data-driven pr
Turning Your Year-End Data Into a Story Reporters Want
By Sunny Goyal · · 10 min read
To turn year-end data into a story reporters want, find one finding that says something new about your market, not your company. Then verify it, write a short method note, package it with a chart and a quotable expert, and pitch it to a few reporters who cover that trend before the December news window closes. The finding is the story. Your brand is the source.
Every December, inboxes fill with “our year in numbers” emails. Most of them are recaps: customers added, features shipped, offices opened. Reporters ignore them because a recap answers a question nobody asked. A data story answers a question readers already have: what changed this year, and what does it mean for next year?
Photo: Cht Gsml on Unsplash.
Why do reporters cover year-end data stories?
Reporters cover year-end data because December and early January are planning months for their readers, and editors need fresh, credible numbers to anchor trend pieces, outlooks, and “what to watch” roundups. Your data is useful to them when it is new, specific, and defensible.
The appetite is documented. In Cision’s 2026 State of the Media report, based on a survey of 1,899 journalists, original research topped the list of what journalists value most in pitches, alongside expert access and embargoed information. The same survey found 66% of journalists rely on PR-provided content for story ideas, and 72% say fewer than a quarter of the pitches they receive are relevant. Data is one of the few things that reliably clears the relevance bar, because it gives a reporter something to report rather than something to repeat.
Muck Rack’s State of Journalism 2026 points the same way. As summarized by The PR Net, journalists ranked clear relevance, credible sources, original data, and strong visuals as the most valuable pitch elements, and 78% said a pitch is relevant when it affects their audience or community, not just their coverage area.
The best-known example of yearly data as a media event is Spotify Wrapped. TechCrunch reported that Wrapped 2025 reached more than 200 million engaged users in its first 24 hours, compared with 62 hours to reach the same mark the year before, and was shared more than 500 million times. You do not need Spotify’s scale. You need the same discipline: a consistent dataset, a clear frame, and a date people expect.
What separates a data story from a company recap?
A data story reports a change in the world that your data can see. A recap reports a change in your company. Reporters write the first and delete the second.
| Company recap | Data story | |
|---|---|---|
| Subject | Your company | Your market, customers, or category |
| Headline shape | ”Company X had a record year" | "Mid-market CFOs cut software vendors by 18% in 2026” |
| Who cares | Employees, investors | Readers of the reporter’s beat |
| Evidence | Internal milestones | Aggregated, anonymized behavior with a method note |
| Reporter’s job | Rewrite your release | Interpret a finding, add context, call other sources |
| Shelf life | One day | Cited for months, often in AI answers |
The test is simple. Remove your company name from the headline. If the headline still interests the reporter’s readers, you have a data story. If it collapses, you have a recap.
Where does a good year-end finding come from?
Good findings usually come from data you already hold, viewed across your customer base rather than your own P&L. The point is to aggregate what many customers did, so the pattern reflects the market.
Common sources inside a company:
- Product usage. What customers did more or less of this year: transactions, searches, bookings, categories, features.
- Pricing and spend. Shifts in average order value, contract size, or budget mix, aggregated across accounts.
- Timing. When demand peaked, how long cycles took, and whether that moved versus last year.
- Geography. Which regions grew or shrank, which is gold for regional and national reporters alike.
- Surveys. A short survey of your customers or a panel, fielded in October or November, with a clear sample and dates.
Our guide to data-driven PR and original research covers how to design a survey that holds up. For year-end work, the faster route is usually behavioral data you already collect, because it is larger and harder to argue with.
How do you turn year-end data into a story, step by step?
The process below is the one we run with clients in October and November so the story is ready before reporters lock their December plans.
- Pick the question first. Write down three questions your buyers or the reporter’s readers are asking about this year. “Did hiring actually slow in Q3?” “Are shoppers trading down?” Choose the one your data can answer best.
- Pull and clean the data. Define the period, the sample, and the exclusions. Compare like with like: same customers, same months, same definitions year over year. Remove outliers you can explain and document the ones you cannot.
- Find the one finding. Look for the change that is surprising, large enough to matter, and easy to say in a sentence. Most datasets produce five interesting numbers. Pick one headline finding and two or three supporting ones.
- Write the method note. In 100 to 200 words, explain where the data came from, the sample size, the time frame, how you anonymized it, and what it cannot tell you. This note is what makes a reporter comfortable citing you.
- Package the assets. Build one clean chart per finding, a one-page summary, a spreadsheet of the underlying aggregates, and a named spokesperson who can explain the “why.” Publish the full report on your site with a stable URL.
- Pitch, then publish. Offer the finding to one or two reporters first, often under embargo, then release it broadly. Our journalist pitching guide covers the mechanics; keep the pitch under 200 words.
What should the method note include?
The method note should let a skeptical reporter judge your data in under a minute. If it cannot, they will either skip the story or call a competitor’s analyst instead.
Use this template:
About the data. This analysis is based on [anonymized, aggregated] [transaction / usage / survey] data from [number] [customers / users / respondents] in [countries or regions] between [start date] and [end date], compared with the same period in [prior year]. [Describe any weighting or exclusions.] Individual customer data was not shared. Figures are rounded. The data reflects [Company]‘s customer base and may not represent the full market.
That last sentence matters. Admitting the limits of your sample makes the finding more credible, not less. Reporters at outlets like Bloomberg and Axios are trained to ask about representativeness. Answering before they ask shortens the path to publication.
