Your newsroom isn't slow — it's stuck where nobody is looking

Every newsroom has a plan for the predictable: election night, the budget vote, or the Friday print deadline. What breaks a newsroom is rarely the event on the calendar. It is the photo nobody can find, the legal sign-off stuck in an inbox, and the one editor who has become the only person able to approve everything. Although such bottlenecks rarely appear in a planning meeting, they do decide how fast — and how well — local journalism reaches its readers. AI hasn't made these bottlenecks disappear. In many cases, it has simply moved them.

Ask a managing editor how the day went, and the conversation will likely turn to everything that happened around the published stories. The courts reporter filed on time, but the photo was sitting on a freelancer's laptop. The investigation was ready on Tuesday, but it waited until Thursday for a lawyer to read it. The web version went live at noon, while the print version was still being rewritten by hand at 8 p.m.

None of these delays were anticipated. Yet together, they consume hours every week that small and mid-sized newsrooms simply no longer have. According to the latest State of Local News report from Northwestern University, newspaper employment dropped by another 8% last year — more than 7,500 jobs.  Fewer people now carry the same number of hand-offs, approvals and channels.

With teams already stretched thin, the time lost between reporting and publishing is becoming harder to absorb.

The bottlenecks that never make the planning board

Most newsroom plans are built around content: which stories run, who writes them and when they are due. The real friction, however, sits in the spaces between people and systems. Five bottlenecks show up again and again:

• The single point of approval: In many lean newsrooms, one senior editor has quietly become the gatekeeper for headlines, sensitive stories, social posts and front-page decisions. When that person is in a meeting, on vacation or out sick, the entire flow stops.
• The asset hunt: The story is done, but the photo is not. It sits in an email attachment, on a local drive, or in a forgotten folder. And once it is found, someone still has to check the rights, write the caption and crop it for every channel.
• The late legal or sensitivity check: Stories involving courts, minors or allegations need an extra pair of eyes. Because that additional check is treated as a final step rather than a parallel one, it often starts only when the clock is already running out.
• The print-digital rework loop: A story that went live online at noon gets cut, rewritten and refit for print in the evening. Corrections made on one channel do not always reach the other, which creates both duplicate work and inconsistent facts.
• The collision of the planned and the unplanned: A storm, a shooting or a surprise resignation does not replace the scheduled work. It lands on top of it, and the same handful of people must handle both.

Why our planning keeps missing them

If these bottlenecks are so familiar, why do they still catch newsrooms off guard? The answer lies less in poor discipline and more in how newsroom planning was originally designed.

Workflows grew by layering, not by design

Most newsrooms did not build their digital operation from scratch. They built on systems originally designed for print, adding new channels like social media, newsletters and podcasts along the way. Each brought its own tools and handoffs, often without changes to the underlying processes. A study by FT Strategies and WAN-IFRA involving 448 newsrooms found that 64% still produce news for a single legacy channel. It highlights how many are still operating with processes that were never designed to support publishing across multiple channels.

Knowledge lives in people, not in systems

When staff numbers shrink, the institutional knowledge of who approves what and where files live continue to narrow. That works until one of those people leaves. Every departure takes a piece of the workflow with it, one that was never formalized.

There is no room left for the unexpected

Finally, bottlenecks are a capacity problem. A team running at full stretch every day has no buffer to absorb a breaking story, a sick day or a broken file transfer. Without that buffer, even a small delay ripples through every step that follows.

What AI changed — and what it didn't

AI was supposed to make bottlenecks disappear. In some places, it has. Transcription, translation, tagging, image cropping and first drafts now happen in minutes rather than hours. The Reuters Institute’s 2026 Trends and Predictions report found that nearly all publishers now consider back-end automation important, and one Kenyan publisher reported a rise in output of roughly 40% after introducing an AI assistant.

But speeding up production at one end of the pipeline creates a new problem at the other. While AI may generate 10 social posts, three headlines and a newsletter summary in seconds, someone still has to check every one of them. The bottleneck has not disappeared. It has merely shifted — from making content to approving it.

The Associated Press’ updated standards from July 2026 require a journalist to review and edit every AI-generated output before it reaches readers. While this is the right principle, it means now review capacity, not production capacity, sets the pace. The strain is already visible. In a recent Reuters Institute analysis based on interviews with 20 newsroom leaders and experts, the middle-management layer was described as overloaded, with AI oversight piled on top of daily quality control. One publisher even introduced limits on how much AI-assisted output staff are expected to review, because of reviewer fatigue.

AI has also introduced two subtler bottlenecks:

• Trust calibration: Editors must decide, case by case, how far they can trust a tool. Without clear rules on which outputs need a full review and which need just a quick check, everything defaults to a full review — or worse, to no review at all.
• Fragmented tools: Many newsrooms adopted AI one tool at a time — each sitting outside the core workflow. Every copy-paste between a chatbot and the CMS is a new hand-off, and every hand-off is a potential delay.

In short, AI did not remove the bottlenecks nobody plans for. It concentrated them on the people newsrooms can least afford to overload: their editors.

Designing a newsroom that bends instead of breaks

You cannot predict when the next storm, scandal or sick day will hit. But you can build a workflow that absorbs the shock rather than amplifies it. That starts with four practical shifts:

• Make waiting visible: Track not just who owns a story, but also where it sits and for how long. A shared, centralized planning and production environment instantly shows where content is stuck, so editors can act before a delay turns into a missed deadline.
• Spread approval rights: Define clear rules for which content needs senior sign-off and which does not. Back them up with role-based workflows, so a second editor can step in without asking where everything is.
• Give every asset one home: A single source of truth for photos, video and graphics, with rights information and automated renditions built in, turns the asset hunt into a simple search.
• Review by risk, not by default: Not every AI output deserves the same scrutiny. A suggested alt text is not an investigative headline. Tiered review rules, with AI built into the core workflow rather than added on top, let editors spend their attention where the stakes are highest.

And crucially, treat the legal or sensitivity check as a parallel track. When reviewers can see a story while it is still being written, the final check becomes a confirmation rather than a last-minute scramble.

From firefighting to flow

The bottlenecks nobody plans for are not a sign of failure. They are a sign that newsroom workflows were built for a different era — one with more people, fewer channels and no machines producing drafts at scale.

The publishers that thrive in the coming years will not be the ones that produce the most content. They will be the ones that move it most smoothly, from the first assignment to the final approval, without wearing down the editors in between. When the flow is right, the unexpected becomes manageable. And newsrooms get back what matters most: time for the journalism their communities depend on.

About WoodWing:

WoodWing enhances your content ecosystem by combining technology with extensive publishing expertise. For over 25 years, we’ve driven innovation in editorial and production workflows, empowering publishing and creative teams to create, manage and deliver content across print and digital channels with greater efficiency, speed and consistency.

Our portfolio includes solutions for collaborative multi-channel content production, digital asset management, quality management, knowledge management and enterprise information management.

Founded in 2000, WoodWing Software BV is headquartered in the Netherlands, with offices in the U.S. and Malaysia, commercial staff in LATAM, a global partner network and over 250 employees worldwide.

For more information, please reach out to Dan Pugliese at dan.pugliese@woodwing.com.