Your e-shop works, orders keep coming in, customers pay. And yet you feel that every new feature takes longer than before, the website keeps slowing down and your developers are afraid to deploy. Most e-shops only deal with a redesign when it is too late. But what if an outdated design is not the main thing holding you back?
“Two years ago it took me two weeks to deliver a feature; today it takes a month. And when I tell the developers to deploy, they say — just make sure nothing breaks.” This quote from Vašek Macíček, CEO of Shopsys, from the Exec Talks #72 interview captures exactly the moment when technical debt becomes unbearable. Most companies start dealing with replatforming at five past twelve, when the old system is literally falling apart. And the problem doesn’t show only on the website, but above all in the data that is key for future growth.
Summary for Those Who Don’t Have Time to Read the Whole Article
- Technical debt shows up as slower development, instability and constant “workarounds”.
- But the worst part is not an outdated design — it’s the poor-quality data hidden behind it.
- Without consistent and clean data, you can’t make meaningful use of analytics, personalisation or artificial intelligence.
- The future of e-commerce analytics is heading towards conversational AI interfaces — and they need clean data more than anything before.
- Replatforming is a unique opportunity to rebuild not only the website, but your whole data infrastructure, and to get ready for the future.
How to Tell That Your Platform Is Holding Back Growth
Technical debt is like rust. At first you don’t notice it, but gradually it eats into the whole structure. It’s not just a feeling that “it’s old”. There are concrete signals that your e-commerce platform is no longer enough and is turning into a brake instead of the engine of your business:
- Delivering features takes many times longer: A simple task that used to take days drags on for weeks. Every change requires disproportionate effort and testing.
- The website is slow even after optimisation: Even when you invest in caching and image optimisation, the core of the system is so inefficient that customers leave for competitors who are two clicks ahead.
- Workarounds instead of solutions: Instead of systemic solutions, you stick one “hack” on top of another. As Vašek Macíček says, “you bend what has already been bent”.
- Fear of deployment: The worst stage. Every new release is a risky operation, because you are afraid that fixing one thing will break three others. Innovation stops.
“An e-shop is just a pretty face. Behind it, you need quality technology — ERP, PIM, a data layer. If the backend doesn’t work, the e-shop can’t be good either.”
Vašek Macíček, CEO of Shopsys (source: Exec Talks #72)
An Old Website Is the Least of It — the Real Problem Is Data
When considering a new platform, most companies focus on design, UX and new features. That matters, but it’s only the tip of the iceberg. The real problem that keeps you from growing is hidden below the surface — in your data infrastructure. We see it with clients all the time. For example, at Super zoo, before our cooperation, we identified more than 80 inconsistent parameters and duplicate tags in GTM. There were no standards for data collection at all.
Petr Voves, owner of Ochutnej Ořech, describes it similarly in our interview: “Before, we did it by feel… I could see the basic figures, but we weren’t able to get to more advanced metrics.” (source: YouTube, Digitální architekti). Add missing server-side tracking, which causes tens of percent of data to be lost, and data locked in separate systems (silos) that can’t be connected to BigQuery or Keboola, and the result is clear: you make decisions based on gut feeling and incomplete numbers, which is a road to ruin these days. Without quality data, you simply can’t build a meaningful report, let alone run a company.
Analytics and AI — Why Technical Debt Hurts Twice as Much
Why does data quality matter more than ever? Because the future belongs to advanced analytics and artificial intelligence. Today’s analytics is no longer just about tracking traffic in Google Analytics. It’s about connecting data from different sources — the e-shop, marketing, ERP, CRM — into one whole. Old platforms, however, often can’t collect and pass on data consistently. What’s more, analytics is quickly moving from static dashboards to conversational AI interfaces. Soon you won’t be clicking through charts; you’ll ask a question: “Which product categories dropped in margin last quarter, and why?”
But artificial intelligence can’t work miracles. Without clean, structured and complete data, it won’t answer such a question. As I say in our video with Ochutnej Ořech: “Where does business creak? In connecting data to artificial intelligence systems.” Projects that invest in data warehouses today “will have a competitive advantage in connecting data to AI systems.” The longer you wait with fixing your data infrastructure, the wider the gap between you and your competitors will be. Without quality data, there simply will be no AI — and you are depriving yourself of the key trends in measurement for the coming years.
Not sure exactly where your e-shop is losing potential? We’ll help you find out.
Final summary: Technical debt is not just a slow website or an outdated design. It is an invisible brake whose force multiplies the moment you want to start working seriously with data, analytics and artificial intelligence. Our clients such as Super zoo and Ochutnej Ořech understood that replatforming is the ideal opportunity not only for a new “pretty face”, but above all for building solid data foundations. The results speak for themselves: consistent data across the company, faster and more accurate decision-making and, ultimately, measurable growth. The first step is to stop burying your head in the sand and take a look at where you really stand.
Frequently Asked Questions
1. How do I know if my e-shop has technical debt?
The main signals are a significant slowdown in developing new features, system instability (frequent outages), a growing number of temporary fixes (workarounds) and a general fear in your IT team of deploying anything to production.
2. Why isn’t a redesign enough?
A redesign only solves the visible part of the e-shop (the frontend). Technical debt, however, most often hides in the backend, the architecture and above all in the way the system collects and stores data. A new design on old foundations is only a temporary solution.
3. Can I improve analytics without replacing the whole platform?
Partly, yes — for example, by implementing Google Tag Manager and server-side tracking, you can make data collection more accurate. But if the platform itself generates inconsistent data (e.g. different formats for the same parameters), it’s only a patch. For truly reliable analytics, the problem has to be solved at the source.
4. What is technical debt from a data perspective?
It is a state in which you have inconsistent, incomplete or poorly structured data. It shows up, for example, as duplicate parameters, missing tracking of key events, data locked in different systems with no way of connecting them, or the absence of a unified data model.
5. Why is it important to deal with data together with the platform?
Because the platform is the primary source of most of your business and customer data. Replatforming is a unique opportunity to design the whole data infrastructure correctly from the start. Fixing data chaos retrospectively is many times more expensive and complicated.
Sources
- Exec Talks #72: Vašek Macíček (CEO, Shopsys). YouTube. Available at: https://www.youtube.com/watch?v=sj2FwD0Tqg4
- Interview with Petr Voves (Ochutnej Ořech) about growth through data and BI. YouTube, Digitální architekti. Available at: https://www.youtube.com/watch?v=yn-vmD7L8KI
- Super zoo — case study. Archetix. Available at: /post/super-zoo-how-we-tamed-data-for-international-expansion/
- Ochutnej Ořech — case study. Archetix. Available at: /post/transformation-of-ochutnej-orech/