Writing And Reporting For The Media

The Practical Workflow Most People Get Wrong

Here is how the job actually goes. You get a deadline. Usually it is three hours from now. You have a source document — sometimes it is clean, sometimes it is forty pages of PDFs someone scanned on a bad day. You open your notes app and start extracting the factual skeleton: who, what, when, where, numbers. That is step one. Step two is checking whether each fact survives contact with reality. I wasted two months early in my career running quotes through a fact-check workflow that was basically “read it once, move on.” One story ran with a misattributed figure because I did not verify the original spreadsheet. The correction cost me two editorials and a real conversation with a reader who was angry at the wrong person. After that I started building a verification step into the process from the start, and the turnaround time actually got faster, not slower, because I stopped having to redo work. The term gets thrown around in journalism schools and newsroom training manuals, but it does not mean much without the details. At its core it is a set of procedures for turning raw information into copy that meets editorial standards under deadline pressure. The “reporting” part is the investigation — pulling documents, talking to sources, confirming data. The “writing” part is the assembly — taking those confirmed facts and structuring them into a format a reader can process in a short window of time. The “media” part means understanding which platform you are writing for, because the difference between a broadcast script and a digital investigative piece is not just word count. It is sentence rhythm, attribution style, and how you handle uncertainty. I will walk through a typical cycle using a concrete example because abstract process descriptions are not helpful. Say you are covering a local government vote on a zoning change. You start by pulling the public meeting minutes and the agenda packet from the municipal website. These documents are usually poorly organized. The key vote tally might be buried in a twelve-page appendix. You identify the decision points: what was proposed, what was voted on, the margin, the dissenting positions, and any conditions attached to approval. You verify the numbers against the official record — never against the news release the city sent out, because those frequently contain rounding errors or outdated figures.

Next you interview people who have a position on the outcome. That means the developer pushing the rezoning, the neighbor opposing it, and the council member who cast the deciding vote. You do not lead with your own opinion. You ask open questions and record what they say. If a source gives you a number you cannot verify from the document trail, you flag it and either confirm it independently or attribute it clearly as their claim. This is where most amateur reporters fail. They trust the first number they hear instead of treating every numeric claim as provisional until documented. After the interviews you draft the story. The structure follows a reverse pyramid for straight news: the most important facts up top, supporting details below, background at the end. For longer features the structure can be more narrative, but the hierarchy of information still matters. Readers decide within the first two sentences whether to keep reading, so the lede needs to carry the essential facts without hedging. “The city council approved the Pine Street rezoning 4-3 on Tuesday” is better than “A controversial rezoning proposal moved forward Tuesday after a lengthy debate.” One tells the reader what happened. The other tells the reader what the article is about.

Source Verification: The Part Nobody Teaches Well

Here is something I learned the hard way. Document-based reporting beats pure sourcing almost every time, but documents lie too. They lie through omission, through formatting tricks, through the difference between a draft and a final adopted version. When you are working with government records, always check whether the version you are reading has an adoption stamp or a clerk signature. An agenda item marked “for discussion” is not the same as an adopted ordinance. I once published a story based on a planning commission draft that had been substantially altered before the city council vote. The final version removed two of the five environmental mitigation requirements. The draft I had quoted from was wrong, and so was the story. I fixed it the next day, but the damage to credibility was real. For financial or statistical reporting, go to the primary source. If a press release says “unemployment dropped by 12 percent,” find the original labor department table and calculate the percentage yourself. Press releases love relative percentages because they look impressive. A drop from 50,000 to 44,000 is a 12 percent decline. A drop from 4 percent to 3.5 percent is a 12.5 percent decline but only half a percentage point in absolute terms. Both are technically correct. Only one tells the whole story. Your readers deserve both numbers in context. Interviews should be corroborated whenever possible. If a source tells you something that changes the story, and you can verify it from a second independent source, the story is stronger. If you cannot verify it, you either omit it or present it as unconfirmed. There is no middle ground that satisfies editors who care about accuracy. The tradeoff is speed. Corroboration takes time. In a breaking news environment you sometimes publish with a single source and update later when you confirm. That is an accepted practice in most newsrooms, but you should flag it explicitly in the copy if you do it, and you should correct it as soon as new information arrives.

Platform-Specific Writing Decisions

The way you write for print is different from the way you write for broadcast, which is different from digital. Print allows for longer sentences and more complex paragraph structures because the reader controls the pace. Broadcast requires short sentences, active voice, and natural pauses because the anchor reads it aloud under time pressure. Digital sits somewhere in between. It rewards skimmability — short paragraphs, clear subheads, data embedded visually when possible — but it also supports depth that neither print nor broadcast can match. Above all, know your audience and your constraints. A wire service story has different requirements from an investigative feature. The former prioritizes speed and universality. The latter prioritizes original reporting and narrative cohesion. Both are valid. Neither is superior. The mistake is applying wire service structure to investigative material or vice versa. It makes the writing feel tone-deaf to the effort that went into the reporting.

The Hard Parts

I will be blunt about what does not work. Deadlines create pressure that encourages shortcuts. Newsroom budgets are thin. Reporters are expected to file multiple stories a day, sometimes under two hours per piece. Under those conditions, the verification steps I described above get shortened or skipped. This is not an excuse — it is a structural problem. The workaround is personal discipline, not institutional support. Build a checklist you follow every time, even when you are exhausted. “Did I verify every number? Every name? Every attribution?” is a simple question, and answering it honestly prevents most errors before they reach publication. Another difficult area is handling anonymous sources. They have a place in serious reporting, especially when the source is exposing wrongdoing at personal risk. But they weaken a story if used carelessly. The best practice is to require on-the-record confirmation of the underlying facts whenever possible, even if the source themselves remains anonymous. If you cannot confirm the facts without the source, you need to explain why in your copy or in an editor’s note. Readers can tolerate anonymous sourcing more than they can tolerate unverifiable claims presented as fact.

A Practical Checklist for Day-to-Day Work

Save this. Use it. Do not skip it because you are in a hurry.

• Identify every factual claim in your draft and tag it with its source. Document, on-the-record source, off-the-record source, or unverified. If you cannot tag it, you do not have enough information to publish it yet.
• Verify every number independently. Even if two sources agree, check the original document. Agreement between sources is better than nothing, but it is not the same as primary verification.
• Confirm names, titles, and affiliations before publication. A wrong name is the easiest error to fix and the most damaging to trust. A quick call to the relevant office or a cross-check against a public directory takes under two minutes and prevents embarrassing corrections.
• Test your lede against the facts. Read the first sentence aloud. Does it accurately reflect the substance of the story, or is it clever for cleverness’ sake? If the lede and the facts diverge even slightly, rewrite the lede or rewrite the story.
• Run attribution on every quote that carries a claim. “The mayor said unemployment will fall,” not “Unemployment will fall.” The difference is the difference between reporting and advocacy. Editors can tell, and readers can tell too.

The process is not glamorous. It is slow, repetitive, and often frustrating. But it is the only way to produce work that survives scrutiny. The alternative is corrections, retraction notices, and the slow erosion of reader trust. I would rather have a slower turnaround and fewer errors than the other option. It has worked for me, and it should work for anyone who takes this seriously.