# Content Strategy Playbook > The 12-month content operating system EntireCommerce AI designs for premium DTC brands. Runs from a single Claude Code prompt. Produces one consolidated strategy document: ICP-language audit, positioning synthesis, 4 to 8 topic pillars, cluster architecture, channel mix, publishing cadence, production workflow, distribution amplification, and KPI framework. Feeds the 30-60-90 publishing roadmap directly into the client's `actions.md`. --- ## Purpose Design the 12-month publishing operating system for a premium DTC brand from first principles. Starts from three inputs the data cannot reveal: the ICP's actual words, the founder's positioning hypothesis, and the competitive whitespace. Outputs the 4 to 8 topic pillars the brand should own, the cluster architecture underneath each pillar, the channel mix, the cadence, the production workflow, and the leading and lagging indicators that say whether it is working. This playbook sits upstream of every tactical content workflow. The content-gap playbook mines data to find holes in an existing programme. This playbook decides what the programme should look like in the first place: which pillars to plant a flag on, which channels to invest in, how often to publish, who produces each piece, and which metrics prove the system is compounding. ## When to run it - **Week 2 of a new engagement.** Follows the Initial GTM Audit. The audit surfaces the binding constraint. When that constraint is "content programme missing or incoherent", this playbook designs the replacement. - **Annual planning cycle.** Re-run at the start of each 12-month horizon. Pillars shift as the category matures and the brand's authority deepens. - **Major positioning change.** New product line, new ICP, new category bet. Pillars have to track the brand's defensible territory. - **Greenfield launch.** Pre-revenue brand with no content yet. This playbook produces the minimum-viable content plan for the first 90 days. ## Who this is for (ICP) Premium DTC brands, AOV $500+ or 2x category median, founder-led, US or English-speaking European markets, with a team or budget capable of publishing 2 to 4 pieces a week across at least two channels. Founder-led brands still deciding where to invest, and fractional CMOs setting a 12-month content plan across multiple portfolio brands, are the two core personas. --- ## Two paths, four modes The playbook flexes on two independent dimensions. ### Axis 1. Content maturity Does the brand have an existing content programme to reset, or is it greenfield? - **Path A. Existing programme.** Blog, social presence, email list, or podcast already live. Needs a strategic reset: which pillars to keep, which to kill, which to add. - **Path B. Greenfield.** Pre-launch, pre-content, or content published ad-hoc with no architecture. Needs a starting point and a minimum-viable plan. ### Axis 2. Scope - **Full.** All eight sections. Used on a 12-month strategic reset or annual planning cycle. 30 to 60 minutes of run time. - **Quick.** Pillar selection plus cadence only. Skips production workflow and distribution depth. 30 minutes. Used when the founder needs the pillars on a page before a strategy offsite. ### The four modes | Mode | Path | Scope | When to use | |---|---|---|---| | Full Reset | A | Full | Existing programme, 12-month strategic refresh | | Refresh Quick | A | Quick | Existing programme, pillar-only reset before a deeper dive | | Launch Full | B | Full | Greenfield, pre-launch or first 90 days, full operating-system design | | Launch Quick | B | Quick | Greenfield, minimum-viable content plan before the first publish | Mode detection happens at kick-off. Check `CLAUDE.md` for existing content inventory. If the brand lists active blog URLs, an email programme, or a published social cadence, default to Path A. Otherwise Path B. Confirm with the founder. --- ## Tool-availability gate The playbook runs in two layers. The External layer always runs. The Internal layer runs only when logged-in tool access has been granted. ### External layer (operator-side API keys, public data, always runs) | Tool | Used for | Data source | |---|---|---| | Ahrefs | Competitor content footprint, topic volume, backlink authority | Public SEO graph | | DataForSEO | Keyword volumes, SERP intent mapping per pillar | Public SERPs | | Keywords Everywhere | Seed-term volumes for pillar and cluster sizing | Public keyword data | | Serper | Live SERP snapshots, AI Overview presence per pillar | Public SERPs | | Google Trends | Pillar seasonality, emerging topic momentum | Public trend data | | Public ICP research (Reddit, X, Quora, niche forums, YouTube comments) | 15-quote