Step‑by‑step prompt frameworks for content marketing planning, including why it works, a ready‑to‑copy prompt, and what to measure next:
90‑Day Content Roadmap Architect
Topic Cluster and Authority Map Builder
Persona‑to‑Journey Message and Objection Matrix
By Daniel Rowles, lead trainer at Target Internet and author of Digital Branding (Kogan Page). Last updated: November 2026.
The best AI prompts for content planning give the AI a clear role, your real goals and data, numbered tasks and a defined output, then ask for two contrasting options so you can choose. Below are five tested copy-and-paste prompts covering a 90-day roadmap, topic clusters, persona messaging, a content calendar and team capacity.
Each prompt follows the same format: what it does, why it works, the prompt itself, and what to measure afterwards. Replace the text [in brackets] with your own details. This guide is part of our series of AI prompt guides for common digital marketing tasks.
AI is good at structuring ideas, spotting gaps and turning a strategy into a schedule. It does not know your audience, your performance data or what your team can realistically deliver unless you tell it. The prompts below work best when you paste in real inputs: a Google Analytics 4 export, your personas, last quarter's content calendar or your Search Console queries.
Three techniques run through every prompt in this guide:

Use a reasoning model, such as the thinking modes in ChatGPT, Claude or Gemini, for the analysis and prioritisation steps. Faster models are fine for spinning out headlines, social posts and tracking links once the plan is agreed. Always review the output: AI can be confidently wrong about numbers it has not been given, and Google's guidance on helpful, people-first content applies however a plan was produced.
What it does: Turns business goals into a prioritised 90-day plan with themes, formats, owners and a short pack to win stakeholder sign-off.
Why this works: It forces a diagnose-then-prioritise approach, ties every idea to a measurable goal, and produces two contrasting plans you can compare before committing. RICE scoring, a prioritisation method developed at Intercom, makes the ranking transparent.
Copy and paste prompt: "Act as a senior content strategist. Build a 90-day content roadmap that aligns to goals, audience needs and channel realities. Inputs: business goal and KPI: [for example, increase qualified demo requests by 20%]; website: [URL]; ideal customer profiles or personas: [list, including market or region]; brand voice: [for example, practical, warm, expert; British English]; priority themes or products: [list]; channels in scope: [website, LinkedIn, email, short video, ads]; resources and constraints: [team roles, budget, compliance notes]; data: [attach GA4, CRM, SEO tool or social analytics exports]; seasonality or events: [list]. Tasks: 1) Ask up to five clarifying questions only if critical inputs are missing; otherwise proceed. 2) Summarise current content performance and obvious gaps by funnel stage. 3) For each persona and stage, list problems, jobs to be done and the proof they need. 4) Generate 25 ideas across formats, tagged by funnel stage, pillar and proof needed. 5) Prioritise with RICE (Reach, Impact, Confidence, Effort) and show a ranked backlog with scores and a one-line rationale. 6) Produce Plan A, a low-risk evergreen cadence, and Plan B, a bold point-of-view plan with 20% of time for experiments, and explain when to choose each. 7) Build a 12-week calendar with owners, draft and publish dates, dependencies and a definition of done. 8) Write a measurement plan with metrics per item, UTM conventions, review cadence and stop criteria. 9) Outline a 10-slide stakeholder pack. Output as tables in British English."
What to measure next: publishing velocity, the percentage of content delivered on time, assisted conversions by theme in GA4, and organic clicks to target pages.

What it does: Designs content pillars, supporting cluster pages and an internal linking plan that builds topical authority while answering real reader questions.
Why this works: It makes the AI state its assumptions about search intent first, compares a search-first structure with a story-first one, and outputs a link plan you can implement in one pass. Pair it with our AI prompts for SEO to feed in real keyword data.
