“…AI didn’t kill analytics. It killed analytics middlemen. …” When the business user can seamlessly — or with very low friction — navigate data through an LLM, ask relevant questions in natural language, and quickly receive highly accurate answers (instead of waiting days or weeks for feedback from the Analytics team), several things will/are happening: Up to 80% of the analytical work as we know it becomes vulnerable. The traditional task of translating business questions into SQL (or other query languages) and present a BI tool to retrieve and provide answers will increasingly be absorbed by AI. Only the top 20% of analysts will truly thrive. These are the ones who deeply understand the business, grasp the causality behind key variables, and know how to ask the right questions — not just the technically correct ones. Decision-making speed will increase significantly. With fewer intermediaries validating data and generating insights, organizations will reduce the cost of making decisions (which is different from the value of the decision itself). The number of analysts will likely decrease substantially. Translator tasks will be automated by AI. However, the analysts who remain will generate multiple times more value than before. Why? Because they will focus less on querying and more on: Framing the right problems Making recommendations Validating assumptions Running scenarios Assigning probabilities to outcomes Connecting insights to business impact In short, they will move from reporting to providing Intelligence. For a deeper perspective on this shift, I strongly recommend reading the well-designed article by Tree Data Point Thursday.
- Pedro
Read on Substack
No comments:
Post a Comment