When to run this checklist
Run it once the data is analysed and the asset page is built, before the story is offered to the first journalist or released under embargo. Run the data group again if the analysis changes after the first pass, because a revised figure is usually the one nobody else has checked. A story pitched in waves, such as an exclusiveA story offered to one writer or title first and alone, used when one prominent piece is worth more than several smaller ones. followed by wider release and then regional cuts, needs the list once before the first wave and a quick look at the outreach group before each later one. A study repeated each year passes the whole list again for every new edition, and the tutorial on running an annual data study covers what changes between editions.
The PR or content lead runs it, with the analyst who produced the figures and whoever owns data protection in the business. The digital PR for links guide explains how the search team will judge the campaign afterwards, which is worth agreeing before the pitch rather than after. When a press release goes out with the story, the press release distribution checklist still runs on the final release text; this list does not repeat it. A team that pitches every week from a bank of stories runs that rhythm through the PR programme playbook, and uses this list for each data-led story within it. A survey commissioned for the story has its own earlier gate: the survey launch checklist runs before the questionnaire reaches its sample, long before the findings are pitched.
The research statements on this page are the ones the data stories guide and the questionnaire pretest tutorial make from the codes of the Market Research Society and ESOMAR: a sample that represents the group the headline describes, neutral questions, and a published method note. What counts as an adequate Sample SizeThe number of users or sessions an experiment needs to detect a given effect reliably. for a given claim is a matter for those codes and a trained researcher, and the list does not state one. Privacy lines are general direction, not legal advice.
Before you start
- The analysed data and its source, with the dates, the question wording for any survey and the cuts planned for each audience.
- The draft headline and its alternatives, in the style of the titles you will pitch.
- The draft methodology note: who or what was measured, when, how and with what limits.
- The asset page, live on a staging address or ready to publish, with its chart images.
- The claims and evidence record for every figure and statement the story makes.
- The target list, segmented, with the pitch and data cut for each segment.
- The draft release or notes to editors, if a release will go out with the story.
- The news values you will test the headline against.
The checklist
The groups follow the order in which a data story is usually checked: the data first, because nothing else matters if it fails; then the method and the release terms; then the page journalists will cite; then the headline; and finally the outreach plan. A line in the data group that cannot be ticked stops the pitch.
The data
Method and release
The asset page
Headline and angles
Outreach
Where each check is explained
The widget keeps the lines short so they can be ticked; the reasoning lives in the guides. Each check links to the page that explains it.
| Check | Explained in |
|---|---|
| Sample represents the group in the headline | Campaign ideation and data stories |
| Published survey wording pretested | Pretesting a questionnaire |
| Customer data figures cleared | Campaign ideation and data stories |
| Source and period cited for public data | Campaign ideation and data stories |
| Each claim checked against its evidence | Building a claims library |
| Methodology note written | Campaign ideation and data stories |
| Method note in the notes to editors | Press releases and the online newsroom |
| Embargo terms set | Press releases and the online newsroom |
| Asset on a stable, relevant page | Campaign ideation and data stories |
| Headline finding on top, method one click away | Campaign ideation and data stories |
| Static chart images supplied | Campaign ideation and data stories |
| Separate campaign domain planned | Microsites and multiple domains |
| Headline tested against news values | Campaign ideation and data stories |
| Alternative headlines for each kind of title | Campaign ideation and data stories |
| Story survives a weaker finding | Campaign ideation and data stories |
| Target list by segment | Campaign ideation and data stories |
| Exclusive and release order decided | Campaign ideation and data stories |
| Outreach log set up | Media relations and pitching |
| Link asked for once, never as a condition | Digital PR for links |
If an item fails
The data group is the hard stop. A sample that does not represent the group the headline describes means either a narrower headline that the sample does support or no pitch; the data stories guide is clear that journalists check, and a thin survey is remembered. A figure from customer data that legal or data protection has not cleared stays out of every version of the story, including the regional cuts, and any cut small enough to point to a person or a client is dropped rather than rounded. Customer data that names or identifies people is Personal DataAny information relating to an identified or identifiable living person, such as a name, email address, location data or an online identifier., and the people who own data protection make that call, not the PR team. A claim with no evidence behind it in the claims library is removed or rewritten until it says only what the evidence supports.
The method and release group is the next stop. A story without a methodology note waits for one, because the journalists most worth reaching will ask for it, and the press releases and online newsroom guide puts it in the notes to editors so they do not have to. Embargo terms that are unclear are settled before anything is sent; an embargo is an agreement with writers, offered only to those you trust, as the pitching guide explains.
The asset page and headline groups can sometimes go ahead with a note. Chart images still being resized can follow the first pitch by a short interval if the page itself is ready. A headline that fails the news values test should be rewritten, not pitched harder. A separate campaign domain with no owner or afterlife is fixed before launch, as the microsites and multiple domains guide describes, because links earned by a domain nobody maintains are lost when it lapses.
In the outreach group, a missing log can be set up on the day, but not skipped: it is the record that later measures earned Share of VoiceYour share of market visibility. and feeds the next story. Any plan that makes a BacklinkA link from another site to yours. the price of the story is dropped. The link building methods guide sets out why links are earned rather than demanded, with Google's spam policies as the boundary.
A self-storage firm pitches a moving-day study (illustrative)
A moving-day study with one cut removed
Grenzivo Storage, an invented self-storage firm with sites in several cities, wants a story from its own booking data on when people move house. Its content lead runs this list with the analyst and the company's data protection manager in late August. The analysis covers 38,000 bookings over three years and finds that the last Friday of the month is the busiest moving day, with bookings on that day running about 2.4 times the average weekday.
The data group catches one problem. A planned regional cut for a small site covers only 14 bookings on the busiest day, few enough that a reader who knew a neighbour had moved might identify the customer. The data protection manager asks for it to be dropped, and the regional story is told for the larger sites only. The claims check finds that a draft line calling the firm the busiest storage provider in its cities has no evidence, and it is cut.
The method note states the period, the number of bookings, that the data covers the firm's own customers rather than all movers, and that bookings are a proxy for moving day. It goes into the notes to editors. The headline is tested against the news values: timeliness through the end-of-month pattern, proximity through the city cuts. An alternative for property trade titles leads on the contract completion pattern behind it, and the team checks that a weaker finding would still support a story about planning a move.
The national angle is offered as an exclusive to one consumer title on a Tuesday, the wider release follows the next day, and the city cuts go to 9 regional titles over the following week. The log records each pitch and reply, and one writer who used the data without citing it receives a single request for a link to the study page.
Common mistakes
- A headline wider than the sample. A study of the business's own customers presented as a finding about everyone is the first thing a sceptical journalist checks.
- Regional cuts nobody checked for privacy. The smallest cut is the one most likely to point to a person.
- No method note, or one buried. Journalists who cannot find the method either ask, delay or move on.
- An asset page built for the brand rather than the reader. A reader arriving from an article needs the finding at the top, not the company story.
- Interactive only. Many titles will not embed an interactive piece and need a static chart image instead.
- One pitch for every title. National, regional and trade titles need different headlines and data cuts.
- A fragile story. A campaign that only works if one finding comes out strongly has nothing to pitch when it does not.
- Pressing for links. A link made the condition of the story damages the relationship and the campaign.
The checklists library holds the lists that sit around this one, including the press release distribution checklist for the release that goes out with the story.
