Coverage gaps on a distributed team, and how to spot them

Spread a team across time zones and a quiet problem appears that nobody planned. There are stretches of the working day when, on paper, you have a full roster, but in practice no one was online: a handoff window where one region has logged off and the next has not yet started, a Friday afternoon that thins out earlier than you realized, a holiday one office observes that clusters with a light day in another. These are coverage gaps: spans of time when a team, a region, or an on-call rotation had nobody available to respond.

Coverage gaps are nearly invisible while they happen. You are not watching Teams in the hour your East Coast people sign off and your West Coast people step out for lunch. The gap shows up later, as a ticket that sat untouched for hours or an incident that waited for someone to wake up, and it reads as a one-off when it usually is not. This post is about treating coverage as something you can see and plan from your team's presence history, as an operations exercise about the schedule, not a question about any one person.

Why coverage gaps hide in the live dot

Microsoft Teams gives every person a presence dot: available, away, busy, offline. It is genuinely useful for the question it was built for, "can I reach this one person right now." It is the wrong instrument for coverage, for three reasons.

• It is one moment. The dot reflects this instant. Coverage is a question about spans of time, and a snapshot cannot describe a span.
• It is one person. You read dots one name at a time. Coverage is a property of the whole group: was anyone online at 3pm Thursday.
• It does not remember. Glance away and the moment is gone. Nothing is recorded, so you cannot look back at last week and ask where the team was thin.

The window where nobody was around is exactly the window you were not watching, and the dot keeps no record of it. To see a coverage gap reliably you have to read presence as history: each person's status sampled through the day and stored, so a span of "nobody available" becomes something you can point at on a timeline.

A note on framing. Presence shows when someone was online in Teams, not what work they did. A coverage gap is a property of the schedule, not a fault of a person, and reading it any other way poisons the conversation you are trying to have. The same history also shows why the green dot is never the whole story: the person who stays green all day but is rarely around, and the person who looks away at 2pm but is steadily online in the evening. Either way, read the pattern, not the glance.

The three gaps that hit distributed teams

Coverage gaps are not random. On distributed teams they cluster into three predictable shapes.

• The handoff window. The seam where one region logs off and the next starts. Handoffs that look continuous on a planning doc often leave a dead hour once people drift offline before their nominal end time.
• The quiet stretch. Inside a single region, coverage thins below one at the edges of the day and on Friday afternoon. Nobody decided to leave the desk uncovered; people just wrapped up.
• The holiday cluster. Holidays and time off do not line up across offices. When two regions both run light on the same date, a window you thought was double-covered drops to zero.

In all three the fix is structural: shift a start time, stagger an end, move an on-call boundary, plan around a holiday. These are scheduling facts you could not see until you looked at presence over time, not performance issues.

A worked example

Make it concrete. You run a six-person support team meant to cover 07:00 to 22:00 US Eastern, split three on US Eastern time and three on US Pacific time. The East Coast opens, the West Coast closes. On paper, continuous. You pull a week of presence history, roll it up to how many people were available each hour, and look for any hour near zero. A simplified weekday slice might look like this.

The signal jumps out: the 15:00 hour is running on one person, right when the East Coast has signed off and two of the three Pacific people are at lunch. That is a handoff-window gap. The customer-facing version is "tickets that arrive around 3pm Eastern wait longest," which you may already have noticed without knowing why.

You did not watch anyone to find this. You used three views from presence history: online-time per period, overlap windows (where regions were available at the same time), and low-coverage stretches (hours below your target). Together they turn a vague "we seem slow mid-afternoon" into a specific, fixable hour.

What to do once you can see the gaps

Seeing the gap is most of the work. The remedies are ordinary scheduling moves, and the data lets you pick the smallest one that closes the hole rather than over-correcting.

• Stagger hours at the seam. If the 15:00 hour is thin, moving one lunch or shifting one East Coast person to start and end an hour later covers it without adding headcount. History shows which hour and how much, so you adjust by the minimum.
• Adjust the handoff boundary. If a region drifts offline before its nominal end time, move the official boundary to match reality, or extend a small overlap so the seam is genuinely continuous.
• Plan around holiday clusters. Once you can see which dates drop two regions at once, you can arrange light cover in advance instead of finding out from a missed ticket.
• Right-size on-call. If a rotation leans on one person during a recurring window, that is the argument for a second name on the slot, grounded in data rather than a hunch.

The throughline: these are decisions about schedules and staffing, not about individuals. The team is the unit of analysis. For the deeper how-to on the time-zone side, we cover it in how to track team availability across time zones.

Doing this responsibly

Tracking presence to plan coverage is reasonable. It also erodes trust fast if done quietly, or if a schedule problem gets pinned on whoever happened to be on shift. A few principles keep it clean.

• Tell the team presence is tracked, and why. Say it openly, coverage planning included. Tracking people do not know about is what costs you trust.
• For coverage, keep the unit the team. Roll presence up per region and per period. The coverage question is "was the window covered," not "what was this individual doing."
• Collect the minimum. Presence and basic directory fields, nothing more. No message content, no screen capture, no keystrokes. Retain it only as long as you need.
• Use it to start a conversation. A thin hour is a prompt to adjust a schedule, not a judgment on whoever happened to be online.

How Presify helps

Presify is a Microsoft Teams presence tracker. It samples Teams presence (available, away, busy, offline) through the Microsoft Graph presence API on a short, regular interval, stores each change, and keeps the history, so you can see who's online, when, and for how long. Coverage is one of the views that history gives you.

• See the pattern. A presence timeline for each person and each team, every recorded status change on a 24-hour day view, and online-time totals and trends. Roll the same record up per team with per-person time zones and a business-hours overlay, and a thin handoff or a quiet Friday shows up as a span you can point at instead of a gap you hear about from a late ticket.
• Spot presence that looks gamed. Six automatic flags surface unusual presence patterns. Four (unbroken Available, no idle, robotic regularity, no activity variation) are the patterns a mouse jiggler or keep-alive script tends to leave; overnight presence and weekend uniformity mark unusual hours, which can just as easily be someone working late. Each flag shows how it's calculated, so it is a pattern to read, not proof. Standard includes the Idle suppression flag; all six run on Pro and Scale.
• Quick and light. One read-only Microsoft sign-in by an admin, about five minutes, and nothing installed on anyone's device. No screenshots, no keystrokes, and no message, chat, call, file, or calendar content.
• Bounded and scoped. Your data is scoped to your Microsoft tenant and processed on US infrastructure, with bounded retention and per-user and workspace-wide deletion built in. Guests are never tracked, and people whose Microsoft 365 location is in the EEA, the UK, Switzerland, or Canada are excluded automatically.

Coverage is one question the history answers. The one most managers start with is simpler: who's online, when, and how much. See how Presify works, read Presify for remote and hybrid teams, or start with one Microsoft sign-in.