Many companies accumulate data hoping that one day an algorithm will give them the definitive answer to their problems. But there is no silver-bullet question. Often, we want to go straight to the why, when the first thing we need is to understand what’s happening and where it’s happening. For data to stop being just numbers and become tools, we must start with the basics, asking the right questions. This is how Christian Santín, Business Analyst Programmer, from MindDen, explains it to us:
- Where are we losing money?
- What is working well?
- Where are there inefficiencies?
- What are our customers actually doing?
A real case: “Nobody reviewed them”.
To illustrate this, there’s no need to go into theory. As Santín explains, he recently encountered a case of “digital Diogenes.” A company had been paying for quite expensive infrastructure and servers for years. They had daily pipelines that loaded millions of records into their data warehouse. The problem? Practically no one reviewed them. They only looked at a couple of basic reports and never validated them against historical data. Upon analysis, they discovered metrics that had changed behavior months ago, duplications in processes, and significant deviations that no one had noticed. They didn’t have a problem with a lack of data; they had a problem with a lack of controls and questions.
Santín recounts that “what’s striking is that the company was investing in quite expensive infrastructure to store and process that data, but it wasn’t even doing something as basic as comparing historical trends or detecting anomalies.” He also reminds us that the true value is not in accumulating more data, but in making better use of the data you already have.
The next time someone on your team suggests adding a new dataset “just in case,” ask them a simple question: What business decision are we going to make tomorrow with that information? If the answer is silence, it’s probably time to stop accumulating and start understanding.