Extrapolix

Wholesale Distribution

Wholesale Distribution

Pre-Implementation Challenges :

Chapter One: The Legacy Years, 2005 to 2015

Extrapolix began working with this distributor in 2005, when the business carried 8 FMCG brands on a single system. As the business grew, the software’s answer to growth was simple: add another brand, add another computer, install another instance. It worked, for a while.

By 2015, ten years later, the business had grown to 20 FMCG brands. The office by then had 20 physical computers, each running its own separate copy of the software for a separate brand. That image, twenty brands and twenty computers standing in for them, is the clearest picture we have of what an architecture looks like once it has quietly reached its limit while the business kept growing anyway.

The consequences were real and expensive. If one customer bought from four of those twenty brands, getting that customer’s true outstanding balance meant logging into four separate systems and adding it up by hand. The final books of accounts were assembled once a year by exporting data out of all twenty systems and reconciling them in another piece of software entirely. Any figure that crossed brand lines meant a manual consolidation in a spreadsheet. As bill volume kept climbing year over year, running one brand per computer was becoming, quite literally, impossible to sustain.

What We Found

The finding that drove what came next was not about a missing feature. It was about an architecture that had run out of room. The twenty-computers state was not a temporary inefficiency to be tidied up. It was the original software doing exactly what it had always done, one instance per brand, long after the business had outgrown that pattern. The cost of that mismatch was not visible on any single day. It was being absorbed quietly, in employee hours, in reconciliation overhead, and in blind spots at the management level. Recognizing that kind of problem means looking at the business as it actually is, not as the existing software assumes it to be. Most software fails before the build begins.

The Harder Finding

The technical answer, a rebuild on a proper multi-brand architecture, was clear well before 2015. What took ten years was something else entirely. When we recommended the move, the founder was skeptical, understandably, given the scale of change involved in rebuilding the system a twenty-brand business depended on. He agreed because of the decade of relationship behind the recommendation, and the trust that decade had earned.

That is not a footnote to this story. It is the story. The software work was buildable by any competent team willing to take the time to understand the business. What made the rebuild possible was ten years of demonstrated reliability, not any single piece of code. Software is easy. People are hard.

What We Built

The 2015 rebuild was a cloud-based trading and distribution platform covering the business end to end, designed specifically around the limits the twenty-computer state had exposed.

Chapter Two: Eleven Years on the Cloud

The platform built in 2015 has now run continuously for eleven years. In that time, the brand count has grown from 20 to more than 50, well over double. Two entirely new product categories, consumer durables and pharmaceuticals, have been added on top of the original FMCG business. Route coverage has expanded past 100 across the distributor’s branches, and billing points now number 60. Daily bill volume has grown from the 2,000-plus it carried at rebuild to more than 3,000 today.

The architecture has absorbed all of that growth without needing to be rebuilt again. Every year the platform keeps running is itself the proof that the 2015 decision was the right one.

Chapter Three: What Comes Next

A migration to newer technology, with a further round of feature enhancements, is now under discussion. It is a substantial undertaking, and it will follow its own round of discovery before timing is set. This third chapter is not a sign that the cloud platform failed. The eleven-year run is the proof that it succeeded. It is being planned because the business, and the technology available to it, have both moved forward, and a twenty-one-year relationship has reached the point where the next generation makes sense.

The Outcome

Twenty-one years in, this is still an active, forward-looking relationship, not one running on momentum. The business is preparing to invest in its next architectural generation with the same team that built the last one.

Let's Start With a Conversation

If you run a distribution business and any of this feels familiar, the next step is a conversation, not a pitch.

Scroll to Top