The feasibility of a metric depends heavily on the availability and quality of its underlying data. This becomes even more complex when sourcing data from global systems, due to different time zones, formats, and processes.

Here are two real-world examples that illustrate these challenges and the strategies we developed to address them.

Cross-Regional Data Synchronization

The first challenge emerged when gathering daily data from three geographic regions.

The time required for data collection, processing and dashboard update resulted in a full day's delay. This substantially impacted the metric's value for our stakeholders.

The solution involved implementing a staggered data ingestion and incremental dashboard updates. Metrics were labeled as "partial" until all data had completely arrived.

System Integration and Data Normalization

In another project, we had to merge transaction data from two different systems. Each system maintained different data structures, requiring us to consolidate the information into a smaller common set.

We analyzed which metrics could still be calculated accurately and documented all limitations. For example, certain dashboard filters could not be implemented due to missing attributes in the merged dataset. This analysis also enabled clear communication with stakeholders about the trade-offs involved.

These examples represent just two of the challenges you may encounter when developing metrics. Transparent communication throughout the process is key. It manages stakeholder expectations, clarifies trade-offs, and enables informed decisions about how to refine or pivot a metric.