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

The anatomy of a data story journalists actually publish

Most pitches that land in a journalist's inbox are dead on arrival, and not because the topic is wrong. They fail because they are built backwards. The sender starts with what they want — a link, a brand mention, a campaign result — and works outward from there, hoping the reporter will excavate something useful from the pile. Reporters do not excavate. They scan, and they move on. A data story that earns coverage is assembled in the opposite direction: it starts with what a reader would find genuinely surprising, then surrounds that finding with just enough structure to make it publishable without extra work on the journalist's part.

Over years of putting research-led stories in front of national and trade press, the same skeleton keeps showing up under the pieces that get picked up. Five parts, in roughly this order: the finding, the hook, the method, the visual, and the quote. Get all five right and a reporter can publish your story almost as it arrives. Miss one and you create friction — and friction is what kills coverage faster than anything.

The finding comes first, and it has to be one sentence

Every data story lives or dies on a single sentence a reader could repeat to a colleague. Not a theme, not a topic area, not "insights into consumer behaviour." A specific, falsifiable claim with a number attached. "Renters in their thirties now spend more on subscriptions than on home contents insurance" is a finding. "We surveyed UK renters about their spending habits" is a press release. The difference is whether someone who hears it raises an eyebrow.

The discipline here is brutal. Before you write a word of outreach, you should be able to state the finding in one line and feel slightly nervous that it might be too strong. If it makes you shrug, it will make a reporter delete the email. A useful test: would this sentence survive being read aloud at a dinner table among people who have never heard of your client? If the table goes quiet for a second, you have a story. If someone politely changes the subject, you do not.

The finding also has to be the thing your data can actually support. This is where many campaigns go wrong — the headline finding is dramatic, but the underlying numbers were never built to carry it. A good practice is to identify the strongest defensible claim in the dataset and let that become the headline, rather than deciding the headline you want and then torturing the data toward it. We cover the question-by-question version of this discipline in survey design for PR that holds up to scrutiny, because a weak instrument cannot produce a strong finding no matter how it is framed.

The hook is why it matters now, to these readers

A finding tells a reporter what is true. The hook tells them why their audience should care today. These are not the same thing, and conflating them is one of the most common reasons interesting data goes nowhere. You can have a genuinely novel statistic that no outlet will run because there is no reason it belongs in this week's paper rather than any other week's.

The hook usually answers one of a handful of questions. Is this number changing — up sharply, down sharply, reversing a long trend? Does it contradict something everyone assumes is true? Does it attach to a moment already in the news cycle? Does it expose a gap between what people say and what they do? Strong data stories tend to ride one of these. The flat, hookless story — "here is an interesting fact about an unchanging situation" — is the hardest to place, because the reporter has to invent the relevance themselves, and they will not.

The most reliable hook is contrast. Something is more than you would expect, or less, or moving in the opposite direction to the conventional wisdom. The job of the pitch is to make that contrast impossible to miss in the first two sentences. If a journalist has to read to paragraph three to understand why the number is interesting, the story has already lost.

It helps to think of the hook as the answer to a reporter's silent question: "So what?" A finding can be perfectly true and perfectly dull. The hook is the layer that converts a fact into a consequence — what changes because this is true, who is affected, what it predicts. When a campaign struggles to place an objectively interesting statistic, the problem is almost always a missing or weak hook, not a weak finding. The fix is rarely more data; it is finding the angle on the data you already have that makes a reader sit up. The same dataset can support a flat story and a sharp one depending entirely on which contrast you choose to lead with.

The method is what makes it trustworthy enough to print

Here is the part amateurs skip and professionals never do. A finding without a method is a rumour. The moment a reporter takes your number seriously, the next thing they need is the ability to defend it to an editor, and increasingly to a sceptical readership. That means they need to know who you asked, how many of them, when, and how the data was gathered. Not buried in a footnote — available, clean, and honest.

A publishable method statement is short and unflinching. Sample size and composition. Fieldwork dates. The source of the data, whether that is a commissioned survey, an internal dataset, a public records request, or analysis of an existing open dataset. If the sample skews in a way that matters, say so. Reporters do not expect perfection; they expect transparency, and they have been burned often enough by inflated claims that they treat a clean method note as a signal of trustworthiness in itself.

