Craft · Jun 25, 2026 · 10 min read · by the Pressfold team

Visualising data so journalists can actually use it

You can have the best dataset and the sharpest angle in the country, and still lose the story to a bad chart. When a reporter opens your pitch, the visual is often the first thing they judge, and a confusing or over-designed graphic tells them the work behind it is probably just as muddled. The opposite is also true: a clean, self-explanatory chart can carry a story across the line on its own, because it gives the journalist something they can publish immediately. Visualising data for the press is not about making things look impressive. It is about making the finding so clear that a busy editor understands it in a glance and trusts it enough to run.

This article is about the craft of building charts and assets that reporters actually reuse. I will cover what makes a graphic publishable rather than just pretty, which formats survive the trip from your studio to the printed or online page, and the specific mistakes that get otherwise good data spiked. The standard to aim for is simple: a reader who never sees your pitch should still understand the chart.

The journalist is your real audience

The most common mistake in data visualisation for PR is designing for the wrong person. Teams build charts to impress their client or their boss, loading them with brand colours, logos, and decorative flourishes. But the person who decides whether your chart gets used is a reporter on a deadline, and the person who has to understand it is their reader. Neither of them cares about your brand guidelines. They care whether the point lands instantly.

This reframing changes everything about how you design. A reporter needs a chart they can drop into a page with minimal editing, that reads clearly at small sizes and in black and white, and that does not need a paragraph of explanation to make sense. A reader needs to grasp the single main message before they consciously decide to keep reading. If your graphic fails either test, it does not matter how polished it looks. Design backwards from those two people and most of your decisions make themselves.

It is worth understanding why this matters so much in practice. A reporter on deadline is making a fast cost-benefit calculation on every asset that lands in their inbox. If your chart needs reworking before it can run, it competes against the dozen other stories they could cover with less effort, and effort usually wins. A chart that is genuinely ready to publish removes that competition entirely. You are not just illustrating your story; you are lowering the price of running it below everything else on their desk. That is the real job of a press graphic, and it is why polish aimed at impressing a client so often produces a chart no newsroom will touch.

What makes a chart publishable

A publishable chart does one job and does it without help. Before you choose a format, write the single sentence the chart is meant to prove. If you cannot, the problem is the finding, not the design. Once you have that sentence, build the simplest possible visual that demonstrates it and resist adding anything that does not serve it.

The qualities that separate a chart a journalist reuses from one they ignore:

• One message per chart. If a graphic is trying to show three things, it shows none of them. Split it. A reporter would rather have three clear charts than one busy one.
• A title that states the finding. Not "Sales by region" but "The north spends twice as much as the south." The title should be the headline, doing the interpretive work for a reader who only glances.
• Labels on the chart, not in a legend. Forcing the eye to bounce between a legend and the data is friction. Label lines and bars directly wherever you can.
• Honest axes. Start bar charts at zero. Do not truncate scales to exaggerate a difference. A reporter who spots a misleading axis will distrust the whole story.
• Readable at small sizes. Most coverage appears on phones. If the chart is illegible at thumbnail size, it is illegible where it counts.

Behind every good chart is a finding worth charting, which is why visualisation work cannot be separated from how the story was built in the first place. The clearer your core finding, the easier the chart, and the relationship runs both ways, as covered in the anatomy of a data story.

Choosing the right format

Most data stories need one of a small handful of chart types, and matching the type to the data is half the battle. Reaching for an unusual format to look distinctive usually backfires, because novelty costs the reader comprehension. The dependable choices:

• Bar charts for comparing categories. They are the workhorse for a reason: people read length effortlessly. Use them whenever you are ranking or comparing discrete things.
• Line charts for change over time. Nothing communicates a trend faster. Keep the number of lines low, because more than four or five becomes a tangle.
• Maps for geographic patterns, but only when location is genuinely the point. A map that could have been a bar chart wastes the reader's effort decoding shapes.
• Simple tables when exact figures matter more than the shape of the data. Reporters often prefer a clean table they can quote precisely.

Avoid the formats that flatter the designer and confuse everyone else. Pie charts with more than a few slices are hard to read; stacked area charts hide the individual trends inside them; dual-axis charts invite misleading comparisons and editors distrust them on sight. Three-dimensional effects, gradients, and decorative backgrounds add nothing but noise. When in doubt, choose the plainer option. The plainest honest chart that proves your sentence is almost always the right one.

