Nights of Uncertainty
In the world of retail, time is gold. Every second matters when it comes to sales, customers, and strategic decisions. For a well-known department store chain in Central America—with operations in five countries, 42 branches, and thousands of daily transactions—the management of sales relied on a critical application: a system that, every night at 11:00 p.m., was supposed to collect data from every store, consolidate it, generate a summary report, and send it to the top executives (CEOs).
In theory, the process was simple: collect, calculate, report. In practice, it was a nightmare. The system, built in an outdated programming language, had become slow, unreliable, and extremely fragile. When everything worked well, it took about 11 minutes to generate the report. But that was the exception. If any of the databases failed to respond, the application would attempt up to three reconnections, with 20-minute intervals between each attempt.
The result was devastating: delays of 20 minutes at best, missing reports at worst, and sleepless nights of uncertainty for executives who needed reliable data by morning.
The frustration was obvious. Imagine being the IT manager: every night before going to sleep, you had to cross your fingers, hoping the report would arrive. Sometimes it came with incomplete data. Sometimes it never came. And when it arrived late, the information had often already lost its value. Worse still, the developer who built the application no longer worked at the company, and the archaic source code was useless as a foundation for modernization. It looked like a dead end—or so it seemed.
The problem was not only technological; it was psychological and organizational. Every day without reliable reports eroded executives’ confidence in their own infrastructure. Morning meetings often began with the same question: “Did last night’s report arrive?”
The IT department, which should have been a driver of trust and support, was instead seen as an emergency unit—always firefighting, always justifying delays, always explaining the inexplicable.
The consequences were deep:
- Delayed strategic decisions: Without reports, there was no certainty in planning the next day’s operations.
- Emotional strain: Top executives ended their days anxious and began the next ones frustrated.
- A culture of resignation: Gradually, the idea settled that “errors were just part of the system.”
A technical problem had now become a matter of corporate trust.
The call came directly, with the clarity of those who know exactly what’s at stake. When headquarters decided to seek help, we were invited to a meeting where the challenge was laid out with an unambiguous message: time was limited.
The deadline: three months. That was the time given to understand the logic of a decentralized system with multiple databases spread across five countries. Three months to deal with an obsolete codebase whose creator was no longer in the organization.
The system responsible for generating daily sales reports had become unreliable. This wasn’t a minor bug or isolated error—it was a critical bottleneck that, night after night, put the company’s ability to make regional decisions at risk.
The challenge was set on the table: “We need a definitive solution, within a realistic but non-negotiable timeframe.” This was not just another project or another client—it was the chance to build a strategic alliance that could define the future for both sides.
Three months to deliver a concrete answer to executives who—understandably—had lost their patience. This was not an experimental project. It wasn’t a “let’s try, and if it works, great; if not, we’ll leave it.” It was a critical necessity, with a deadline.
In that context, pressure was not the enemy; it was the catalyst. We knew what was at stake wasn’t just a report—it was the credibility of an entire division within a regional organization.
The second obstacle was even greater: the failed application was useless as a reference. Its logic was a black box, trapped in dead code, built in an obsolete language, undocumented, with no traceability. Nothing could be salvaged. We had to start from scratch.
But first, we needed to decipher the real heart of the business: the Microsoft Dynamics RMS sales system. That’s where everything lived. Diving into it was like opening an endless book: dozens of databases with mismatched structures, store-specific rules, and relationships that varied by country. What seemed like straightforward daily sales by store turned into a massive puzzle once scaled to the regional level.
Each store ran on its own server, each country with its own configuration, each database with its quirks.
Structures were not uniform: some tables accumulated sales by daily cuts, others by weekly closures; some records updated in real time, others only at cash register close. Integrating that information meant unraveling a system full of exceptions, foreign keys, and local business rules.
For weeks, most of the work wasn’t programming but understanding: mapping tables, tracing transactions, reconstructing the logic behind every closing process. Each SQL query was like lifting another piece of an endless puzzle.
Time was moving relentlessly. But we knew that without deep understanding, any solution would be fragile. We needed to build on solid ground, even if the mountain of data seemed insurmountable.
The day of the presentation arrived. Expectations were high. The IT manager and executives watched closely. The room was tense, filled with memories of past disappointments.
The IT manager launched the new application. Within seconds, the report appeared on the screen. Silence broke with a startling comment:
—“It doesn’t work!” exclaimed the IT manager.
Glances were exchanged. Sighs of disappointment filled the room. Had we failed?
The CEO interrupted:
—“Why doesn’t it work? What’s wrong?”
The IT manager, in disbelief, replied:
—“The data isn’t real… or it isn’t updated. It’s impossible for the report to generate in less than six seconds when it usually takes more than ten minutes!”
Our response was firm, calm, and confident:
—“Believe it. What you see is the new application. No more endless waits. No more midnight uncertainty. The information is here—real time.”
To remove all doubt, the commercial manager picked up the phone and called one of the 42 stores at random. He asked for that day’s sales figure. The number matched exactly with the report on screen.
Silence turned to astonishment. Astonishment to relief. Relief to excitement. At that moment, we hadn’t just solved a problem—we had changed how they viewed data management.
At the end of the meeting, the IT manager approached us, still torn between disbelief and admiration. He couldn’t understand how not only had we eliminated failures, but also transformed an inefficient process into a fast and reliable system.
The impact was immediate:
- From 11 minutes to 6 seconds.
- Real-time reports, updated every 30 minutes.
- Elimination of failed or incomplete reports.
- Information available in multiple currencies with the flexibility the company needed.
- An improved, smarter, more attractive graphical interface.
But beyond the technology, what truly transformed was the organizational culture: from distrust to enthusiasm, from resignation to proactivity, from “this is just how it works” to “what else can we achieve?”
This case is not just the story of an application. It’s the story of how a department store chain chose not to settle for the obsolete, and how a technical challenge became an opportunity to redefine the future.
In just three months, what looked like a dead end became a bridge to a new way of working—faster, more reliable, and more strategic.
The lesson is clear: the real limits lie not in technology, but in how we choose to face problems. And when innovation meets determination, results don’t just meet expectations—they exceed them, inspire and mark the beginning of a new era.