Almost every monday.com environment we come into has dashboards. Very few have dashboards that anyone in leadership actually opens. That is a shame, because a portfolio view that works does something concrete: it replaces the status meeting. Instead of spending an hour a week retelling where things stand, the time goes to what deviates.
The difference between a dashboard that gets used and one that dies rarely lies in the tool. It lies in how it is built.
A dashboard answers questions, it does not show data
The most common mistake is building from the data: here is everything we have, neatly laid out. Leadership does not have the question "what data is there?" It has questions like "are we delivering what we promised?", "where are we burning more than planned?" and "which decisions are waiting on me?".
So we always start from the other end: which three to five questions should this view answer, and for whom? Anything that does not help answer them goes. A good leadership view is almost always smaller than the first sketch.
The numbers need an owner
A status nobody is responsible for turns green on its own. For a portfolio view to be trustworthy, every number in it needs a clear owner and a clear rhythm: the project manager updates status and forecast at a set time each week, and the view pulls everything from there.
This is less a technical question than a question of how you work, and that is why a dashboard can never be introduced separately from the process it reflects. Build the view without the rhythm and you get a nice picture of stale data.
Build across teams, not per team
Individual teams often have excellent boards. The problem arises one level up: when status is called different things, dates are used differently and priority means different things, nothing can be summed up. The portfolio view forces a shared language: the same status scale, the same definition of done, the same way of assigning ownership.
Our advice is to set those conventions before the portfolio view is built, not after. It is a small job when done early and a big one when twenty boards have to be redone afterward.
The most common mistakes
- Too many widgets. A view with thirty charts is read by nobody. Fewer numbers, more clarity.
- Red with no way forward. A deviation without an owner and a next step just spoils the mood. Tie every deviation to an owner.
- Manual numbers next to automatic ones. As soon as one number is pasted in by hand, the whole view starts to be distrusted. Either the number comes from the system, or it is not included.
- One view for everyone. Leadership, project managers and teams need different views of the same data, not the same view.
How you can tell it works
The clearest proof is the calendar: status meetings get shorter or disappear, and the meetings that are held are about deviations and decisions. The second proof is that the questions change character, from "where do we stand?" to "what do we do about this?".
When we build portfolio views for customers, we always do it in tandem with the process: conventions first, then the views, and a walkthrough with the people who will use them. A dashboard is done when it survives its first month without anyone asking for the old report.