web analytics
Back to blog

The best content marketers are editors first

Extractable summary

Content editing, the ability to look at a draft and see what’s structurally broken, inaccurate, off-brand or missing, is what separates a senior content marketer from a junior one, and most hiring processes never test for it. Writing is generative and editing is evaluative, and judging a draft against a standard is the rarer skill. It works on three levels (structure, substance and voice), with proofreading a distant fourth. Now that roughly half of new web articles are AI-generated, the judgement to tell a good draft from a merely fluent one is worth more than the ability to produce one. To hire for it, hand candidates a flawed piece and ask what’s wrong with it.

Content editing is the skill that carries content marketing, and almost every hiring process tests for something else. Every one I’ve seen obsesses over writing samples: send us three articles, show us your portfolio, prove you can write. Writing counts, obviously; the trouble is that it doesn’t separate a senior content marketer from a junior one, and I’m convinced most teams are testing for the wrong skill. The best content marketers I know would rather be judged on their editing than their writing, because the work depends on looking at a draft and seeing precisely what’s wrong with it, what’s missing, what’s structurally broken and what quietly sounds like the wrong company.

Comparison for hiring content marketers: a writing sample proves clean prose under ideal conditions, while a content editing test reveals whether someone can see what's broken, which claims to check and where the voice slips.

Why is content editing harder than writing?

Because writing is generative and editing is evaluative, and evaluating against a standard is the rarer skill by a distance. Anyone with a topic and a few hours can produce a draft, but far fewer people can pick that draft up and tell you, specifically and correctly, why it doesn’t work and what would fix it.

Producing words is the easy part now: Graphite’s 2026 analysis of 55,400 English-language articles found that primarily AI-generated articles have made up roughly half of new articles since early 2025. When a machine can hand you a competent-looking first draft in seconds, knowing whether those words are any good becomes the scarce part, and the value moves to editorial judgement: whether an argument holds, whether a structure serves the reader, whether a claim is true. AI made it easier to publish without making it any easier to publish something good, and content editing fills the gap between the two.

What are the three kinds of content editing?

Structural, substantive and voice editing, with proofreading a distant fourth. People hear “editing” and think typos and commas, which is the smallest and least valuable layer:

LevelThe question it asksWhat it catches
StructuralDoes the argument work?A buried lead, sections in the wrong order, a piece that wanders off before concluding anything, a premise that’s wrong
SubstantiveIs it accurate, and is it useful?Stats with no source, confident sentences that are subtly false, pieces that fill a calendar slot without helping anyone
VoiceDoes it sound like the brand?Generic content-mill prose, the tone of whoever wrote it that day, words the brand never uses
ProofreadingIs it clean?Typos, commas, formatting

Structural editing asks whether the piece opens where it should, builds in an order that makes sense and lands the point it set out to make, or whether it buries the lead under three paragraphs of throat-clearing. It means rearranging, cutting, and sometimes deciding the whole premise is wrong. It also serves how people read: Nielsen Norman Group’s classic study found 79% of users scanned web pages and only 16% read word by word, and copy that was concise, scannable and objective together tested 124% better on usability.

Substantive editing is the layer that’s become non-negotiable: fact-checking every claim, catching the stat with no source, and asking whether the piece helps the reader. Drafts now arrive sounding authoritative whether or not they’re true, and AI tools are confidently wrong more often than people assume: when the Tow Center for Digital Journalism tested eight AI search tools on 1,600 queries, they answered more than 60% incorrectly. As researchers Klaudia Jaźwińska and Aisvarya Chandrasekar put it, “Most of the tools we tested presented inaccurate answers with alarming confidence.” An AI-assisted draft inherits that confidence, and only an editor checks whether it’s earned. It’s the same discipline I wrote about in writing for technical audiences, where one wrong claim loses the reader for good.

Voice editing is the most subjective layer, and without a defined standard behind it, it’s just personal taste, so the standard has to exist first.

Content editing levels in order: structural asks whether the argument works, substantive asks whether it's accurate and useful, voice asks whether it sounds like the brand, then proofreading last.

Why do most content problems turn out to be editing problems?

Because the failure almost never lives at the sentence level. You hire a good writer and the content is still mediocre: the sentences are clean, the grammar’s fine, and the piece somehow doesn’t land. The problem is the structure that buries the point, the unchecked claim that erodes trust, the argument that doesn’t quite hold, the tone that’s a half-step off the brand. Those are content editing failures, and no amount of pretty prose fixes them. B2B marketers half know this already: in CMI’s 2025 research, 43% named differentiating their content as a challenge, more than the 39% who struggled to create enough of it.

