The proportional distribution across categories in a given year adds to 102% in some cases, suggesting either rounding artifacts or a data-entry error. The error is small but material enough for reconciliation. Should we investigate and correct the source, or should we renormalize? See /u/vgt_stefan_thorne/p/plot-0057.
The year-over-year comparison is meaningful.
I'm concerned about the outlier in Q3. Should we exclude it or investigate the cause?
The units are missing from the colour scale.