Picture an accounting manager at a mid-sized trading company. The firm just installed a new business intelligence (BI) system. The software runs, the server hums, the reports are ready. But every morning, the manager falls back into old habits: opens Excel, types in numbers by hand, prints a summary, and faxes it to the general director. The new system sits untouched. This scene repeats itself across many small and medium-sized businesses. The problem is not the software. The problem is whether people actually open it.
A BI system is a computer program that pulls data from different parts of a company — sales, inventory, accounting, production — and turns it into tables and charts that managers can read quickly. The core idea is straightforward: a manager arrives in the morning, glances at a screen, and answers questions like ‘what did we sell last week, which product made money, which customer has an overdue balance’ without digging through folders or asking three different people. On paper this sounds very useful. In practice the picture often looks different.
Many BI projects make one key mistake: the software vendor or the consultant measures success by how many reports they deliver. The result is a system with hundreds of reports. Sales report, inventory report, customer report, region report, product group report — the list grows. Who actually opens these reports? Usually very few people. Understanding which report does what requires its own expertise. A sales representative or a warehouse clerk looks at that long list and has no idea where to start. So they do not start. They close the screen and go back to what they know. The system becomes decoration.
The right measure here is not how many reports exist, but how many people open how many reports on a regular basis. Tracking this does not require anything complicated. The software itself keeps a record of who opened which report and when. Call this a ‘usage log.’ If the system holds a hundred reports and eighty of them have never been touched, those eighty reports represent wasted time and wasted money — both the licence cost and the hours spent building them.
The fix starts with cutting the report list down. For each department, identify the two or three reports that genuinely matter. For the sales team, perhaps ‘top-selling products this month’ and ‘overdue receivables’ are enough. For accounting, ‘monthly income and expense summary’ and ‘current cash position’ may cover most needs. These reports should be clean and simple — not packed with columns, colour codes, and nested charts. A manager should look at a report and understand it in five seconds. Simplicity drives usage. Usage builds trust. Trust makes the investment worthwhile.
The second step is real user training. This does not mean a two-hour presentation in a meeting room. Real training means a sales manager sits at his own desk, opens his own data, and learns to read his own report — with someone knowledgeable sitting next to him, answering questions as they come up. This kind of hands-on, one-to-one work is far more effective than any group session. People learn when they see their own numbers. The moment a person says ‘this chart shows what my region sold last month,’ the software stops being foreign to them.
There is also the matter of resistance. When people switch to a new system, they naturally push back. This resistance usually does not come from laziness — it comes from uncertainty. ‘What if I misread a figure?’ ‘What if I take the wrong number to my director?’ These are real fears. Addressing them means creating an environment where opening a report, clicking around, and asking questions is safe and encouraged. Wrong clicks do not break the system. Questions are not embarrassing. Without this environment, adoption does not happen regardless of how good the software is.
For a small or medium-sized business owner investing in a BI system, the question to ask six months after launch is simple: how many people on my team open this system every day? If the answer is satisfying, the project is working. If the answer is ‘two or three out of twenty,’ then no matter how many reports the system contains, the investment has not paid off. Choosing the right software matters, but tracking actual usage after the launch — that is where the real work begins.
This article was originally written in Turkish by Gökhan MERCANOĞLU on March 31, 2003 and has been automatically translated into English and other languages using machine translation.