By the end of this module, participants will be able to:
- understand the difference between operational dashboards and strategic dashboards
- recognize what different internal profiles need to see
- organize a visualization using layers of interpretation
- identify signals of a poorly designed dashboard
From generic panels to useful dashboards
At a basic level it is common to create dashboards that attempt to show everything to everyone. The result is usually a screen full of indicators that nobody uses consistently.
At a more mature level the starting question changes:
Who needs to see what, and in order to make which decision?
Dashboards by role
Not all roles need the same information. For example:
- a section editor may need daily monitoring of pieces, recirculation, and engagement
- an audience manager may need acquisition, recurrence, and channel behavior
- a business profile may need conversion, retention, and contribution by product
The usefulness of a dashboard depends on its ability to adapt to the context of use. The reason is not simply that different roles care about different topics — it is that they make decisions on different timescales and act on different signals. A section editor is asking “what should I commission tomorrow?” and needs daily, piece-level data. An audience manager is asking “is our subscriber base growing or eroding?” and needs weekly trend data. A CEO is asking “is this business sustainable?” and needs monthly cohort and revenue data. When these three people receive the same dashboard, at least two of them are looking at information that does not match the cadence or granularity of their actual decisions.
Reading hierarchy
A useful dashboard is usually organized in layers:
- main indicators
- explanation or breakdown
- the possibility to explore segments, channels, or content in more depth
This hierarchy allows for quick interpretation while also enabling deeper exploration when necessary.
In practice, the first layer might show five numbers: total subscribers this week, churn rate, conversion rate, the top-performing article, and newsletter open rate. The second layer breaks these down — which editorial sections drove subscriptions, which user segments are showing elevated churn, which traffic sources produce the most engaged readers. The third layer allows teams to investigate a specific cohort, article, or channel in detail. The discipline in designing this hierarchy lies in resisting the temptation to promote everything to the first layer. When a dashboard shows twenty equally prominent metrics, the result is that nothing feels urgent — and nothing gets acted on.
What a dashboard should not do
A poorly designed dashboard often presents some of the following problems:
- too many metrics without prioritization
- absence of time context
- mixing incompatible indicators
- visualizations that are difficult to read
- inability to explore the data further
The most common and damaging problem is the first one: too many metrics without prioritisation. When a dashboard places forty indicators on the same screen with equal visual weight, the implicit message is that all of them matter equally. In practice, this means nothing is prioritised, reviews take longer, and teams tend to gravitate toward the metrics they already understand rather than the ones that would actually inform a decision. A focused dashboard that shows eight metrics clearly and consistently is almost always more useful than a comprehensive one that tries to show everything at once.
From reporting to conversation
A mature dashboard is not only a place where numbers are observed. It is a tool for better conversations within the organization. It helps align teams, detect opportunities, and frame decisions more clearly.
Consider the difference between two weekly reviews. In the first, a section editor and an audience manager look at a shared dashboard and one says: “Scroll depth on long-form pieces is high, but recirculation is almost zero — readers are finishing the article and leaving.” In the second, the same team looks at a single traffic number and concludes that things are going well. The first conversation is only possible because the dashboard surfaces a specific tension that neither person would have raised on their own. That is the difference between a reporting tool and a conversation tool.
A good dashboard does not show everything: it shows what is necessary to make better decisions.
Try it yourself
Your newsroom sends one weekly performance report to the entire team. It contains these 18 metrics, listed in no particular order:
Page views · Unique users · New users · Returning users · Sessions · Avg. session duration · Bounce rate · Scroll depth · Top 10 articles by traffic · Traffic by source · Mobile vs. desktop split · Newsletter open rate · Newsletter click rate · Social reach · Paywall exposures · New subscriptions · Subscription cancellations · Revenue from subscriptions
Three people receive this exact same report:
- The section editor — makes daily decisions about what to publish, when, and how to promote it
- The audience development lead — responsible for growing and retaining the subscriber base
- The CEO — reviews overall business health once a week
Consider:
- Select 4–5 metrics from the list for each person. Which ones genuinely help them make their specific decisions — and which are just noise for their role?
- Both the editor and the audience lead care about “returning users” — but for different reasons. What does each of them actually need to know about that metric?
- Which metrics on the list are operational (day-to-day) and which are strategic (trends and direction)? Does the current report make that distinction at all?
- If you could add one metric that isn’t on the list — specific to just one of the three roles — what would it be and why?
One dashboard for everyone usually means a useful dashboard for no one.