Skip to content
Saturday, August 29, 2026
PUBLIC MEDIA REVIEWPUBLIC INTEREST MEDIA · INNOVATION
S&P 500−0.35%FTSE 100−0.17%Euro/Dollar+0.22%Brent Crude+1.25%10-Year US+1.40%
PUBLIC MEDIA REVIEWPUBLIC INTEREST MEDIA · INNOVATION
Home / Innovation
Innovation

How Newsrooms Share a Data Project Without Sharing a Newsroom

Collaborative data investigations let small outlets do work none could afford alone — if the partners settle methodology, credit, and embargo rules before the first record request.

WE
William Elliott, · February 14, 2026 · 4 min read
ShareXFacebookLinkedInTelegramEmail
Close-up of a spreadsheet printout annotated by hand across a desk

Small newsrooms can do accountability data work they could never fund alone by sharing the labor: one outlet obtains the records, one analyzes, several contribute local verification, and every partner publishes the same weekend. Documented collaborations in the United States — from statehouse reporting alliances to the regional cooperatives formed as bureau sharing spread after 2018 — show the pattern holds across scales. What makes or breaks the project is not the analysis; it is the memo the partners write in week one.

What goes in the collaboration memo?

Five decisions, made before any records arrive:

  1. Data custody: who holds the records, where they live, and who can query them.
  2. Methodology sign-off: every partner approves the written methodology — definitions, exclusions, error handling — before findings are drafted.
  3. Local verification: each partner checks the rows for its own market; local editors are the only ones who know the name that is spelled wrong.
  4. Credit and bylines: who leads, who contributes, how partner logos appear, and what the shared boilerplate says.
  5. Embargo and exclusivity: a fixed publish window, with each outlet free to angle the story for its own audience.

Why does methodology need unanimous sign-off?

Because the methodology is the story's defense. When a statewide dataset names a local official, the first call the newsroom gets is a dispute about the data, not the journalism. A written methodology — what counts as a case, how duplicates were removed, what the records omit — converts that call from an argument into a document. Partners who have approved the methodology cannot disown it under local pressure, and that is precisely the protection a small outlet needs when its partner in the capital is not standing in the room.

How is the work actually divided?

The durable pattern assigns roles by comparative advantage. The partner with the records-desk experience files and manages requests; the partner with the data analyst owns cleaning and analysis; market partners verify cases and write local sidebars that make the collaborative project feel native to each audience. Version control and a shared workspace matter more than the specific tool — what matters is that every partner sees the same dataset and the same change log, so no outlet is publishing numbers it cannot trace.

What does the audience see?

Each partner runs its own framing, but the shared spine is consistent: a common headline format, the full methodology published alongside every version, and a consistent credit line naming all partners. Readers in one market should recognize the project when they encounter it in another. Transparency extends to funding — if the collaboration is grant-supported, every partner discloses the funder in the published package, following the disclosure norms that journalism nonprofits have made standard practice.

What are the classic failure modes?

Three recur. The big partner publishes first and the small partners' local angles become old news — prevented only by an embargo with teeth. The methodology is written after the findings, when it is a justification rather than a plan. And the verification step is skipped because the deadline is near — the single most common source of corrections, and the reason the memo puts local verification ahead of the publish date in the schedule, not after it.

Where should partners look for a first project?

Start with a dataset that is public everywhere but analyzed nowhere: inspection results, campaign finance at the county level, overtime spending, or 911 response times. A first collaboration should have a dataset small enough to finish, because the project's real product is not the story — it is the working relationship that makes the second, harder project possible.

Frequently asked questions

Who owns the records if one partner paid for them?

The memo decides. The common arrangement is joint editorial use with the paying partner credited; what a memo must never leave silent, because the dispute arrives exactly when the data proves most valuable.

Can a two-person newsroom contribute anything real?

Yes — local verification is the scarce resource in every regional data project. Small partners who verify carefully earn analysis help in return.

Should partners publish identical stories?

No. Shared spine, local angle. Identical stories waste the collaboration's main editorial advantage: the same data means different things in different markets.

Frequently Asked Questions

Who owns the records if one partner paid for them?
The memo decides. The common arrangement is joint editorial use with the paying partner credited; what a memo must never leave silent, because the dispute arrives exactly when the data proves most valuable.
Can a two-person newsroom contribute anything real to a data collaboration?
Yes — local verification is the scarce resource in every regional data project. Small partners who verify carefully earn analysis help in return.
Should partners publish identical stories?
No. Shared spine, local angle. Identical stories waste the collaboration's main editorial advantage: the same data means different things in different markets.