When should you pitch year-end data?
Pitch year-end data in November and the first two weeks of December, with a second window in the first week of January. Editors plan year-end and outlook packages weeks ahead, so a finding that arrives on December 22 has usually missed the slot.
| Window | What reporters are building | What to offer |
|---|---|---|
| Early to mid November | Holiday, budget, and planning stories | Early finding with an exclusive offer |
| Late November to mid December | Year-in-review and “what changed” pieces | Full report, chart, spokesperson |
| Late December | Skeleton crews, pre-written outlooks | Usually too late for new data |
| First week of January | Predictions and “year ahead” pieces | Data reframed as a forward signal |
If your data supports a forecast, pair it with a clear prediction. Reporters prefer predictions that are grounded in something, and a dataset is the best grounding you can offer.
A worked example (hypothetical)
This scenario is hypothetical and illustrates the process. It does not describe a Digitize client.
A Series B payroll software company serves about 4,000 small and mid-sized employers. In October, its comms lead asks the data team a single question: did small employers really slow hiring this year, as the headlines claim?
The team pulls anonymized, aggregated payroll counts for the same 3,100 employers active in both 2025 and 2026. The headline finding: net headcount at these employers grew 2.1% in 2026, versus 4.4% in 2025, but hourly roles grew faster than salaried ones for the first time in the dataset’s history. That second finding is the story. It is specific, it challenges a common narrative, and it is easy to say.
The comms lead writes a 150-word method note, builds two charts, and publishes a report page. In early November, she offers the finding to one small business reporter at a national outlet under a short embargo, with the CEO and an independent labor economist available for interviews. After the exclusive runs, she sends the report to a list of regional business editors with state-level breakdowns, each tailored to the editor’s state.
Result in the scenario: one national story, several regional pickups, and a report page that other writers cite for months. None of it mentions the company’s own growth, and all of it names the company as the source.
How does a year-end data story help AI search visibility?
A well-published data story gives AI engines a primary source to attribute, and earned coverage of that story gives them independent corroboration. Both raise the odds your brand is named when someone asks an assistant about the trend.
Muck Rack’s State of PR 2026 found that half of PR professionals are creating more authoritative or data-driven content to improve visibility in AI-generated responses, and 55% are securing coverage in high-authority publications for the same reason. The two work together. Your report page should state each finding in a plain sentence, include the date and sample, and use a stable URL that coverage can link to. Our guide to how tier-one press drives AI citations explains why the earned layer matters so much.
What do most teams get wrong?
Most teams start with the data they have and hope a story appears. The better approach starts with the question and only then asks whether the data can answer it.
- Leading with the brand. “Company X’s 2026 in review” is a blog post, not a pitch.
- Too many findings. Twelve stats in one email means the reporter has to find the story. Give them one.
- No method note. Without sample, dates, and limits, careful reporters will not cite the number.
- Cherry-picked comparisons. Comparing a strong month with a weak one, or a changing customer base year over year, is the fastest way to lose a reporter’s trust permanently.
- Missing the window. A finding sent the week before Christmas usually waits until it is stale.
- No owned home. If the data lives only in a PDF attachment, there is nothing to link and nothing for AI engines to cite. Put it on a page in your newsroom; our guide to building a newsroom page that gets cited covers the structure.
- Forgetting to reuse it. A strong dataset can feed bylines, sales decks, and investor updates for a quarter. Our guide to earned media amplification shows how.
Year-end data story checklist
- One question the reporter’s readers already care about
- Like-for-like comparison with documented exclusions
- One headline finding and two or three supporting ones
- Method note with sample, dates, anonymization, and limits
- One chart per finding, readable on a phone
- Report page with a stable URL and plain-sentence findings
- Spokesperson briefed on the “why,” plus an independent voice if possible
- Pitch sent before mid December, with an exclusive or embargo plan
- Regional cuts ready for local and trade reporters
- Plan to reuse the data in January outlook pitches and bylines
Frequently asked questions
How big does a dataset need to be for a reporter to use it?
There is no fixed threshold, but the sample has to be large enough that the finding is not noise and described clearly enough that a reporter can judge it. A few thousand customers or transactions with a consistent definition usually carries more weight than a small survey. Always state the sample size and time frame in the method note.
Can we use customer data in a press story?
Yes, if it is aggregated, anonymized, and permitted by your contracts and privacy policy. Check with legal before you publish. Never share data that could identify an individual customer, and never disclose a named customer’s figures without written permission.
Should we offer the data as an exclusive?
Often, yes. Offering one reporter first access under a short embargo increases the odds of a strong first story, and other outlets frequently follow. Pick the reporter whose recent work best fits the finding, set a clear deadline, and release the data broadly once the exclusive runs.
What if the data shows something unflattering about our market?
That is often the best story. Reporters trust sources who publish findings that are not purely promotional. As long as the finding is accurate and does not misrepresent your own business, a candid number builds more credibility than a cheerful one.
How is a data story different from a predictions pitch?
A data story reports what already happened, with evidence. A predictions pitch forecasts what will happen next. The strongest January pitches combine them: here is what changed this year, and here is what it suggests about next year.
If you have a dataset and are not sure whether there is a story in it, our media relations team can pressure-test the finding, build the method note, and take it to the right reporters. Start a conversation before the December window closes.
