Quick answer
Choose a chart that makes your question easier to answer. A trend, a channel comparison and a conversion journey need different views. If people need three minutes of explanation before they can read the dashboard, simplify it.
Choose the chart from the question
What should someone understand after looking at this chart? Start there. Adding another visual to a crowded dashboard won’t help if nobody knows which comparison matters or what they’re supposed to do about it.
| Question | Useful form | Watch for |
|---|---|---|
| What changed over time? | Line chart | Inconsistent periods and incomplete data |
| Which category is larger? | Bar chart | A misleading value-axis baseline |
| Where do people stop? | Funnel or stage table | Different populations at each stage |
| How do cohorts compare? | Cohort table or lines | Unequal follow-up time |
| How are values distributed? | Histogram or box plot | Averages hiding variation |
The UK Government Analysis Function’s chart guidance covers choosing and presenting common chart types. For a few exact values, a well-labeled table may be clearer than a graphic.
Name the denominator
Conversion rate needs an explicit base: sessions, unique visitors, inquiries or eligible accounts. Two charts using different denominators should not share an unqualified conversion label.
For example, 50 inquiries from 1,000 measured visits is a 5% inquiry rate for those visits. It is not a customer acquisition rate. State the time window and relevant exclusions beside the figure, and show counts when a small sample could make a percentage unstable.
Make fair comparisons
Compare complete periods or label partial ones. Allow enough time for later outcomes, particularly in long sales cycles. A recent campaign may appear weak simply because its opportunities have not matured.
Use a zero baseline for bars when bar length represents magnitude. If a line chart uses a narrower axis to show variation, make the scale clear. Avoid decorative effects that change apparent area or make values harder to compare.
Use color with a purpose
Reserve emphasis for the series or finding that matters. Label lines directly where practical and do not rely on color alone to distinguish categories. Provide a text account of the finding and accessible data when readers need the underlying values.
Finish the chart with the decision it supports: investigate a source, adjust a test or wait for more data. Our marketing analytics guide explains how to connect reporting with a regular decision-making routine.
A clear chart still needs the right denominator, whether it reports search visibility or the quality of purchased impressions. See our guides to SEO performance measurement and display advertising quality.
