Content Production

Content plans do not fail on strategy

They fail in month three. The plan is agreed, the first few pieces go out, and then a busy quarter arrives and publishing stops. Six months later the strategy is blamed and a new agency is appointed to write a new one.

The failure is operational. The plan assumed people with full-time jobs would find several hours a week to write, and they did not, because writing was never their job and nothing was taken off their desk to make room.

Production is the half of content marketing that nobody sells, and it is the half that determines whether anything gets published.

What a production system actually consists of

• A brief that says what the page must cover, who it is for, what it must not claim, and what it should link to
• A source of substance — usually thirty minutes of someone's expertise, captured rather than requested in writing
• A drafting step that turns that into a page
• An editorial pass against a house style, so twelve pieces by four people read as one organisation
• A factual review by whoever is accountable for being wrong in public
• Publishing, with the internal links and metadata done rather than deferred
• A refresh schedule, because pages decay

Every stage has a named owner and a realistic turnaround. A pipeline where one stage depends on an unspecified person finding time is a pipeline that stops at that stage.

The brief is the whole leverage point

A weak brief produces a draft that needs rewriting, which costs more than writing it properly would have. A good one makes the draft nearly right first time.

Ours specify the search intent being served, the questions the page must answer, the depth expected, the internal links in and out, the terminology to use and avoid, and the claims that are off limits. For regulated clients that last item is not optional — it is what makes the draft approvable.

Getting expertise out of busy people

The single change that rescues most content programmes: stop asking experts to write, and start interviewing them.

A half-hour recorded conversation about work someone does every day produces more usable substance than three weeks of waiting for a document that never arrives. They talk, we transcribe, we draft, they correct. Correcting a draft takes ten minutes and people reliably do it. Producing one from nothing takes hours and people reliably do not.

This is also what makes content distinctive. The specifics that separate a page from the twenty generic ones competing with it live in someone's head and come out in conversation.

Refreshing usually beats publishing

Most sites with a few years of content have more to gain from improving what exists than from adding to it. Pages that once ranked and have slipped, pages covering a topic that has moved on, near-duplicates competing with each other.

Updating a page that already has history and links is faster than earning both from scratch. We would typically split effort between new pages and a standing refresh cycle rather than treating the archive as finished.

Where we use AI, and where we do not

Worth being direct, because most agencies are not. We use it for transcription, for structuring interview notes, and for first-pass research. We do not publish generated articles.

The reason is practical rather than principled: generated copy is fluent, average, and contains no information that was not already on the internet. It reads acceptably and gives a reader nothing to remember. Content only works when it contains something the competing pages do not have, and that has to come from a person who knows the subject.

What we do

• Brief writing against a keyword and intent map
• Subject-matter interviews, recorded and transcribed
• Drafting, editing and house style
• Internal linking as part of publishing, not a later audit
• Publishing into your CMS, or handover-ready files
• Refresh cycles on existing pages, feeding the wider content strategy
• A calendar sized to what your team can sustain

Cadence honestly

We would rather agree two well-made pages a month for a year than eight a month for two months. Search rewards consistency and depth, and an archive of twenty-four solid pages beats sixteen rushed ones followed by silence.

If your team can only support one interview a fortnight, that is the input the plan is built around. A plan that requires more attention than you have is not a more ambitious plan, it is a plan that will stop.