Visualizing is not decorating
A visualization is not meant to beautify data but to make it understandable. The main criterion is not aesthetic but functional: the chart should allow someone to understand something faster, compare more easily, or detect a pattern that would be less visible in a table.
In a newsroom, this matters greatly. Decisions are often made quickly and with limited time. A poor visualization can hide the important conclusion; a good visualization makes it obvious.
Which tool to use
Excel remains a useful tool for organizing, cleaning, and exploring basic data. For many simple tasks, it is sufficient.
BI or dashboard tools allow a more structured approach when metrics need to be monitored regularly, periods compared, or information shared between teams.
Visualization tools oriented toward publishing or communication—such as certain chart builders—can be especially useful when narrative clarity or ease of use is required.
The ideal tool is not the most sophisticated one, but the one that solves a specific need well.
Principles of good visualization
A good visualization usually respects five principles:
Clarity: it is quickly understood
Simplicity: unnecessary decoration is removed
Comparability: it makes categories, periods, or segments easy to compare
Visual hierarchy: it highlights what is important
Context: it provides the necessary framework to interpret the figure
A number without context can be ambiguous. A chart without clear labels can be as well.
Common mistakes
Among the most common errors are:
• using too many colors or decorative elements
• truncating axes in confusing ways
• choosing inappropriate charts for the type of comparison
• mixing too many variables in a single visualization
• creating charts that are visually striking but cognitively difficult
A very common mistake is using pie charts for complex comparisons. They perform poorly when there are many categories or small differences. In most of these cases, a bar chart communicates better.
Choosing the right chart
If the goal is to compare categories, a bar chart usually works best.
If the goal is to show change over time, a line chart is usually better.
If the goal is to observe the relationship between two variables, a scatter plot is appropriate.
If exact precision is more important than visual impact, a table may be the best option.
Learning to choose the appropriate chart does not require mastering advanced tools. It requires understanding which question you want to answer.
Try it yourself
Your team needs to present monthly performance to the editorial board. A colleague has prepared four charts. Read the descriptions and assess each one:
Chart A — A pie chart with 9 slices showing traffic by source: Search, Direct, Newsletter, Facebook, Instagram, Twitter/X, Other social, Referral, Internal. The two smallest slices (Instagram and Twitter/X) are each under 2%.
Chart B — A bar chart showing page views per month for the last 12 months. The Y-axis starts at 850,000 instead of 0. All bars look roughly the same height.
Chart C — A line chart comparing new vs. returning users over 6 months. Two clearly labelled lines. Y-axis starts at 0.
Chart D — A table showing conversion rate, average reading time, and scroll depth for each of your 15 editorial sections, sorted alphabetically.
Consider:
- Which chart could actively mislead the reader — and how?
- Which chart would you replace, and with what?
- Chart D is technically accurate. When is a table the right choice — and when would a chart serve better?
- If you had 90 seconds to brief your editor on performance, which single chart would you show — and why?
Good visualisation isn’t about making data look impressive. It’s about making the conclusion impossible to miss.