One nuance is worth holding onto: the right format also depends on the kind of outlet you are pitching. A national newspaper's data desk may want the raw numbers so they can rebuild a chart in their own style, while a trade publication or a regional site will often run your image exactly as supplied. Smart practice is to hand over both the finished graphic and the clean data, so the format decision belongs to the reporter rather than to you. You are not trying to control how the story looks. You are trying to make every reasonable choice they might make easy, which is a more generous and far more effective goal.

Building the asset pack a newsroom wants

A single chart is rarely enough. Reporters work in different formats and to different deadlines, so the teams that get reused hand over a small pack that covers the likely needs without forcing anyone to ask. Done well, this pack is the difference between a reporter running your story today and parking it until they have time to rebuild your visuals, which usually means never.

A practical asset pack tends to include:

• A high-resolution image of each chart, in a standard format, sized for both web and print, with the source and date baked into the image so it stands alone if forwarded.
• The underlying data in a clean spreadsheet, so a data journalist can rebuild the chart in their own house style if they prefer. Many serious outlets will, and refusing to share the numbers reads as something to hide.
• A short, plain caption for each chart that a reporter can lift directly, stating the finding and the source in one or two sentences.
• A clear source line naming the data, the publisher, and the period, identical across every asset so attribution is never ambiguous.

The principle is to remove every reason a reporter might have to do extra work or come back with a question. The data file should be tidy and labelled in plain language, the same standard you would apply when working from external sources, which connects directly to the sourcing discipline in public datasets that make great PR stories. Friction is the enemy of coverage, and an asset pack exists to eliminate it.

One detail that earns disproportionate goodwill is anticipating reuse beyond the single article. A chart that carries its own source line travels safely when an editor crops it for social, embeds it in a newsletter, or hands it to a sister title. The version without a baked-in source becomes orphaned the moment it leaves the original page, and an orphaned chart is one a careful editor will decline to run at all. Spending two minutes making each asset self-contained means your finding keeps its attribution wherever it ends up, which protects both the reporter and your own credit. It is a small courtesy that quietly multiplies how far one study can spread.

The mistakes that get charts spiked

Some errors are fatal not because they look bad but because they break trust. A misleading axis is the classic example. Truncating a scale to make a small change look dramatic might pass internal review, but a sharp editor will catch it, and once they catch one manipulation they will assume the whole study is spun. Honesty in the visual is not a nicety; it is what keeps your data usable.

Other recurring failures are quieter but just as costly. Charts overloaded with data points so the main signal drowns. Colour schemes that fall apart for colour-blind readers or in greyscale print. Tiny fonts that vanish on a phone. Branding so heavy the chart reads as an advertisement, which makes editors reluctant to run it at all. And the subtle one: a chart that technically shows the data but does not make the point, leaving the reporter to figure out why you sent it. Each of these is avoidable with a single habit, which is to show your draft chart to someone outside the project and watch whether they understand it without a word from you.

That test is more powerful than any design checklist because it catches the failures you have gone blind to. After hours inside a dataset, you know what the chart means before you even look at it, so you cannot judge whether it communicates. A colleague who has never seen the data has no such advantage, and their first reaction is the closest thing you have to a reporter's. If they pause, squint, or ask what they are looking at, you have found a problem while it is still cheap to fix. If they read the headline number out loud without prompting, you are done. Build that thirty-second test into your process and you will ship far fewer charts that get quietly ignored, which is the most common fate of a press graphic and the hardest one to diagnose, because nobody ever writes back to tell you your chart was confusing.

Clarity is the whole job

If you take one idea from this, let it be that clarity beats polish every time. The temptation in data visualisation is always to add: another colour, another data series, another flourish to justify the effort. The discipline is to subtract until only the finding remains. The chart a reporter reuses is almost never the most beautiful one in the pack. It is the one whose meaning is impossible to miss.

Build for the reporter and the reader, prove a single sentence, choose the plainest honest format, and hand over a pack that leaves no question unanswered. Do that consistently and your visuals stop being something you attach and start being the reason your stories get picked up. The numbers may earn the coverage, but the chart is what lets a journalist say yes without a second thought, and that yes is the entire point of the work.