\

Content team roles: a functions-first guide for small teams

Written from several years running lean content operations, two and three-person teams covering strategy, editing, and distribution without a dedicated hire for any of them.

Picture a content team and you probably picture a writer. That's the reflex, and it's why so many small operations end up with someone who can write a clean sentence but has no clear brief for what to write or why. Writing is one job among several. Someone has to decide what gets made, someone has to check it holds up, someone has to make sure it can be found, and someone has to look at what happened after it went out.

On a twelve-person team, those are different job titles. On a team of two, or a team of one, they're different hats worn on different days. CMI's 2025 B2B research found that 54% of companies with a dedicated content team run it with two to five people, and another 24% have no dedicated team at all, just the work spread across other roles, vendors, and whoever has a free afternoon (source). Most teams are small. The question isn't whether you can afford eight job titles. It's whether you've named the eight jobs.

Start with functions

Content team roles work best when you stop thinking about them as headcount and start thinking about them as functions that need an owner, even if one person owns four of them.

There's a rough sequence to the work: decide, research, write, edit, get found, get seen, learn. That figure comes from CMI's own survey methodology, which asks marketers to self-report team size and structure rather than pulling from payroll data, worth knowing before you treat it as a precise headcount. Skip the naming and the sequence still happens, just badly. Someone still decides what to publish, they just decide it by reacting to whatever felt urgent that week. Someone still edits, they just do it by skimming their own draft five minutes before it goes live. The functions don't disappear when nobody's assigned to them. They just get done worse, by whoever's nearest.

Here's what that looks like at two team sizes:

One-person team. Strategy, research, and analysis happen in a Monday planning block. Writing runs Tuesday through Thursday. Friday morning is a cold edit of Monday's draft, plus an hour on distribution, turning last week's post into two social snippets and a newsletter blurb. SEO checks happen before drafting starts, not after.

Two-person team. One person owns strategy, research, and analysis, essentially the thinking work. The other owns drafting and editing swaps with them on alternating pieces, so nobody edits only their own copy. Distribution and SEO get split by whoever has capacity that week, tracked on a shared doc rather than assigned permanently.

That's the case for treating this as a functions list rather than an org chart. A team of two can cover strategy, research, writing, editing, SEO, distribution, and analysis between them. A team of one can cover the same ground, just sequentially rather than all at once. What breaks small teams isn't the number of functions. It's pretending one of them doesn't need an owner because there's no separate person to give it to.

Strategy

Strategy is the most commonly missing function on small teams, and when it's missing, publishing turns reactive almost immediately.

Without someone owning what gets made and why, a team drifts toward whatever's loudest that week: a competitor's post, an idea from three weeks ago that never got finished, a topic that seemed clever on a Tuesday. None of it is wrong exactly. It just doesn't add up, because nothing ties it to an audience, a funnel stage, or a set of searches worth owning. A blog built this way can read fine post by post and mean nothing as a whole.

On a one or two-person setup, strategy doesn't need a dedicated hire. It needs an hour, ideally the same hour every week, spent deciding what the next few pieces are for. Say a two-person team spends that hour mapping the month against one goal, a segment they're trying to reach, a stage of the funnel that's underperforming, and says no to everything that doesn't serve it. That's the whole job. Skip it and the writer becomes the de facto strategist by default, deciding topics on the fly with no time to check whether they're worth the effort.

Research and source material

Research and writing are different functions, and treating them as one is how you end up with content that reads fine and says nothing.

The researcher's job is finding the substance the writer will shape, a data point, an expert's actual opinion, a close read of what already exists on the topic. Cut this step and you get a competent-sounding piece with nothing underneath it, the twentieth version of an argument that was already tired by the tenth. Good prose can't manufacture a fact it doesn't have.

Part of this is competitive, not just factual. Before committing to a topic, someone needs to check what's already ranking and where the actual gap sits, otherwise the team ends up polishing an angle that's already been covered better elsewhere. This is also where search intent and SEO belong, at the front of the process rather than as a fix applied after publishing. Understanding what a searcher actually wants from a query should shape the outline before a word of the draft exists. A well-argued piece can sit on page four because nobody checked, before writing, what people searching that term were actually trying to do. Fixing that after publishing means rewriting the piece, not tweaking a meta tag.

