Slow Queries. Frustrated Users
To begin their transformation, Apotek 1 decided to migrate from an on-premises Oracle database to the Oracle Autonomous Data Warehouse in the cloud because the promise of improved performance, scalability, and cost savings made a logical first step. The company also implemented IBM Cognos Analytics as its BI and reporting layer and the migration seemed successful at first, with notable performance gains in the short term.
However, the company soon encountered issues with performance consistency. Query response times varied significantly, ranging from seconds to hours, making it difficult for users to rely on the system for timely insights. Dashboards and reports often timed out or returned errors, severely limiting the scope and depth of analysis. Users were forced to aggregate data and restrict the historical range they could analyze, which had an impact on Apotek 1’s ability to uncover timely insights.
One notable example was the inability to process data from the 6-8 weeks prior to each campaign, which hindered Apotek 1’s capacity to identify patterns, anticipate issues, and resolve stock availability problems proactively. At the same time, their move towards a self-service analytics model, where users could create their own reports and dashboards, became brittle and prone to breaking. This frustrated users, limiting data interactivity, analysis and adoption.
The analytics team spent hours tuning and optimizing the system, with little success. As frustrations grew, it became clear that the technology was impeding Apotek 1’s analytics goals, rather than enabling them.