When we founded the agency, one of our first clients came to us after a bad experience with their previous agency. The project had promise, but the codebase they inherited was a mess. It was full of tightly coupled components and quick fixes that made it difficult to improve anything without breaking something else.
So we did a full audit and presented our findings. The client agreed. There was serious technical debt. We agreed on a monthly fixed budget to work on improvements, stabilisation, and refactoring.
At least, that was the plan.
What actually happened was very different.
Over time, the focus quietly shifted. Instead of using the budget to clean up the codebase and invest in long-term quality, the money went toward new features and urgent bug fixes.
We did our best. We stabilised the platform. Solved hundreds of issues. Onboarded more of their clients. But the messy foundation remained untouched.
And now, years later, the cracks are showing. The platform is outdated. New technology has outpaced them. And they are stuck reacting to problems rather than driving improvements.
Part of the challenge lies in the business model. The service was being offered at a price point that didn’t realistically support the level of complexity involved, which made it difficult to invest in the technical improvements the platform needed.
That number might sound okay on the surface, but it simply doesn’t work.
We’ve built Grow As Root, our own modular ecommerce stack. Even with no customisation, we start at 1,000 euros per month. If clients want custom features, they pay on top.
Trying to offer deep support and feature development for half that cost, without even a basic technical investment buffer, is a race to the bottom.
But the bigger problem is confidence. The client isn’t comfortable raising prices or setting proper boundaries. Not even for new customers. So the gap keeps growing, and the ability to act keeps shrinking.
Another issue is indecisiveness. Internally, the client struggles to commit to a strategy. One day they hear advice from one person and follow it. The next day someone else tells them something different and they change direction again.
There’s no stable plan. No roadmap. Just reaction after reaction.
Without clarity, every decision becomes expensive. Teams waste time building, unbuilding, and rebuilding. And eventually, people burn out or give up.
The worst part is that the longer you wait to address technical debt, the more expensive it becomes.
Right now, the platform is outdated. Any investment they make today costs far more than it would have a year or two ago. And because there’s no plan, every euro spent feels like a gamble.
As a partner, this is frustrating. We want to help. But without proper prioritisation, we’re limited to applying band-aids.
This whole project reminded me of going to the dentist. Skip your check-ups for too long and one day you’re told you need an eight-hour procedure that costs a fortune. Had you gone every six months, the issues would have been spotted early and fixed quickly.
The same logic applies to software.
A small code issue today can evolve into a security risk, a scaling bottleneck, or a support nightmare tomorrow. The deeper these issues go, the more they erode the rest of your platform. And when you finally decide to act, the fix is no longer simple or cheap.
Too many people treat software as a one-off project. You finish your online shop, pay the final invoice, and think the job is done.
But software is never done. It evolves. Your customers evolve. Your industry evolves. Your competitors are improving their platforms every month.
You need to treat your shop like a living system. That means allocating budget not just for new features, but for maintaining what you already have.
We often get asked what a sensible budget looks like for maintenance and upgrades.
There’s no one-size-fits-all number, but here are some rough guidelines:
If your shop is only 10 to 15 percent of your overall revenue, your investment can reflect that. But if ecommerce is your primary sales channel, you cannot afford to cut corners.
Launching a new shop or system is just the beginning. No matter how clean the initial build, things will need to be changed, optimised, and eventually rebuilt.
Over time, you’ll discover better ways to do things. Some features will become obsolete. Others will prove more popular than expected and need scaling. You may even outgrow parts of your architecture.
That’s all normal.
But if you treat your launch as the finish line, you’ll fall behind fast.
Your future stability depends on decisions you make today. Are you tracking which parts of your platform need refactoring? Do you have documentation for all your third-party tools? Are you removing unused code? Do you have a plan for managing technical debt?
If not, now is the time to start.
You don’t need a full rebuild. You just need a mindset shift. Think in systems. Prioritise maintainability. And work with partners who help you stay focused on the long term.
Any software platform is a living organism. It needs ongoing attention, even when everything looks fine on the surface.
If you stop investing, things decay. And when they break, fixing them costs a lot more than maintaining them would have.
The most resilient ecommerce brands don’t just chase new features. They manage their architecture like an asset. They plan upgrades. They schedule clean-ups. They build for tomorrow.
If you're unsure where to start, we’re happy to take a look at your current setup and help you map a strategy. And if you’re already feeling the weight of past decisions, we can help you break the work into manageable steps and get things back on track.
Better late than never. But sooner is always better.
Image from insung yoon on Unsplash
Every dev has that one nightmare codebase. Share yours — or ask us to review your current setup and map a path toward cleaner, scalable architecture.
02.12.25
16.11.25
