Production as a workflow
A piece of content passes through the same stages whether it is a guide, a video script or a newsletter issue. Naming the stages matters because each has a different owner, a different input and a different definition of done. When a team skips the brief, the editor ends up doing strategy; when it skips review, the reader does the fact-checking; when it skips repurposing, the asset reaches a fraction of the people it could. The table below is the workflow this section assumes. Teams of one still pass through every stage; they simply wear every hat in turn, and the brief is what stops them drifting between the hats.
| Stage | Input | Output | Usually owned by | Guide |
|---|---|---|---|---|
| Brief | Calendar slot, audience research, goal | A one-page brief with question, angle, outline and review criteria | Content lead or strategist | Writing a content brief |
| Research and sourcing | The brief | Notes, sources, quotes and proof points | Writer | Writing a content brief, Interviewing a subject-matter expert |
| Draft | Brief plus research | A complete first draft that follows the outline | Writer, sometimes with AI assistance | AI-assisted content with human review |
| Edit | The draft | A draft that is structured for scanning and reads plainly | Writer, then editor | Writing for the web |
| Review | Edited draft | Fact, source, brand and legal sign-off | Subject expert, editor, sometimes legal | AI-assisted content with human review |
| Build | Approved draft | Formatted page or asset with images, metadata and links | Producer or web editor | Writing for the web |
| Publish and distribute | Built asset | Live asset plus the distribution plan in motion | Channel owners | Distribution hub |
| Repurpose | Published asset | Derivative formats scheduled across channels | Content and social teams | Repurposing content |
| Refresh or retire | Performance review | Updated, merged or removed asset | Content lead | Content governance |
The stages are sequential but not slow. Most of the time in production is spent waiting for review and approval, not writing; the fix is to agree reviewers and turnaround at the brief stage rather than to write faster.
Briefs, writing, repurposing, AI: the core guides
- Writing a content brief is a step-by-step tutorial for the document that prevents rewrites: the goal, the audience and their stage, the exact question the piece answers, the angle, the proof it needs, the outline, the format, the calls to action, the distribution notes and the criteria a reviewer will use. It assumes the calendar exists and the slot has been chosen.
- Writing for the web covers how people read on screens and what that means for structure: front-loaded paragraphs, headings that carry meaning on their own, lists where lists help, plain language, evidence over adjectives, and the accessibility habits that make writing usable by more people. It ends with an editing routine that works pass by pass.
- Repurposing content is the method for getting several assets from one: which source pieces are worth repurposing, the format ladder from long to short, how to adapt rather than copy, how to sequence releases, and how to track derivatives so the source gets credit.
- AI-assisted content with human review sets out a workflow where AI drafts and people decide: prompts built from the brief, fact and source checks, editing for voice and specificity, disclosure, and a review gate nothing passes without. It is deliberately tool-neutral.
The core guides are written to be read in that order the first time, and then used individually. Search-specific craft, such as matching the format a results page rewards or structuring headings for a target query, is owned by the SEO content section; the production guides here apply to every content type, including the ones search will never see.
Further guides
Further guides cover the people a piece depends on and the content types with rules of their own:
- Interviewing a subject-matter expert — a tutorial for turning an expert's knowledge into a publishable piece: briefing the interview, asking for cases and decisions, recording with agreement, drafting and agreeing the byline.
- Writing a customer case study — a tutorial for a customer story people believe: choosing the story, getting written permission, interviewing the customer, checking results, approval and keeping it current.
- Working with freelance writers — finding, trialling and working with freelancers: paid test pieces, briefs and style guides, useful feedback, and rights, AI terms and bylines agreed in writing.
- Writing product reviews and roundups — reviews and best-of roundups readers trust: a testing method, evidence, scoring, disclosure of gifts and affiliate links, and updates.
- Building a glossary for your audience — a tutorial for a glossary readers and search engines can use: terms chosen from real questions, definitions, links to guides and avoiding thin pages.
- Hiring and evaluating a content marketing agency — a guide for handing part of the programme to an agency or studio: what stays with the business, the questions to put in writing, a bounded trial piece, the warning signs, scope, authorship and disclosure, reports that hold up and the handover file to keep from the start.
Quality gates before publishing
A gate is a question with a yes-or-no answer that stops a piece until the answer is yes. Gates sound bureaucratic and are the opposite: they replace vague "does this feel ready" conversations with a short list anyone on the team can apply. Five gates cover most content.
- Does it answer the brief? The question in the brief is answered in the opening, the outline is followed or the deviation is explained, and the call to action matches the stage the brief named. If the piece drifted into a different question, either the brief or the piece is wrong, and it is faster to decide which now than after publication.
- Is every checkable claim checked? Every fact, figure, quote and product detail has a source the reviewer can open. Claims that cannot be sourced are rewritten as opinion or removed. This gate is the one AI-assisted drafts fail most often, which is why the review workflow guide treats it at length.
