People choose what to share
A benchmark uses volunteered milestone intervals. It is not a complete count of every person or case moving through a process.
Journey benchmarks
A little more context for the in-between.
Shared milestone reports can add context to the waiting between steps. We show only groups that pass privacy checks, keep official agency information separate, and never turn a cohort into a prediction for your case.
There is no current public comparison to show.
Togetherward publishes a result only when the current contributed evidence can support a careful summary and passes the privacy checks. A quiet page is more honest than a number without enough context.
This page gives group context only. It does not evaluate an individual case or call for any action.
How to read this page
A shared picture, with its edges left visible.
Pending reports stay in the analysis
Cases that have not reached the next recorded milestone are treated as still being observed, rather than dropped or counted as completed.
Every published result describes a group
We show a broad sample band and rounded estimates. There are no individual timelines, case matches, or contributor details in a public result.
See how a comparison is built
Each comparison follows one supported milestone pair and a quarter in which the first milestone was recorded. Different journeys, offices, and periods are not blended into one result.
The estimator is Kaplan–Meier. A report without the next milestone remains in the at-risk group through its latest recorded follow-up instead of being dropped. The displayed midpoint is rounded to weeks, and its uncertainty interval uses Greenwood variance with log-log confidence bounds. NIST’s worked Kaplan–Meier example shows the estimator with completed and right-censored observations.
The method does not model every journey shape. Transfers, reopened records, competing events, and intervals whose exact timing is unknown are outside this version’s estimate. A group that cannot pass the privacy checks stays off this page; internal privacy thresholds and suppressed counts are not published.
To question a published summary or report a correction, contact Togetherward support. Identify the public journey and start period, and leave private case dates, government reference numbers, and other personal details out of the message.
Looking for official information?
Community observations and government processing reports describe different things. We keep them separate so a group summary is never mistaken for an agency estimate.