Drafting and editing

Drafting turns research into something a reader follows from the first line to the last, and editing is the separate function of checking that it actually does.

The writer's job is structure and pacing as much as vocabulary, knowing which sentence should be short and which one can carry a longer thought, and doing it consistently enough that ten posts from the same operation sound like one publication rather than ten moods depending on who was tired that day.

Editing is the function small teams skip most often, and skipping it is usually the difference between content that's fine and content that's good. An editor reads the way a stranger would, catching the point buried in paragraph four or the headline promising something the piece never delivers. A rough working rule from lean-team practice: a solo cold edit the next morning catches maybe half of what a second reader would, structural gaps, weak openings, a claim that needed a source and didn't get one, but it rarely catches tone drift the way a genuinely fresh pair of eyes does. You can't fully edit your own work, because you read what you meant to write instead of what's actually on the page. On a solo setup, the workaround isn't talent, it's time: draft something, leave it overnight, and read it back cold the next morning. It's not as reliable as a second person with no stake in the piece, but it catches the worst of it, and it's a habit worth building even when there's no one else to hand the draft to.

Findability and repurposing

Getting found and getting reused are two separate functions, and neither one happens automatically once a piece is published.

Findability starts before drafting, not after. Ahrefs points to Sparktoro data showing that 63.41% of all US web traffic referrals come from Google (source), a figure drawn from Sparktoro's clickstream sampling rather than Google's own reporting, so treat it as a strong directional estimate rather than a census. It's still a plain argument for why someone on the team needs to own search visibility even when the team is two people. That ownership increasingly stretches beyond classic search rankings too. Whether a piece is structured clearly enough to get cited by AI answer engines is now part of the same job, not a separate one.

Repurposing is the second function, and it's easy to treat publishing as the finish line when it's really the starting point. One well-built article can become a week of social posts, a newsletter section, or a short video script, but only if someone owns turning it into those things on a regular cadence rather than as an occasional afterthought. Ahrefs' write-up on Wirecutter's SEO work makes the same point from a different angle, that publishing strong content isn't sufficient on its own without a deliberate plan for getting it distributed across more than one channel. Buffer runs its own content operation with a team of three people and leans on AI to clear repetitive work so the team's time goes to the judgment calls, strategy, editing, deciding what's worth repurposing and how (source). That's a reasonable model for any small team deciding where its own hours are best spent.

Close the loop

Analysis is the function that turns last month's publishing into next month's plan, and without it, every decision is a guess untested by the one before it.

Someone has to look at what happened after a piece went live and turn it into input for the next round: which topics actually drove the goal, which ones sank without a trace, what the pattern looks like across six months rather than one lucky post. On a two-person team, this often means the difference between renewing a topic cluster and dropping it. Say one post in a cluster of five brings in the bulk of qualified signups while the other four sit flat. Without a recurring look at the numbers, the team keeps producing more of the flat four because they felt fine to write, not because anyone checked what actually worked. Semrush's guidance on content workflows makes a related point worth keeping in mind here, that even lean teams benefit from clearly assigning who handles publishing and follow-up tasks rather than assuming they'll get covered by whoever's free (source). On a team of two, this doesn't need a dashboard. It needs a recurring half hour where someone actually looks at the numbers and asks what to do differently, not just whether traffic went up.

This is also where content team roles stop being an abstract list and start being a habit. HubSpot's framing of content operations as people, process, and technology working across strategy, creation, publishing, distribution, and management is a useful shorthand for the whole loop (source). None of that requires eight job titles. It requires eight jobs getting done, by however many people you've got, with someone accountable for each one even when that someone is the same person twice over. HubSpot's own guidance for teams of one to five marketers backs a lighter setup too, recommending a shared doc for planning and a simple board for tracking rather than dedicated software for each function (source). The tooling can be minimal. The ownership can't be.