Performance improvement

# Make [screen or call] load in under [target] ## Intent[Screen or call] takes [today's number] for [who]; it needs to take under [target] because [impact]. ## Context- Measured with: [tool, dataset, device or network]- Suspected cause: [query, render, payload], see [`path/to/file`]- Data size: [rows, items, payload size] ## What to do1. Measure and record the baseline with [method].2. Fix the largest cause first; [ideas if you have them].3. Measure again the same way and add the numbers to the ticket. ## Constraints- Results stay identical, including sorting and permissions.- No caching that shows stale data for longer than [time]. ## Out of scope- [Other slow screens] ## Open questions- [Which data size and device is the target measured on? Remove it when settled.] ## Acceptance checks- Measured with [method], [screen or call] takes under [target] on [dataset]- Results are identical to before for [sample]- Before and after numbers are recorded on the ticket