HR Case Analyzer
Employee relations cases and HR help-desk tickets: how long they really take, how old the backlog is, and where cases keep coming from.
Reads .xlsx and .csv inside your browser. Nothing is uploaded, and no employee ID is shown or downloaded.
The case log
Drop a spreadsheet here
One row per case or ticket with the date it was opened, and the date it closed if it has. Category, priority, owner and department all help.
The average only counts the cases that ended
“Cases close in twelve days on average” is worked out from closed cases, and the hardest cases are the ones still open. So the figure gets better while the backlog gets worse: close the easy ones quickly, leave the difficult ones, and the average improves. Beside it sits the median time to close with every open case counted for as long as it has been open, the same method the Time to Fill Calculator uses for open roles. Leaving cases open cannot improve it.
A mix of quick questions and slow cases can make even that look fine overall, so it is worked out inside every category too. Relations cases that take months can sit behind a help desk that answers pay questions in a day.
A target missed is missed before the case closes
The usual rate of cases closed within target leaves out every open case already past it, until the day it finally closes. So the rate is given twice: among closed cases, and among every case whose result is already known, counting an open case past its target as missed now. Targets can differ by priority.
Age is where the risk sits
Forty open cases is ten days’ work or a year of neglect depending on how old they are. Open cases are grouped by age, and the oldest are listed by case ID, with whoever holds them. A case with nobody assigned is called out, because it is the one that waits longest.
A repeat is a pattern, not a finding
The same person with several cases in a year, or the same manager named in several, is worth reading together. Repeat help-desk questions usually mean a process that is hard to follow. The tool counts repeats; the employee column is used only to count, and no ID is ever shown. Department counts under the minimum are hidden, since in a small team a case count can identify somebody.
What this does not do
It never judges a case, its merits or its outcome, and it does not read case notes. It counts what the log records: dates, categories, owners. Whether a pattern means anything is for the people who handle the cases to decide.