I built a dashboard, then studied what it changed

For my MBA term project at Istanbul Gelisim University I did not write about a problem. I built something that solved one: a business intelligence dashboard that pulls multi-channel ecommerce operations into a single view. The building was the research. That is what Design Science Research means, and it is the most honest kind of research I know.

The MBA is non-thesis, on the English track, and the term project gave me a frame for something I had been doing instinctively for years anyway.

What Design Science Research actually is

Most research describes. You study a problem, measure it, and write down what you found. Design Science Research does something different: you build an artifact that addresses the problem, then you evaluate what it changes. The knowledge comes from the making and the measuring, not from observation alone.

It is, in academic clothing, the same instinct I have had since my engineering degree: the proof of an idea is whether it runs.

The problem: one business, five versions of the truth

A multi-channel ecommerce operation does not have a data problem. It has a data location problem. The Amazon numbers live in Seller Central. The Shopify numbers live in Shopify. Ad spend lives in two or three platforms. Inventory lives somewhere else again, and none of them agree on the same day.

So the operator does what operators do: they keep five tabs open and reconcile reality in their head. That works until it does not, which is exactly the point where a business is growing fast enough to need the clarity most.

The business did not have a data problem. It had a data location problem.

The artifact: one view

The dashboard I built does one stubborn thing well. It pulls the operational signals that actually drive decisions, across channels, into a single view: where revenue is really coming from, which inventory is healthy and which is dragging, where the tracking and the sales disagree, and what needs attention this week.

The hard part was never the charts. It was deciding what to leave out. A dashboard that shows everything is just the five tabs again, stacked. The research question underneath was: what is the smallest set of signals that lets an operator make the right call without opening anything else?

What it changed

The measurable change was not “prettier reports.” It was faster, calmer decisions. When the truth lives in one place, the weekly operations review stops being an archaeology dig and becomes a five-minute glance. The arguments about whose number is right disappear, because there is one number.

That is the finding, and it generalizes: clarity is an operational asset, not a cosmetic one. A team that shares one version of reality moves differently from a team assembling it fresh every Monday.

Why this is how I think about everything

I did not pick Design Science Research because it was on a syllabus. I picked it because it is already how I work. I do not hand a client a report on what is wrong. I build the system that makes it right, and I judge the work by whether the business runs differently afterward. The dashboard was just the version of that with a citation list attached.