Copy and paste prompt: "Act as an SEO content architect. Create a topic cluster plan and internal link map that balances search demand and brand narrative. Inputs: primary topic or pillar: [for example, HR analytics software]; related entities and subtopics: [list]; geography: [for example, UK and EU]; funnel emphasis: [awareness, consideration or decision]; brand voice: [for example, authoritative and approachable; British English]; constraints: [claims to avoid, regulated terms]; competitor angles to counter: [list if known]; keyword or Search Console data: [paste if available]. Tasks: 1) State the dominant search intent for each subtopic and your confidence level. 2) Map the sub-questions and entities a complete answer needs, marking gaps as high, medium or low priority. 3) Produce Outline A, search-first and optimised for coverage and AI answers, and Outline B, story-first and optimised for differentiation, then recommend one. 4) Plan one pillar page and 6 to 12 cluster pages with target length, a short brief and the proof each needs. 5) Design internal links: pillar to cluster, cluster to pillar, related links and anchor text suggestions. 6) Suggest structured data types. 7) Add editorial guardrails for tone, accessibility and inclusion. Output as tables."
What to measure next: rankings and impressions for the pillar keyword, appearances in AI Overviews and featured snippets, click-through on internal links to the pillar page, and growth in non-brand organic clicks.
What it does: Maps messages, objections and proof to each stage of each persona's journey, then turns the gaps into content briefs.
Why this works: Role play surfaces the questions real buyers ask, and the matrix converts them into prioritised briefs with a clear belief shift for each piece. It works best when you start from well-researched marketing personas rather than asking the AI to invent them.
Copy and paste prompt: "Act as a buyer enablement lead. Build a persona-by-journey content matrix that resolves objections and proves value. Inputs: personas: [for example, HR Director, CFO, People Operations Manager]; desired action and KPI: [for example, book a demo; demo-to-opportunity rate]; brand voice: [for example, practical and empathetic; British English]; known objections: [price, integration, data privacy]; proof available: [case studies, benchmarks, certifications]; legal or compliance constraints: [list]. Tasks: 1) For each persona and stage, list their top questions, anxieties and decision criteria, as if summarising customer interviews. 2) Build a message house: core promise, three to five supporting messages and the proof for each persona. 3) For each persona and stage, propose two content ideas with format, the belief we want to change, call to action and proof. 4) Write two contrasting introductions for one key piece: one high-clarity and scannable, one a narrative hook with authority signals. 5) Prioritise the top 10 pieces using RICE and explain the trade-offs. 6) Write copy-ready briefs for the top five pieces with key points, outline and accessibility notes. Output as tables and bullet points."
What to measure next: progression from one funnel stage to the next in your CRM, content-assisted pipeline, and whether sales teams hear the targeted objections less often over time.
What it does: Blends a steady evergreen cadence with reactive, news-driven content, using a library of triggers and a quick publish-or-pass decision process.
Why this works: It protects the compounding value of evergreen content while still capturing spikes in interest. The decision tree stops teams chasing every trend by making them weigh impact against effort first.
Copy and paste prompt: "Act as a newsroom-style content editor. Create a content calendar that mixes evergreen output with moment-led opportunities. Inputs: evergreen themes: [list]; triggers to monitor: [industry news, product updates, competitor moves, seasonal events]; channels: [website, LinkedIn, short video, email]; team capacity: [for example, six posts a month plus two reactive slots]; brand voice: [for example, confident and helpful; British English]; risk tolerance: [low, medium or high]. Tasks: 1) Propose a sustainable evergreen rhythm by channel. 2) Define a trigger library: what to watch and which alerts or feeds to use. 3) Create a publish-or-pass decision tree that can be applied within 30 minutes, scoring impact, relevance, originality and effort. 4) Provide quick-turn formats and templates for reactive posts. 5) For one sample trigger, write a control treatment, a practical explainer, and a challenger treatment, a bold point of view. 6) Write a weekly run sheet: who monitors what, stand-up questions and sign-off times. Output as checklists and tables."