ICP-language audit | Public conversations | ### Internal layer (logged-in tool access, runs in Path A only) | Tool | Used for | Access required | |---|---|---| | Google Search Console | Existing-page performance, queries already winning | Restricted access | | GA4 | Content page engagement, time-on-page, conversion attribution | Viewer access | | Shopify Admin API | Category and product inventory for commerce-linked pillars | Custom app token | | Klaviyo or Brevo | Email cadence, sent volume, open and click patterns | API key | The Internal layer makes Path A sharper. Path B does not need it. --- ## Inputs the playbook accepts | Input | Required | Example | Notes | |---|---|---|---| | Brand name and URL | Yes | `Riyanika Jewels, riyanikajewels.com` | Used in output paths and headings | | ICP in one line | Yes | `British-Indian women 30-55 buying lab-grown diamond jewellery for self-gifting` | Feeds Section 1 ICP-language audit | | Founder positioning hypothesis | Yes | `Lab-grown diamonds at a fraction of mined-stone cost with heritage Indian design` | Feeds Section 2 positioning synthesis | | Current content inventory | Path A | `Shopify blog 12 posts, Instagram 8K followers, email list 3K` | Skipped on Path B | | 3 to 5 competitor URLs | Yes | `diamonfire.co.uk, astteria.com, pandora.net` | Feeds competitor whitespace analysis | | Channels currently active | Path A | `Blog, Instagram, email` | Skipped on Path B | | Team size and content budget | Yes | `2 people, £3K/month production budget` | Feeds Section 6 production workflow | | Publishing cadence goal | Yes | `2 to 4 pieces a week, minimum 1 long-form per week` | Sizes the plan | | 12-month revenue or audience target | Yes | `£500K revenue, 25K email list, 50K Instagram` | Feeds KPI framework | If any required input is missing, ask the user. Never proceed with invented inputs. --- ## Output: one consolidated document, two formats ``` {client_folder}/docs/content-strategy/{YYYY-MM-DD}/content-strategy.md (canonical) {client_folder}/docs/content-strategy/{YYYY-MM-DD}/content-strategy.docx (Word deliverable) ``` Supplementary evidence (15 ICP quotes, keyword pulls per pillar, competitor crawl notes, cluster-level sub-topic tables) lives at: ``` {client_folder}/data/content-strategy/{YYYY-MM-DD}/ ``` Length target: 5,000 to 9,000 words on Full mode, 2,000 to 3,500 words on Quick mode. ### Document structure (fixed order) ``` # Content Strategy: {Brand Name} ({YYYY-MM-DD}) Mode: {Full Reset / Refresh Quick / Launch Full / Launch Quick} Path: {A existing / B greenfield} ## Executive summary 3 to 4 sentences naming the pillar commitment, the channel bet, and the cadence the team will hold. ## Cross-cutting themes 3 to 5 bullets synthesising insights across sections. Produced by the Synthesis pass. ## The three highest-leverage moves for the next 90 days Top 3 actions from Section 8 surfaced at the top with a one-line "why this first". ## 1. ICP-language audit ## 2. Positioning and differentiation synthesis ## 3. Topic pillar selection ## 4. Cluster architecture per pillar ## 5. Channel mix and cadence ## 6. Production workflow ## 7. Distribution amplification ## 8. KPI framework ## 9. ICE action list and 30-60-90 publishing roadmap ## Conclusion 2 to 3 paragraphs. Booking CTA link. ``` --- ## Section 1. ICP-language audit Section 1 runs the canonical audience-insights research prompt at `EntireCommerce/playbooks/audience-insights-master-prompt.md`. Do not rewrite that prompt inline. Load it, feed it the three required inputs (target audience, #1 struggle, what the brand sells), and capture the 15 verbatim quotes from Reddit, X, Quora, niche forums, and YouTube comments. After the 15-quote pass, layer the content-strategy-specific synthesis: - **Five to ten recurring themes.** What do the quotes keep coming back to? Name each theme in 3 to 5 words. Tie each to 2 or more supporting quotes. - **Themes mapped to commercial intent.** For each theme, classify as informational, comparison, transactional, or brand-aware. This classification decides which channel carries which theme later. - **Themes ranked by emotional charge.** The themes that produce visible emotion in the quotes carry most of the content weight. Rank them high-to-low. Save the 15-quote appendix to `{data_folder}/icp-voc-mining.md`. Cross-link from Section 1. ### Quality bar - Every theme traces to quotes in the appendix. - No paraphrased clichés. Use the audience's own words. - Recency: last 12 to 24 months. Flag anything older. --- ## Section 2. Positioning and differentiation synthesis Three inputs land together in this section: 1. The founder's positioning hypothesis from `CLAUDE.md`. 