A list of topics is not an editorial plan. “AI,” “email,” or “SEO” names a subject; it does not specify what a reader needs to decide or do.
Start with a real audience task: compare options, diagnose a failure, complete a process, reduce a risk, or explain a result. Gather wording from support questions, sales conversations, search queries, community discussions, or product use—without exposing private information. Separate verified observations from assumptions.
For every asset, write a compact brief: 1. Reader and context. 2. Question or task to resolve. 3. One-sentence answer or promise. 4. Evidence, example, or method that makes it distinct. 5. Limits and what the material cannot prove. 6. Primary destination and useful supporting links. 7. How it will be adapted for each distribution channel. 8. A review date or condition that would make it stale.
Create one substantial source asset before deriving smaller formats. A guide can become a checklist, demonstration, email lesson, or discussion prompt—but each adaptation should fit the platform and add context. Do not publish near-identical copies just to multiply URLs. That creates editorial clutter and weakens the reader’s reason to choose one version.
A human review should check accuracy, permissions, clarity, and whether an example is real or synthetic. Label synthetic examples honestly. Measure the content’s intended job—qualified discovery, useful completion, informed contact, or reduced support friction—not just how many pieces were produced.
Arabic content needs transcreation and claim review
A literal English-to-Arabic translation can retain a metaphor that sounds unnatural or shifts the meaning. Rewrite for the target reader, then check dialect consistency and grammatical address. For claims—especially health, beauty, finance, or measurable performance—keep a source and reviewer beside every statement that could affect a decision. A synthetic scenario is useful for demonstrating a workflow only when clearly labeled; it is never a client case study or measured outcome.
Turn an interview into a publishable question
Ask customers to reconstruct the last time they tried to solve the task: what triggered it, what they searched, what they misunderstood, what they compared, and what made them trust the next step. Capture exact phrases with permission and remove identifying details. Code the notes by task stage, not by demographic stereotype. Then publish one answer that resolves a real decision and link to the next useful resource. Do not treat one interview as proof of market-wide demand; use it to form a hypothesis to check against search data and additional conversations.
Optional AI-assisted work brief — Customer language synthesis
Using these consented, de-identified interview notes, group exact customer questions by task stage. Preserve useful wording, distinguish quotes from interpretation, and turn the strongest unresolved question into an article brief with evidence to gather and a next-step resource. Do not generalize from a small sample.
Three honest hooks for one useful article
Problem hook
“Your report says 18 leads. How many were actually eligible?”
Contrast hook
“A click, a chat, and a qualified request are three different events.”
Artifact hook
“Here is a four-state worksheet to find where the contact path breaks.”
Payoff rule
Use only the opening the article genuinely answers; never imply a measured portfolio result without evidence.
Worked scenario — Interview language becomes a useful asset
A customer says, “I can see the WhatsApp click in the report, but I do not know if anyone answered.” That exact gap can shape a guide about state reconciliation and inbox ownership. Ask follow-up questions about the last report they reviewed and what decision they needed. One quote inspires a hypothesis; it is not evidence that all customers share it.
AI disclosure: AI tools assisted with drafting. This guide presents a general method and does not claim a verified client result.