How do I compare Modules and time periods in Fab?
Use the Module selector and period controls above the dashboard. For Issue metrics and context cost, by default, Fab compares the measured period with the period of equal length immediately before it. Open the header options menu and select About this data to inspect their time boundaries, counts and definitions.
- Open Fab and choose All Modules or the Module you want to examine.
- Choose 1D, 7D, 14D, 30D or 90D. Issue metrics and context cost use a trailing period ending at the report cutoff.
- For calendar dates, choose Custom, select the start and end dates, and select Apply. You can select up to 90 days. For Issue metrics and context cost, custom ranges include both selected UTC dates. A range that includes today ends at the latest source cutoff.
- To choose a different comparison, open the Comparison chip, choose Custom start…, select a UTC date and select Apply. The comparison must end by the measured period's start and begin within 180 days before the report cutoff. Choose Previous period to restore the default. Select a metric to view its daily chart; the comparison series can be hidden with its chart control.
- Open the header options menu and select About this data and confirm the period and coverage before comparing results.
The preset periods for Issue metrics and context cost can begin and end partway through a UTC day. Their first and last daily points are therefore partial days. A custom range follows UTC calendar boundaries, which can differ from your local midnight. Today remains a partial day until it ends and the source covers it. The comparison has the same measured elapsed length. A custom comparison retains the measured period's UTC start time of day on the chosen date. The source cutoff can shorten the measured range. Daily comparisons are omitted where elapsed coverage differs; period totals remain comparable under these time-window rules. Inspect About this data for exact boundaries and source coverage.
Flash counts and Module rankings use a separate read. Their latest time can differ from the work report cutoff shown in About this data. Preset Flash periods end when that read starts. Custom Flash periods cover the full selected UTC dates.
With a Module selected, context figures count Flashes targeting that Module. Direct requests and dependency inclusions remain separate in the Module ranking.
If a comparison is unavailable, inspect About this data before treating it as no change. Compare similar kinds of Issues; splitting work into smaller Issues can increase throughput without reducing the work required.