Power BI Automation

Lifewood's Power BI datasets were refreshing inconsistently or not at all, so reports served outdated data to stakeholders. With a Pro subscription capped at eight daily refreshes per dataset, manual refresh requests were ad-hoc and unreliable. The team needed a strategy that maximised data freshness within the licensing constraint, without adding infrastructure complexity.

A tiered refresh architecture that respects the eight-refresh-per-day limit while keeping the critical data fresh. Microsoft Fabric handles dataset performance and enables incremental refreshes, which cuts processing time. Schedules are then allocated by how much the data actually moves: critical operational data refreshes eight times daily across business hours, analytical datasets two to three times daily, and historical datasets once daily off-peak.

The whole thing runs on Power BI's native scheduling combined with Power Query optimisation and Fabric's underlying compute. No custom orchestration, nothing extra to maintain.

Constraints force creativity: the eight-refresh limit pushed me toward incremental loads and query optimisation rather than brute-force full refreshes. I also learned that refresh cadence is a business decision, not just a technical one: operational dashboards need hourly updates, historical analysis does not. If I revisited it, I'd add dynamic scheduling based on how often each source actually changes, and automated compression of older partitions to keep the semantic models lean.