2. The five to ten recurring ICP themes from Section 1. 3. The competitive whitespace from a light competitor content-footprint scan (Ahrefs + Serper against the 3 to 5 competitor URLs). The synthesis produces: - **The defensible territory.** One sentence describing the category intersection the brand can own. Must sit at the intersection of what the ICP cares about and what competitors are ignoring. - **Three proof points.** Concrete evidence the brand owns the territory: product claim, founder expertise, customer outcome, data, or asset. Each proof point traces to a specific source. - **The belief shift.** What mental-model change the buyer must make before buying. State what the audience currently believes, state what the brand is teaching them to believe. - **Category claim.** The emerging-category term the brand wants to own (for example "lab-grown diamond heritage jewellery", "launch-monitor-backed short-game training"). Include even if current search volume is zero. Claim the category before it becomes crowded. ### Quality bar - The defensible territory cannot be "premium quality" or "best service". Specific only. - The belief shift must be testable. If the buyer's mental model is not visibly different after 12 months of content, the belief shift is not real. --- ## Section 3. Topic pillar selection **Step 0 — competitor-set + benchmarking opt-in.** Run the competitor-set decision tree from `PLAYBOOK-TEMPLATE.md`. If a competitor set exists, render `BT-CONTENT-publishing-cadence.md` (posts/month, total post count, content depth, internal-linking density) alongside the pillar selection — pillar prioritisation reads better when the reader can see how the brand stacks up on content volume against the field. Skip cleanly if the user opts out. **Upstream inputs.** Pillar selection is strongest when an industry keyword universe is already available. When the brand has a prior GTM audit (`industry-keyword-universe.csv` + `industry-clusters-page-map.md`), SEO audit, or content-gap audit in its data folder, read those artefacts first. The clusters are the pillar candidates, the intent tags feed the commercial-intent test, the opportunity scores feed brand-authority prioritisation. When those artefacts are missing, Section 3 here runs the consolidation inline. The GTM/SEO/content-gap playbooks are the canonical home of that step. Pillars are the 4 to 8 categories the brand commits to owning over 12 months. Each pillar must satisfy three tests: 1. **ICP relevance.** Traces to at least one recurring theme from Section 1. 2. **Commercial intent.** At least 30% of the pillar's cluster terms carry commercial or transactional intent. Pure-informational pillars starve the pipeline. 3. **Brand authority.** The brand has or can build credible authority in this pillar. Founder expertise, proprietary data, customer case studies, or category experience. ### Pillar sizing - **Exactly 4 to 8 pillars.** Three or fewer cannot carry 12 months of content without cannibalisation. Nine or more dilutes focus and no pillar accumulates authority. - **One head pillar.** The single strongest intersection of ICP, commercial intent, and brand authority. Gets 30 to 40% of publishing volume. - **Two to four body pillars.** Strong on two of the three tests. Each gets 15 to 25% of publishing volume. - **One to three edge pillars.** Strong on brand authority only, low immediate commercial volume. 5 to 10% of publishing volume each. These are long-tail authority bets. ### Pillar evidence package (per pillar) Each selected pillar gets a one-page evidence block in the report: - Pillar name (2 to 5 words) - ICP themes it addresses (from Section 1) - Competitive density (how many of the 3 to 5 competitors are publishing in this pillar, and how deep) - Search volume band (total monthly volume across seed terms) - Commercial-intent ratio (approx % of pillar terms with buyer intent) - Brand authority evidence (why this brand can win here) - Reject reasoning for discarded candidates (keep this visible; shows the decision was deliberate) --- ## Section 4. Cluster architecture per pillar Each pillar decomposes into 8 to 20 sub-topics. The cluster architecture is the routing map telling every future piece of content where it lives. ### Per-pillar output - **Pillar page.** One cornerstone URL owning the head term. Target the pillar's strongest commercial-intent keyword. 2,000+ words. Gets internal links from every cluster page underneath. - **Cluster pages.** 8 to 20 supporting pieces, each targeting one sub-topic. Linked up to the pillar page. Clustered around intent: informational cluster (guides, explainers), comparison cluster (versus pages, alternatives), transactional cluster (product-led, buying guides), and AEO cluster (FAQ hubs, definition pages, direct-answer blocks). - **Supporting formats.