Testing BI Deliverables (A Checklist)
Analytics, reporting, and business intelligence are all about making sense of data and presenting insights that drive business decisions. This often means building reports and dashboards. A key challenge in this process is ensuring the quality of the deliverable before it reaches the stakeholders.
Imagine this: a business stakeholder asks, “How many items did we sell last week in Australia?” An analyst picks up this question, dives into the data warehouse, and finds the relevant fields. They might also explore other data sources like Google Analytics, HubSpot, or platforms like SnowPlow or Amplitude. Once all the data is compiled, the analyst builds a dashboard using tools like Tableau or Power BI to present the required information as tables, scorecards, or visualizations (e.g., bar charts, pie charts).
But here’s the tricky part: how do we make sure the analytical result is accurate and useful? How can we ensure the deliverable is accepted without feedback like, “These results don’t make sense” or “I can’t use this”? Here’s a checklist of quality checks to complete before delivering the report to the business.
- Is the report title descriptive?
- Is the business request addressed properly and completely?
- Is the visualization relevant? (e.g., avoid using column charts for continuous data and line charts…