The single fastest way to lose a serious outlet is to dress up something flimsy as something rigorous. A poll of 200 self-selected newsletter subscribers is fine for an illustrative blog point and fatal if you present it as a representative national finding. Honesty about limitations does not weaken the story — it is the thing that lets a good reporter say yes. The stronger the method, the bigger the outlet that can afford to run it.

The visual lets a reporter publish without doing your work

A data story is a visual story whether you provide the visual or not. If you do not, the reporter either makes one themselves — which is friction, and friction loses — or runs the piece without one, which buries it. If you do provide one, and it is clean and accurate and properly labelled, you have removed an entire production step from their day. That is a gift, and gifts get remembered.

The visual should do one job: make the finding obvious at a glance. A single chart that shows the contrast at the heart of the story beats a deck of twelve. The temptation is always to show everything you found; resist it. The reporter is not running your research, they are running one story, and the visual that supports that one story is the only one that matters. The rest belongs in a supporting data table they can dig into if they want depth.

Practical details decide whether the visual gets used. Is it legible at the size it will actually appear? Are the axes labelled? Is the source credited on the chart itself, so it travels intact when the image is shared? Will it survive being dropped into a content management system by someone in a hurry? The principles of building charts a journalist can use immediately deserve their own treatment, and we go deeper in visualising data for journalists.

The quote gives the story a human voice

Numbers establish that something is true. A quote tells the reader why it matters and gives the piece a named, accountable source. Reporters need a quote because a story with only statistics reads like a spreadsheet, and because editors expect attribution. The quote you supply should not restate the number — the chart already did that — it should interpret it. What does this finding mean for an ordinary person? What is likely to happen next? What should someone do differently because of it?

The best supplied quotes sound like a person talking, not a brand issuing a statement. "We're seeing households make trade-offs they wouldn't have considered two years ago" is usable. "Our innovative solutions empower customers to optimise their financial wellness journey" is unusable, and a reporter will either rewrite it or, more likely, drop the story. The quote is also where the named expert earns their place; it should add genuine insight, not corporate throat-clearing.

One underrated function of the quote is to pre-empt the reporter's next call. A story with only numbers leaves the journalist needing a human source, and if you have not supplied a credible one they may go and find their own — possibly a critic of your client. A well-chosen quote from a named expert with a real title closes that gap. It should be attributable to a person whose expertise is plausible on the subject, not a marketing manager opining outside their lane. Reporters quietly downgrade quotes that come from a job title with no obvious connection to the data, because their readers will too.

Keep the quote short enough to use whole. A two-line quote that makes one sharp point survives editing intact; a six-line paragraph gets carved up, and the carving rarely flatters you. Offer one strong line rather than a buffet, and if you want to give the reporter options, supply two or three distinct short quotes rather than one sprawling one. The aim is a sentence that could be lifted directly into the piece without a single edit.

How the five parts fit together

The order matters because it mirrors how a reporter reads. The finding hooks their attention. The hook tells them it belongs in this week's coverage. The method lets them trust it enough to put their byline on it. The visual lets them publish it without extra labour. The quote gives them the human line that turns data into a story their readers will actually finish. Remove any one and you reintroduce work or doubt, and either is usually fatal.

This is also why the structure is unforgiving of shortcuts. You cannot compensate for a weak finding with a beautiful chart, and you cannot rescue a flimsy method with a punchy quote. Each part carries a distinct load. The most common failure is over-investing in the parts that feel like marketing — the visual, the quote — while under-investing in the parts that feel like work, the finding and the method. Reporters notice immediately, because the parts you skimped on are exactly the ones they are professionally trained to check.

The payoff for getting all five right is that you stop pitching and start supplying. A complete data story is not a request for a favour; it is a finished product a busy newsroom can use. That shift — from asking to providing — is what separates campaigns that earn coverage from campaigns that beg for it. Build the skeleton properly, and the story does most of the persuading for you before anyone has to be convinced of anything at all.