** Each cluster gets a supporting-format mix: long-form blog, short-form social, YouTube Shorts, newsletter segment, podcast episode. Not every cluster needs every format. Specify which formats carry which clusters. ### Cluster scoring Every sub-topic scores on: - Search volume (1 to 10) - Difficulty (1 to 10, inverse: lower difficulty scores higher) - Commercial intent (1 to 10) - ICP theme alignment (1 to 10, from Section 1) Sum. Rank sub-topics within the pillar. Top-ranked sub-topics ship first. ### Output artefact A cluster table per pillar, saved to `{data_folder}/pillars/{pillar-slug}-clusters.csv` with columns: sub-topic, target keyword, volume, difficulty, intent, theme alignment, format, priority rank. --- ## Section 5. Channel mix and cadence The channels the brand commits to, each with a role, cadence, and format mix. ### Channel inventory | Channel | Role | Cadence range | Format | |---|---|---|---| | Blog (SEO + AEO) | Compounding organic traffic, pillar authority, AEO citations | 1 to 3 long-form per week | 1,500 to 3,000 word articles, FAQ hubs, comparison pages | | LinkedIn (founder voice) | Operator reputation, inbound, B2B-side relationships | 3 to 5 posts per week | Long-form single posts, text, screenshots, short video | | X (founder voice) | Real-time positioning, community, media amplification | 5 to 10 posts per week | Single posts, threads only for strong narrative arcs | | YouTube Shorts | Short-form video reach, hook-testing | 2 to 5 per week | 30 to 60 second vertical video, UGC or founder talking-head | | Podcast guest appearances | Borrowed audiences, authority transfer | 2 to 4 per month | 45 to 60 minute interviews on niche-relevant shows | | Email newsletter | Owned audience, retention, conversion | Weekly | 300 to 800 word founder-voice letter | | Community (Discord, Circle, private) | Customer depth, qualitative signal, retention | Daily presence | Threaded conversations, AMA cadence, member-generated content | ### Channel selection rules - **Pick no more than 4 channels to start.** Cadence discipline beats channel sprawl. Four channels held for 12 months beats seven channels dropped after three. - **One compounding channel minimum.** Blog or YouTube. Both if budget allows. The compounding channels are the long-term asset. - **One founder-voice channel minimum.** LinkedIn or X. This is where the belief shift from Section 2 gets preached. - **One owned channel minimum.** Email newsletter or community. Algorithms shift; the owned channel survives. ### Channel role assignment For each selected channel, specify: - **Role.** What this channel is optimising for (reach, authority, retention, conversion). - **Pillar allocation.** Which pillars this channel carries. Not every channel carries every pillar. - **Format mix.** Which formats run on this channel (long-form, short-form, video, text). - **Cadence.** Concrete per-week numbers. A single integer per channel. - **Example post types.** 3 to 5 example formats with one-line descriptions. --- ## Section 6. Production workflow The operating system that turns the plan into shipped work. ### Roles - **Founder or subject-matter expert.** Supplies the point of view, the stories, the positioning. Spends 2 to 4 hours a week on content. This is the irreducible founder time. - **Writer or producer.** Turns the founder's raw material into polished drafts. Full-time, fractional, or Claude-assisted. - **Editor.** Enforces voice, fact-checks claims, catches AI clichés and negative parallelisms. Often the same person as the writer on small teams. - **Video producer.** Edits short-form video, captures B-roll, manages UGC submissions. Often fractional or vendor. - **AI assistant (Claude Code).** Handles research, first drafts, repurposing, cluster mapping, keyword analysis, schema markup. Reduces writer throughput time by 3 to 5x. ### Workflow stages Each piece of content moves through these stages: 1. **Brief.** Claude Code builds the brief from the cluster table: target keyword, pillar, ICP theme, source quotes from the 15-quote appendix, competitive SERP snapshot. 2. **Draft.** Writer drafts in Claude Code against the brief. One brief produces one draft. 3. **Edit.** Editor runs the voice gate (em-dash scan, negative-parallelism scan, four-word-sentence floor), fact-checks claims, tightens the opening. 4. **Approve.** Founder reads the final draft, approves or redirects. Single-round approval is the goal. 5. **Ship.** Publish, schema-mark, internal-link to the pillar page, trigger the distribution ladder. 6. **Measure.