope, and the kind of collaboration where good work actually happens.

Software is a relationship, not a transaction.

Lesson 3: Maintenance Is The Product

The first version of any software is maybe 20% of the work. The other 80% is what happens after launch:

  • Feature requests from real usage
  • Edge cases nobody anticipated
  • Integration with new systems
  • Security patches
  • Performance optimization
  • The thing that worked fine until it didn’t

Most software companies want to build new things. We’ve learned to love maintaining old things. That’s where the real value is.

Shipping is the beginning, not the end.

Lesson 4: Simple Beats Clever

Early in our career, we wrote clever code. Abstractions on abstractions. Patterns from the textbook. Impressive architecture diagrams.

Now we write simple code. Obvious code. Code that a developer in 2035 will read and immediately understand.

Clever code is fun to write and miserable to maintain. Simple code is boring to write and trivial to maintain.

Optimize for the maintainer, not the author.

Lesson 5: Say No More Often

Not every project is a good project. We’ve learned to recognize the signs:

  • Unclear requirements that keep shifting
  • Budgets that don’t match ambitions
  • Stakeholders who can’t make decisions
  • “We need it yesterday” urgency
  • Solutions looking for problems

Saying yes to the wrong project means saying no to the right one. It means stressed team members, scope creep, and work we’re not proud of.

The best projects are the ones we almost didn’t take.

Lesson 6: Belgian Problems Need Belgian Solutions

We tried the “scale globally” thing. Build something universal. Target everyone.

It doesn’t work. Not for us.

Belgian traffic fines have specific formats. Belgian rental law has specific requirements. Belgian accounting has specific standards. Belgian health indices have specific calculations.

Our best products—the Loops—work because they’re obsessively Belgian. We understand the context because we live in it.

Global ambition, local execution.

What We’d Tell Our 2013 Selves

  1. Trust takes years to build and seconds to destroy. Protect it.
  2. The first version is never the last version. Plan for evolution.
  3. Clients don’t want technology. They want problems solved.
  4. Documentation is a gift to your future self.
  5. Charge enough to do the work properly.
  6. The best marketing is work that speaks for itself.

Still Here, Still Building

Twelve years in, we’re more convinced than ever: good software is a craft. It takes time, care, and commitment.

We’re not chasing unicorn status or exit strategies. We’re building useful things for Belgian businesses, one project at a time.

Here’s to the next twelve years.