A thirty-day DutyMap rollout.

Week zero — the site survey

Before any crew member is invited into DutyMap, an operations lead walks one site with a site supervisor and a dispatcher. The goal is to understand the physical and procedural reality of attendance at that site: where crew actually enter the site, where the geofence boundary should sit, which roles rotate into and out of the site on a typical shift, and which edge cases the supervisor handles on a weekly basis. The survey output is a one-page document that becomes the site’s reference sheet for the next three weeks.

Week one — one site, one shift

The first deployment week is intentionally narrow. Pick one site and one shift — typically the weekday day shift — and roll DutyMap to that cohort only. Every crew member on the shift gets a device invitation and a brief five-minute onboarding; the site supervisor is in the room for the first morning. The goal of week one is not adoption speed; it is to discover every friction the survey missed. The exception queue will look loud. That is fine — most of those exceptions are artifacts of the shift cohort learning the workflow, and the pattern stabilizes by the end of the week.

Week two — add the second shift, keep the site

In week two, the same site adds a second shift — the overnight or the weekend — with the same device-level onboarding. The second shift is where the first site-specific edge cases appear: handover behavior, mid-night device issues, and coverage signals that only show up when a smaller crew is on the floor. The supervisor and dispatch pair are the same from week one, and the reference document is updated as the shift teaches the team. If the exception queue on the day shift is still noisy by Friday, do not expand in week three. A rollout that earns trust beats a rollout that hits a date.

Week three — expand to two sites

Week three introduces a second site. The two-site stage exposes the dispatch-level pattern: one dispatcher watching two sites, with the same coverage lane showing both, and the exception queue now ranked by severity across sites rather than within a single one. This is the first time the map is the queue claim gets tested on a real operator. Budget a day in the middle of the week to sit with the dispatcher, watch how they triage, and tune the exception sort order if the defaults do not match how the dispatcher actually works.

Week four — region rollout and the retro

In week four, DutyMap expands to the remaining sites in the region. Every site that joins brings its own survey, its own geofence tuning, and its own supervisor orientation, but the operations-lead time per site is now half of what it was in week one because the playbook is known. At the end of week four, the operations lead runs a written retrospective with the supervisors and dispatchers covering three questions: what did DutyMap surface that you were not already seeing; what did it mis-surface; and where is the largest single remaining pain point. The retro feeds the month-two plan.

Role changes that go with the rollout

A DutyMap rollout does change two roles in subtle ways. The site supervisor gains the same coverage view as dispatch, which narrows the gap between the two roles and generally pushes more handling authority to the site. The dispatcher, in exchange, moves from tracking shifts to triaging exceptions, which is usually the work they wanted to be doing in the first place. Both changes are usually welcome, but they are real, and an operations lead who names them aloud in week zero keeps the rollout from stumbling on expectations.

What to measure

Three numbers tell you whether the rollout is landing. First, verified check-in rate — the share of shifts that open with a verified arrival inside the geofence window. A healthy deployment is in the high-nineties by week three. Second, exception time-to-resolution — the median minutes between an exception entering the queue and closing in any of the four valid end states. A healthy deployment halves the median between week one and week four. Third, coverage-gap minutes — minutes per week where at least one site is under its target headcount; this should trend down as the scheduling and the rollout co-evolve.

Things to avoid

Do not run a dark-launch where crew check in through DutyMap but no one looks at the data; that trains the team to distrust the tool. Do not run DutyMap in parallel with a legacy attendance system for more than a week; the two systems will diverge and the supervisor is suddenly maintaining both. And do not grade supervisors on the numbers in the first month; exception counts are noisy in a new deployment and turning them into performance metrics will recreate the buddy-punching problem in a different form.

After the first month

At the end of the first month, DutyMap is embedded in the shift. The operations lead then shifts from running the rollout to running the monthly review. The weekly miss report, the drift tallies, and the coverage-gap summary become part of the operations cadence. The rollout is complete when the tool disappears into the work, and the conversations at shift handover are about the site, not about the app.