
This article is part of the guide Project dashboards: strategic, analytical, operational, which covers the three types and how to choose between them.
The quarter past nine meeting, six people standing. Without an operational project dashboard the first twenty minutes go on establishing what is actually late, who is working on what and whether the thing that blocked everyone yesterday is still blocking them. With one, the same meeting takes ten minutes and produces decisions instead of a status update.
That is the whole value proposition of the operational project dashboard, and it explains its two hard constraints: one screen, thirty seconds, read standing up. Anything that does not survive those constraints belongs on a different page.
In short
- It answers what do I do today. Its readers are the team and whoever coordinates it, and it is read standing up, in about thirty seconds.
- Five reports: late tasks, workload this week, what is blocked, open issues and changes waiting for a decision. Every row carries a name.
- The test for what goes in: can it change what someone does today or tomorrow? If not, it belongs in the analytical dashboard.
- The part almost nobody designs is the return path: actual dates and remaining work flowing back into the plan at the end of the day.
- A read-only operational dashboard leaves data collection on the project manager, and the plan still gets updated in a rush the day before the meeting.
Who reads it, and how
The readers are the people doing the work and the people coordinating them: team leads, a project manager running the day rather than the reporting, sometimes a supplier with access to their own slice. They read it several times a day, never for long, and usually on their feet or between two other things.
Two consequences follow, and they drive every design choice on this page. The first: it has to fit on one screen, without scrolling, because a second screen is a second decision to keep reading and it will not be taken. The second: it is read by exception. What is going fine gets no comment and no space; the page exists to surface what is not.
There is a third, less obvious one. Every row needs a name attached. A late task with no owner is a fact nobody can act on, and it will sit there red for three weeks precisely because it belongs to everybody.
The five reports
- Late tasks, with the person responsible. Sorted by how late, not by project. It is the first thing on the page and usually the only thing the ten-minute meeting needs.
- Workload for this week. Per person, this week and the next. This is where you catch the overload before it turns into a delay, which is the only cheap moment to catch it.
- What is blocked right now. Not what was blocked, what is blocked, with who is waiting and on what. Blockers age badly and this is the report that makes the ageing visible.
- Open issues, and for how many days. The number of days matters more than the issue count: five issues open for two days is a busy week, two issues open for five weeks is a problem nobody owns.
- Changes waiting for a decision. Requests parked pending an answer. They are invisible everywhere else and they are a common reason a team is quietly stalled.
Five is not a rule of thumb, it is what fits on one screen while staying readable from a metre away. If a sixth report genuinely earns its place, something else comes off.
The team’s day, in three moments
The page gets used at three distinct points, and only the first two usually get designed.
Morning. Ten minutes standing up: the late tasks at the top, who is overloaded this week, what stayed blocked from yesterday. Read by exception, decisions taken out loud, nobody narrating their week.
During the day. Alerts on whatever blocks a task, deadlines with nobody on them, supplies that are not arriving. This part needs a written rule about what is allowed to raise an alert. Without one everything lands there within a month, and then nobody looks at any of it.
End of day, and this is the one that matters. Actual start and finish dates, estimated remaining work, two seconds per person. This is the return path, and it is what keeps the other two moments alive: the plan updates from the people doing the work, rather than from the project manager chasing them.

If the dashboard is read-only, collecting the data stays on the project manager, and the plan gets updated in a rush the day before the meeting.
This is worth being blunt about, because it is where most operational dashboards fail. A page that only displays is a page that somebody has to feed, and that somebody is always the project manager. Two seconds per person per day is a tiny cost distributed across the team; two hours the day before the meeting is a large cost concentrated on one person, and it is the reason plans stop being current.
What goes in, and what does not
The test is a single question: can this change what someone does today or tomorrow? Everything that passes goes in. Everything that fails is interesting, possibly important, and belongs on the analytical page.

The request always comes about a month in, and it always sounds harmless: while we are at it, let us add the cost trend. Then the performance indices. Then a comparison with the other project. Each one is defensible on its own, and together they turn a thirty-second page into a two-minute one, which in practice means a page the team stops opening.
The mistakes I see most often
- Rows with no owner. A late task belonging to everybody belongs to nobody, and it will still be there in three weeks.
- Read-only. Data collection stays on the project manager and the plan goes stale between meetings. The return path is not a nice-to-have.
- No entry rule for alerts. Within a month everything raises one, and the team stops reading all of them.
- Two screens. The second screen is not read. If it does not fit, something has to come off rather than move down.
- Analytical content creeping in, one reasonable request at a time, until the page takes two minutes to read.
- Using it instead of talking. The page makes the ten-minute meeting possible; it does not replace telling someone directly that you are stuck.
When you do not need one
A team of four or five sitting in the same room, on one project, does not need an operational dashboard. They already have the fastest possible version of it, which is talking to each other, and the plan’s task list covers the rest. Building a page for them adds a maintenance job and removes nothing.
It starts to pay when the team is distributed, when people work across more than one project, or when the coordination has to survive somebody being away. Those are the conditions where a shared page beats a conversation, and not before.
Frequently asked questions
What should an operational project dashboard contain?
Five reports: late tasks with the person responsible, workload per person for this week and the next, what is blocked right now, open issues with how many days they have been open, and change requests waiting for a decision. Everything fits on one screen, and every row carries a name.
How often should it be updated?
Daily at minimum, continuously where the data allows it. It is the only one of the three dashboards where yesterday’s numbers are genuinely useless, because the decisions it supports are about today and tomorrow.
How do I decide whether something belongs on the operational page?
Ask whether it can change what someone does today or tomorrow. Late tasks, workload and blockers pass. Cost trends, performance indices and comparisons between projects do not: they are useful, and they belong on the analytical page where somebody has half an hour to read them.
Should the team be able to enter data into the dashboard?
Yes, and it is the part that decides whether the dashboard survives. Two seconds per person at the end of the day, entering actual dates and remaining work, keeps the plan current. A read-only page leaves all of that on the project manager, and then the plan gets updated in a rush the day before the meeting.
Does an operational dashboard replace the daily stand-up?
No. It shortens it, from twenty minutes of establishing the facts to ten minutes of deciding what to do about them. The conversation is where the decisions get made; the page is what stops the conversation being spent on finding out what is going on.
Further reading on this blog
- How to collect actuals from your team and update the plan
- Project dashboards: strategic, analytical, operational
- The analytical project dashboard: five reports and the drill-down
- The strategic project dashboard: what to put in it
- How to find the late tasks in Microsoft Project
- The tasks in trouble are the ones nobody was assigned
- How to update a Microsoft Project plan: baseline, status date and progress
Disclosure: this is the blog of a company that makes project-control software. Nothing above needs our tools: a shared board and a disciplined ten-minute meeting cover most of it.
The return path is the part that is hard to build by hand. GanttTimesheet emails each person their own assignments, imports what comes back, lets you approve it line by line and writes it into the plan, so the actual dates and remaining work arrive without anyone having to chase anyone.
#ProjectDashboards #TeamManagement #ProjectControl #MSProject #PMO