Create a reproducible performance comparison and interpret coverage.
Open Analysis to compare sites over the same period. The selection is limited to 12 sites so the comparison remains manageable. Measurements are requested when you start the analysis.
The results can include production, normalized production, PR, availability, expected-production comparisons and a comparison table. Values depend on the measurements and configuration available for each selected site.
Compare the same period across sites and use specific production to account for differences in installed capacity. Fictional demonstration data.
For a broader saved view, an asset comparison dashboard can contain up to 100 selected installations. This is a different scope from the focused Analysis comparison above, which accepts up to 12. Neither view should be presented as a single 360-site comparison chart.
The example dashboard, North portfolio · first 100 sites, covers 100 of the North group's 120 installations. Use the same completed month and compatible measurements to find contrasting rows. Then choose a few relevant peers for Analysis, or open one installation to inspect its time series. A low result is a starting point for investigation, not a diagnosis by itself.
To review all 360 sites, define four non-overlapping working batches of 100 + 100 + 100 + 60. If reviews must stay within the three regional groups, use 100 + 20 per group instead. Keep a record of which batches have been reviewed so the last 20 or 60 sites are not omitted.
For one combined deliverable covering all selected installations, use Reports. The dashboard and report screenshots show separate fictional snapshots; their displayed monthly values are not intended to reconcile.
Specific production uses energy and DC capacity. Portfolio-level normalized values must be read with their applicable sites and coverage; they are not a simple arithmetic average of every visible percentage. A missing site value does not mean that site produced zero.
Comparisons over a period including today are incomplete. Unify avoids presenting a percentage change as a finished-period comparison when the current day is involved. Review the reference dates carefully when comparing against the previous year.
If two similarly sized sites have different yield, first compare specific production and coverage. Then open the weaker site's chart. Check irradiation or expected production, alarms and controller restrictions. A production difference by itself does not prove a hardware defect.
Use the available measurement download for further analysis, or create a report using the same sites and period. Record the chosen range and any incomplete-data notes with your conclusion. A dashboard that moves with today's date is not a fixed historical report.