Small teams win on clarity rather than headcount, and the difference shows up in who decides what. This is how a four-person setup divides the work, what each role is actually responsible for, and the routine that keeps the output consistent when someone is on leave.
The four roles and the boundary between them
Lead development owns the site itself: clean structure, fast pages, mobile performance and the technical foundations that make everything else measurable. When a page ranks worse than it deserves, this is usually where the answer lives, in render speed or markup rather than in the copy.
SEO and strategy turns business goals into a roadmap: keyword research, content optimisation, technical priorities and a plan tied to numbers instead of impressions. The output is a short list of what to do next, not a hundred-page deck.
Web analytics keeps the measurement honest. Dashboards are only useful if someone is willing to delete a metric that never changes a decision, and that is a recurring job, not a setup step.
Copywriting connects search intent with sentences people finish reading. Every piece starts from the query it targets and ends with a tone that survives editing, which is harder than it sounds when four people touch the same draft.
The editorial routine that holds it together
- One planning session per month: pick topics, confirm intent and format for each.
- Drafts go to a reviewer who did not write them, always.
- Publishing happens on fixed days, so the developer and analyst are not on call at midnight.
- A short performance review six to eight weeks later decides rewrite, merge or leave alone.
- The calendar carries one buffer slot per fortnight for opportunistic pieces.
Why ethical practice is an operational choice
Working within search engine guidelines is usually framed as reputation management, but on a small team it is a capacity decision. Tactics that risk a penalty create cleanup work that nobody has spare days for, so the rules function as a shortcut: they remove whole categories of argument before they start.
What clients actually notice
The two comments that recur are about transparency and pace: reports in plain English that show what was done and what it achieved, and month-to-month engagement rather than long contracts. Both are cheaper to promise than to deliver, which is exactly why they are worth measuring. In practice they show up as fewer surprises, fixes explained in a sentence, and a roadmap that survives contact with a busy quarter.
The documents we actually keep
Small teams drown in documentation nobody reads, so we reduced it to five pages that each answer one question and get updated when they go wrong.
- Publishing checklist: the exact order of steps from approved draft to live URL.
- Review standard: what a reviewer checks, and what they are explicitly not allowed to rewrite.
- Metric ownership: one line per number, naming who investigates a sudden drop.
- Escalation path for a page that starts losing traffic, with a time limit on each step.
- Quarterly note on what we stopped doing, which is the only document anyone enjoys writing.
Metrics we deleted on purpose
- Word count per draft, because it rewarded padding and everyone knew it.
- Keyword density, which no writer has ever improved by following.
- Daily traffic screenshots, which measured the analyst's anxiety more than the site.
- Social shares on B2B posts where nobody had shared anything in a year.
- Time in draft, once we noticed it made people publish before the review.
Deleting a metric is a real decision that needs nerve, because dashboards are easy to add and awkward to shrink. The test we use is simple: if a number changes and nobody can name the action it triggers, it comes off the report. What remains fits on one screen and takes ten minutes to discuss, which means the meeting spends its time on decisions instead of definitions.
When the calendar slips
It always slips: someone is ill, a launch moves, a review takes twice as long as planned. The response we settled on is deliberately unheroic.
- Drop the buffer piece first, never the piece with a deadline attached to it.
- Move publishing dates in the calendar the same day the delay is known.
- Tell the reviewer before the writer, because the reviewer is the bottleneck anyway.
- Publish one piece properly rather than two pieces that skip the review.
- Write down the cause of the slip in one line, then fix the cause, not the symptom.
- Protect the review slot even when the publishing slot moves, because unpublished drafts are cheap and unreviewed ones are not.
Three slips in a row usually mean the calendar was optimistic rather than that the team got slower, and the correction is to shrink the plan rather than to add evenings.
Keeping it running when someone is away
Small teams fail quietly when knowledge lives in one person's head. The fix is boring: shared checklists for publishing, a documented review order, and ownership of each metric named on a single page. If you can hand a colleague the whole process in ten minutes, the team is smaller than it looks but far harder to stop.
The next useful step is to write down who signs off each stage of your own publishing, because most delays are not writing problems, they are approval problems, and naming the approver usually shortens the week by a day.