A breach percentage without its eligible ticket count can send a service lead to the wrong site. A small location can show a sharp rate from very few tickets, while a larger location may account for most of the actual breaches.
Latch’s Service delivery report starts with closed service tickets that have an SLA deadline. It excludes Duplicate, Rejected, and Cancelled tickets. Its resolution view keeps the total, met count, breach count, and eligible denominator together before breaking the same cohort down by geography.
Historical capture from 17 September 2026 with synthetic records. Daily totals, attainment, median, and P90 describe eligible closed tickets in the selected window.
Follow one cohort from the daily trend to a site
The overview answers three separate questions: how many eligible tickets closed, how many met their deadline, and how long resolution took in working time. The daily chart uses closure dates. Median and P90 show the middle and slower ends of the resolution-time distribution. A day with no eligible tickets has no rate to report, so it remains a gap rather than a fabricated zero.
From there, the geographic breakdown groups outcomes by region, city, and site. Each row shows its volume, met and breached counts, attainment rate, and working-time median and P90. Unknown regions, cities, and tickets without a site remain visible in the breakdown; they do not disappear from the denominator. The report computes each level from its underlying tickets rather than averaging site medians into a regional value.
Historical capture from 17 September 2026 with synthetic records. The location names and counts are illustrative, not customer outcomes.
Geography comes from the ticket’s resolved site and the site’s current location fields. It does not report a staff member’s live position or prove that someone visited a location. If a site record changes later, the report does not establish the historical location at the time of every ticket. Treat the chart as a way to select work for review, not as proof of cause. Compare eligible volume and deadlines before changing a process; the rate is a signal to investigate, not an explanation on its own.
Inspect the tickets before changing the process
Suppose one site appears to have a lower attainment rate. Compare its eligible count with the rest of the table, then open the contributing tickets. Check the original deadline, recorded working-hours policy, close time, assignment history, and any reopen or recurrence pattern. A location can reveal concentration; it cannot tell whether access, parts, triage, or policy caused the delay.
The reporting workspace guide explains why created-date and close-date cohorts differ. For work spanning several locations, the multi-site rollout view follows site tickets to completion. Inspection records can capture device coordinates and accuracy as supplied fields, as described in coordinates in an inspection record. To check the clock arithmetic against business hours, holidays, and a pause assumption, use the SLA deadline calculator.
Start with the count behind the rate, then follow the affected ticket. A geographic signal earns action only after its denominator and case history support the decision.