The fastest wins I’ve seen came from editing, too: in my current role at OpinionX, I audited and updated the whole existing blog library for AI search, and average Google search position moved from 18.4 to 8.8 in five weeks, with search click-through up 54% on the Q3 average. The work behind that was editing what was already published, and the audit method is in the AI readiness audit for a content library.

A good writer with no editorial eye produces clean drafts that are wrong in ways they can’t see. Give me a merely decent writer with a sharp editorial eye instead, someone who turns out work that’s structurally sound, accurate and on-brand because they catch the problems before anyone else does. That second person is the one you want and the harder one to find, which is why testing for writing alone keeps handing teams the wrong hire. I made the case for writing quality as a business asset in good writing is a competitive moat in B2B; content editing is the skill that produces it, and its absence is a big part of why B2B content sounds the same.

Stat card on content editing at OpinionX: updating the existing blog library moved average search position from 18.4 to 8.8 in five weeks and lifted search click-through rate 54%.

How do you develop an editorial eye?

Learn to read your own draft as a hostile stranger. Most people read their own work and see what they meant to say, which is why self-editing is so hard. Steven Pinker calls the underlying problem the curse of knowledge, the difficulty of imagining what it’s like not to know what you know, and writes that it “is a major reason that good scholars write bad prose.” Distance helps: sleep on it, change the font, read it aloud, do anything that tricks your brain into seeing the words on the page instead of the ones in your head.

The other half is having a standard to edit against, which turns content editing from a matter of opinion into a matter of fact. At a B2B SaaS company serving higher education institutions, building the editorial standards from scratch (the brand voice, the rules, what we said and what we never said) was the most important work I did there, worth more than any single article I wrote. Once that standard existed, I could measure every piece against something concrete instead of arguing over taste, and “is this on brand?” stopped being a shouting match and became a question with an answer. Putting the standard at the centre of how you produce content is the systems thinking in you don’t need a content team, you need a content system.

The same standard is what makes AI drafts usable: when I delegate first drafts to AI, the editing pass against a written standard is what turns them into something I’d publish, and without it the draft stays fluent and generic.

How do you test content editing when hiring?

Stop auditioning only the writing; hand candidates something flawed and ask them to tell you what’s wrong with it. A good test piece has one structural problem, a couple of unsourced or wrong claims and a stretch where the voice slips, and a strong editor finds all three before mentioning a single typo. The ones who can see it are the ones worth hiring, and they’re rarer than the portfolio makes them look.

Hiring test for content editing: plant a structural problem, two bad claims and a voice slip in a draft, and strong editors find all three before mentioning a typo.

Where does this argument fall short?

Writing still counts: an editor who can’t write can diagnose a problem without being able to fix it, so the best hires do both; the argument is about which skill separates seniors from juniors, and neither is optional. Editing also needs time and authority: a content marketer with a sharp eye but no standard to point to, or no power to send a piece back, ends up with opinions nobody acts on. That’s why the editorial standard comes before the editing, and why it needs leadership behind it.

Frequently asked questions

For senior roles, generally yes. Writing is the baseline skill, the thing that gets you in the door, but content editing is what separates a senior content marketer from a junior one. Editing is evaluative rather than generative, which makes it the harder and rarer skill: plenty of people can produce a competent draft, while far fewer can look at one and identify exactly what’s structurally broken, inaccurate, or off-brand. As AI makes passable drafts effortless to generate, the editorial judgment to tell good from merely fluent has become the higher-value skill.

Structural editing asks whether the argument works: whether the piece is ordered well, makes its point, and doesn’t bury the lead. Substantive editing asks whether the content is accurate and useful: fact-checking claims, catching unsourced stats, and confirming the piece genuinely helps the reader. Voice editing asks whether it sounds like the brand rather than a generic content mill or whoever happened to write it. Proofreading, the typo-and-comma layer most people picture, is the smallest and least valuable part of the job.

Stop auditioning only their writing samples and hand them a flawed piece to critique. Ask them to tell you what’s wrong with it: where the structure fails, which claims they’d want verified, and where the voice slips off-brand. A strong editor will spot the structural and substantive problems well beyond the surface typos. This reveals far more about seniority than a polished portfolio, which only proves they can write under ideal conditions rather than fix what’s broken underneath someone else’s draft.

How did this land?No reactions yet
Solange Rainha
Solange Rainha
Content Marketing Manager | 10+ Years B2B SaaS & AEO/LLMO