- Would a subject expert sign it? Google's guidance on helpful content asks whether content shows first-hand experience and expertise, the qualities often summarised as E-E-A-TExperience, Expertise, Authoritativeness, Trust.. The practical test is whether a named person with real experience of the subject is willing to put their name on it; if not, the piece needs their input before it goes out.
- Can a person scan it? Headings describe their sections, the first sentence of each paragraph carries the point, lists are used where items are parallel, and link text says where a link goes. The accessibility checks in the writing guide sit here.
- Is it findable and measurable? Title, description, headings and internal links are in place for search (SEOImproving visibility in unpaid search results.), the page is grouped for reporting, and campaign links carry tracking parameters. A piece that cannot be measured cannot be defended at the next review.
The content governance guide covers who holds each gate and what happens when a gate is skipped under deadline pressure; the production guides cover how to pass them.
Capacity and cadence
Most content calendars fail on capacity, not ideas. A team plans a cadence it would like, discovers in the second month that review and build take longer than drafting, and then either misses slots or publishes unreviewed work. The honest approach is to measure the hours each stage takes for each content type, multiply by the planned cadence and compare the total with the hours the team actually has after meetings, reporting and reactive requests. The content pillars and calendar guide covers sizing the calendar; the production question is how to make the hours go further.
Three levers move capacity without lowering standards. Briefs reduce rewrites, which are the single largest hidden cost. Repurposing turns one researched, reviewed asset into several channel-ready pieces for a fraction of the original effort. AI assistance can shorten the first-draft stage for some content types, provided the review stage is funded properly, because the hours saved in drafting partly move into checking. Where drafting really is the bottleneck, freelance writers add capacity; where review is, more drafts only lengthen the queue. A cadence a team can keep for a year beats a cadence it can keep for a month.
Sizing a small team's cadence
A software company's content team is two people: a writer-editor and a part-time designer who also builds pages. They plan four long guides, four customer stories and eight short posts a month. Timing a month of work shows a long guide takes about 14 hours from brief to publish once review waits are counted, a customer story about 9 hours including interview scheduling, and a short post about 2 hours. The plan therefore needs roughly 108 hours of production a month against the 95 hours the two of them genuinely have for it. They drop to three guides, keep the stories, and repurpose each guide into three of the short posts instead of writing them fresh. The new plan needs about 84 hours, leaves slack for a reactive piece, and holds for the rest of the year.
Common production problems
Most production trouble shows up as one of a handful of symptoms. The table maps each to the guide that deals with it.
| Symptom | Usual cause | Guide |
|---|---|---|
| Drafts come back needing a full rewrite | No brief, or a brief that named a topic but not a question and angle | Writing a content brief |
| Pieces read well but say nothing specific | Evidence and examples were never gathered; adjectives filled the gap | Writing for the web |
| Readers leave before the point | Preamble before the answer; headings that do not describe their sections | Writing for the web |
| Good assets published once and forgotten | No repurposing step in the workflow; derivatives not planned at brief stage | Repurposing content |
| Social and email versions are pasted excerpts | Copying instead of adapting to each channel | Repurposing content |
| Factual errors or unsourced claims reach publication | AI or rushed drafting without a fact and source check | AI-assisted content with human review |
| Everything sounds the same | Drafts not edited for voice; one prompt template reused across topics | AI-assisted content with human review |
| Calendar slots missed every month | Cadence set by ambition, not measured capacity | Content pillars and calendar |
| Pieces take weeks to move from idea to publication | Waits in briefing, review and approval rather than in writing | Content production too slow |
| Nobody can say whether a piece worked | Not grouped for reporting; no tracking on distribution links | Measuring content performance |
| Expert pieces repeat what every other page already says | A draft requested from the expert instead of an interview; generic questions; no follow-up | Interviewing a subject-matter expert |
| Customer stories refused or cut back at approval | Interviewed and drafted before written permission was agreed | Writing a customer case study |
| Freelance drafts need as much editing every time | Commissioned without a brief or style guide; feedback based on taste | Working with freelance writers |
| An outside provider's work cannot be judged, or leaving it would lose the briefs and sources | Strategy, standards and records not kept inside the business; reports read on the provider's own definitions | Hiring and evaluating a content marketing agency |
| Readers do not trust a roundup's winners | Method written after testing; untested products added; disclosure only in the footer | Writing product reviews and roundups |
| Glossary pages that hold a sentence each | A page for every term; terms chosen from a dictionary rather than readers' questions | Building a glossary |
Click-through (CTRShare of impressions that result in a click.) from search and from distribution channels is where production quality first becomes visible outside the team: a title and opening that match what the reader came for earn the click, and a page that delivers earns the next one. The content marketing pillar places production between strategy and distribution; this section is the middle of that system.
