How to Create Infographics with Data: A Survey Walkthrough

You have survey results worth sharing, and a 40-slide deck will not reach the people who need them. An infographic can, because it puts one finding in front of a client, customer or colleague in under a minute. Most of the work is editing, not drawing. This article walks through the method, and it builds on our guide to charting survey data, which covers each chart type in more depth.

Key takeaways

• To create an infographic with survey data, pick one message first, then choose three to five numbers that prove it and match a simple chart to each.
• Write the message as a finding a reader could repeat rather than a topic, and show the baseline so a change means something.
• Use the simplest chart for each number, keep every bar axis at zero, and use one large figure for a single result that needs no chart.
• Label every base, add the question wording, fieldwork dates and source, and check that group percentages rebuild the headline total before you publish.
• Make it accessible with direct value labels, text contrast of at least 4.5 to 1, and alt text plus a text version of the figures.

How do you create an infographic with survey data?

Work in this order, because each step narrows the next one.

1. Pick one message.
2. Choose three to five numbers that prove it.
3. Match a chart to each number.
4. Lay the page out so the message reads first.
5. Add the base, the source and the dates.
6. Make it accessible.
7. Review it against a checklist.

To keep this concrete, we will follow one illustrative example. Imagine a meal-kit brand that surveyed 1,000 US adults this year, and 1,000 last year, about how they plan weeknight dinners. The numbers below are made up to show the method. They are not real survey data.

How do you pick the one message?

State the finding as a sentence a reader could repeat. “Most people decide what to cook on the day” works. “Results of our cooking survey” does not, because it names a topic and says nothing about it.

A good message connects to something the audience already cares about. Technical data takes more effort, so ask what changed, who differs, or what surprised you. If you are collecting new data, write the questions around the message from the start.

Your story should strengthen the message, not bend the data. If the message is a rise in satisfaction over two quarters, show the earlier quarters too, even when they were low. The baseline is what makes the rise meaningful.

In the example, the message is: “Six in ten people decide what to cook on the day, and it is rising.”

Which three to five numbers should you show?

Pick the figures that prove the message, then cut the rest. Three to five is a practical limit, because each number needs room to be read.

A useful test is to ask what each number answers: how big, how different, how much has it changed, why. In the example:

• How big: 62% decide on the day itself.
• How different: the share varies sharply by age group.
• What changed: it was 55% a year ago.
• Why: the main reasons given for deciding late.

Before you commit to a comparison, check the sample behind it. Each group you show needs enough respondents for its percentage to be steady. With 300 people in a group, a result near 74% carries a margin of error of about 5 points at 95% confidence, assuming a simple random sample. That is fine for a broad gap, but too loose for a gap of two points.

Which chart fits each number?

Choose the simplest chart that shows the comparison without distortion. Our guide to charting survey data covers the options in full.

Number	Question it answers	Chart
62%, up from 55%	How big, and what changed	Large figure with the prior value beside it
74% / 63% / 49% by age	How different	Bar chart, axis from zero
Reasons for deciding late	Why	Sorted bar chart

• A single yes/no result rarely needs a chart. One large number carries it. If you do draw the split, a pie chart works for two or three slices that add to 100%.
• Two points in time need only two values. Use a line once you have three or more.
• Use a line graph for a longer trend, a map when geography is the point, and a stacked bar chart when you need parts of a whole across several groups.
• Keep every bar axis at zero, so bar length matches the value.

How should you lay out the infographic?

Put the message at the top as a headline, then the biggest number, then the supporting charts in the order the story unfolds. Readers scan before they read, so hierarchy does the persuading. Give each section one question and one chart. A page split down the middle suits before-and-after comparisons.

Keep the design consistent. Use one font family, one line width and one palette across all charts, and let one accent color mark the number you want noticed. Leave white space around each chart, and choose clear over clever.

What sourcing and bases should it carry?

Every infographic needs a footer that lets a skeptical reader check it. Include:

• The base: who was asked and how many. For the example, “1,000 US adults, 18 and over”.
• The question wording for the headline figure, or a link to it.
• The fieldwork dates and the source.
• Sub-bases where they differ. The late-decision reasons come only from the 620 people who decide on the day (62% of 1,000), so label that chart “Base: 620”.

Watch the sums. If respondents could pick more than one reason, the bars can total over 100%, so say “multiple answers allowed”. In the example, 41% of those 620 cite schedule changes, 33% cite nothing in the house and 29% cite wanting something different. Those add to 103%, which is fine here, and is why the note matters.

For the age chart, check that the groups reconcile to the total. If 18 to 34 is 30% of the sample at 74%, 35 to 54 is 38% at 63%, and 55 and over is 32% at 49%, the weighted result is 0.30 x 74 + 0.38 x 63 + 0.32 x 49 = 22.2 + 23.9 + 15.7 = 61.8, which rounds to the 62% headline. If your parts do not rebuild the total, find out why before you publish.

For the change from 55% to 62%, a seven-point gap between two samples of 1,000 is larger than the margin of about 4.3 points for that comparison, so it is a real shift rather than noise, again assuming simple random samples.

How do you make an infographic accessible?

Plan for readers who use screen readers, have low vision or are color blind, or view on a phone.

• Do not rely on color alone. Label bars directly with their values, so the chart works in grayscale.
• Check contrast. Body text needs at least a 4.5 to 1 contrast ratio against its background under WCAG 2.1 AA. Avoid light gray text.
• Add alt text and a text version. Put the message and key numbers in the image description, and publish the figures as plain text or a table beside the graphic.

What mistakes should you avoid?

• Starting with the chart. Picking a visual before the message gives you decoration. Write the headline sentence first.
• Showing everything. If a figure does not prove the message, cut it.
• Truncating the axis. A bar chart that starts at 40% makes small gaps look large.
• Hiding the base. A percentage without a sample size cannot be judged.
• Over-reading small groups. A difference between two tiny subgroups may be noise.

What is the review checklist?

Before you publish, check that:

• The headline states a finding, not a topic.
• There are no more than five numbers.
• Every bar axis starts at zero.
• Every chart shows its base, and the footer shows source, dates and question wording.
• Percentages and parts reconcile to totals.
• Fonts, colors and line widths are consistent.
• It reads at a glance on a phone and in grayscale.
• Alt text and a text version exist.
• A colleague can state the message back to you.

How do you do this in practice?

The slow work sits before the design: weighting, cutting results by group and checking which differences are real. Halo Reports provides crosstabs and banner tables with weighting and significance testing, plus charts and interactive report pages, so the numbers you place in an infographic come from one checked source. To start from what has changed, mTab Pulse monitors trackers and studies and surfaces meaningful movement.

Start with the sentence. Write your message, list the three to five numbers that prove it, and note the base beside each. When that page makes sense on its own, the design is the easy part. If the audience wants a one-page update rather than a graphic, an executive brief covers what changed, why it matters and what to do.