What to measure next: time to publish for reactive content, reach and engagement on moment-led posts compared with evergreen, follower growth and brand mentions.
What it does: Turns ambition into realistic output by modelling capacity by role, setting turnaround times and defining what "done" means so the plan sticks.
Why this works: It replaces wishful calendars with evidence-based throughput. Comparing a conservative and an aggressive scenario helps secure buy-in and protects quality, which matters when you need stakeholder support for resources.
Copy and paste prompt: "Act as a content operations lead. Build a capacity-aware plan, service levels and governance so the calendar is deliverable. Inputs: team roles and weekly hours: [for example, writer: 12 hours, editor: 6 hours]; external partners or budget: [if any]; required outputs by channel: [for example, three LinkedIn posts and one long-form article a week]; review and compliance needs: [for example, financial services regulations, head of department review]; definition of done: [style, accessibility, inclusion, legal]; brand voice: [British English]; historic data: [attach previous content schedules showing what was actually published]. Tasks: 1) Estimate realistic monthly output by format using historic velocity, or state the defaults you assume. 2) Build a service level matrix for drafting, review, legal, design and publishing. 3) List bottlenecks and mitigations in a risk register. 4) Model Scenario A, conservative and quality-first, and Scenario B, aggressive and speed-first, with trade-offs and a recommendation. 5) Define governance: a RACI, a definition of done checklist and a change-control process. 6) Produce an eight-week schedule with owners, buffers and checkpoints. Output as tables and checklists."
What to measure next: on-time rate, revision rounds per asset, errors caught at quality checks, and cycle time from brief to publication.
Prompts like these are only as good as the strategic thinking behind them. The Target Internet platform helps you build that, with interactive courses across digital marketing and AI, skills benchmarking to show where your gaps are, accredited certification, over 400 episodes of the Digital Marketing Podcast, and monthly live masterclasses.
For more prompts in this series, see our guides to AI prompts for SEO and AI prompts to improve existing content.
If you are ready to take a structured approach to building your digital marketing skills, explore Target Internet membership options and discover how the platform's interactive courses, benchmarking tools, podcast, accredited certification and live masterclasses can help you plan content that delivers results.
Yes. AI can turn your themes, channels and capacity into a dated calendar with owners and deadlines. Give it your real constraints, such as team hours and approval times, and ask it to separate evergreen content from reactive slots, otherwise it tends to produce a plan your team cannot deliver.
Any of the leading assistants, such as ChatGPT, Claude, Gemini or Microsoft Copilot, can run these prompts. For the analysis and prioritisation steps, use a reasoning or thinking mode. The quality of your inputs matters more than the tool you choose.
RICE is a scoring method developed at Intercom that ranks ideas by Reach, Impact, Confidence and Effort. You multiply reach, impact and confidence, then divide by effort. It helps content teams compare very different ideas on the same scale.
Give the AI real data, such as analytics exports, customer questions, personas and past performance, and ask it to diagnose the current situation before planning. Asking for two contrasting options and a rationale for each also reduces bland, middle-of-the-road suggestions.
AI can help draft content, but planning is where it adds most value with least risk. Keep human experts responsible for the final content, add first-hand experience and original insight, and check facts, because search engines and readers reward content that is genuinely useful.
Daniel Rowles is lead trainer at Target Internet and the author of Digital Branding (Kogan Page). He co-hosts The Digital Marketing Podcast and delivers digital marketing and AI training for organisations around the world, including digital transformation and future skills. You can find out more on his Target Internet profile.
1. Intercom (n.d.). RICE: simple prioritization for product managers. https://www.intercom.com/blog/rice-simple-prioritization-for-product-managers/
2. Google Search Central (n.d.). Creating helpful, reliable, people-first content. https://developers.google.com/search/docs/fundamentals/creating-helpful-content
Traditional SEO was about proving relevance to a search algorithm to rank a link. AEO is different - it's about training an AI to understand your brand as a credible "Entity" so it cites you as the answer.