** Ingest into the weekly-review dashboard. Leading indicators at 14 days, lagging at 90. ### Tooling - **Brief + draft.** Claude Code with the brand's `CLAUDE.md` loaded. - **Editorial calendar.** Airtable or Notion. One row per planned piece, columns for pillar, cluster, channel, status, owner, publish date. - **Publishing.** Shopify blog, WordPress, Ghost, or equivalent CMS. - **Distribution.** Buffer, Hypefury, or native scheduler per channel. - **Measurement.** GA4 + GSC + Clarity feeding the daily dashboard. ### Throughput and cost Given one founder (2 to 4 hours/week) + one writer (20 to 30 hours/week) + Claude Code, a typical premium DTC brand ships: - 2 long-form articles per week - 4 short-form posts per channel per week (so 8 to 12 posts across two founder-voice channels) - 1 weekly newsletter - 2 to 4 short-form videos per week Total monthly cost: £3K to £8K depending on whether writer is fractional or full-time. Lower if the founder writes. --- ## Section 7. Distribution amplification Every long-form piece earns its compounding traffic by being the seed for 5 or more downstream assets. ### The repurposing ladder One long-form piece (2,000 words or a 45-minute podcast) cascades into: 1. **One newsletter segment.** 300 to 600 words adapting the core argument for the owned audience. 2. **Three to five LinkedIn posts.** One per sub-point from the long-form. Founder voice, text-first. 3. **Five to ten X posts.** One per quotable line, one per contrarian take, one per data point. 4. **Two to three short-form videos.** YouTube Shorts, Reels, TikTok. 30 to 60 seconds each. 5. **One infographic or carousel.** Instagram, LinkedIn carousel, or Pinterest pin. Budget 2 hours of writer time per long-form for the repurposing pass. Claude Code accelerates each step. ### Syndication partners - **Podcast guest appearances.** 2 to 4 per month. Source via the target ICP's listening habits from Section 1. - **Newsletter swaps.** Cross-post with 3 to 5 non-competing newsletters in adjacent niches. One per month. - **Guest articles.** Contribute one long-form piece per quarter to a higher-authority publication in the category. - **Affiliate and creator network.** Seed product or commission-based partnerships with 5 to 20 creators in the niche. Each creator's audience is a distribution channel. ### DM-lead-capture comment playbook Every social post with a downloadable asset uses the comment-to-DM capture flow. The post ends with "Comment {TRIGGER_WORD} and I'll DM you the link." This converts passive engagement into a traceable lead. - Pick one trigger word per asset. Unique across the whole content library. - DM the link within 24 hours. Automation via tools like TapLio or ManyChat if volume grows. - Log trigger-word conversions in the daily dashboard. --- ## Section 8. KPI framework Two sets of metrics. Leading indicators track whether the machine is running. Lagging indicators track whether the machine is producing revenue. ### Leading indicators (weekly review) - **Publishing velocity.** Pieces shipped per week, by channel. Target holds the cadence from Section 5. - **Audience growth.** Net new subscribers on each channel. Report the absolute number. - **Time on page.** Average engaged time per pillar page. Target 2+ minutes on long-form. - **CTR on organic listings.** Click-through rate from GSC impressions. Target above 3% on page-1 rankings. - **Comment and reply rate.** Engagement depth on founder-voice channels. Rising = belief shift landing. - **DM-trigger conversions.** Trigger-word DMs captured per post. ### Lagging indicators (monthly review) - **Organic traffic.** Total monthly sessions to pillar and cluster pages. 90-day rolling view. - **AEO citations.** Count of queries where the brand is cited across ChatGPT, Claude, Perplexity, and Google AI Overview. Re-run monthly. - **Qualified leads or signups.** Email captures, demo requests, product trials attributable to content. - **LTV of content-sourced customers.** Compare to paid-sourced LTV. Content-sourced should win over time. - **Pillar-level revenue attribution.** Revenue attributable to each pillar via GA4 + Klaviyo + Shopify. Re-balances the pillar allocation mix quarterly. ### Review cadence - **Weekly (10 min).** Leading indicators on the daily dashboard. Kill or double down on anything visibly breaking. - **Monthly (30 min).** Lagging indicators plus cadence adherence. Adjust the editorial calendar. - **Quarterly (2 hours).** Pillar-level performance. Re-balance allocation. Kill a pillar if it fails 2 quarters in a row. - **Annually (half day).** Full playbook re-run. New pillars claimed, old pillars retired. --- ## Synthesis pass (mandatory before writing to disk) After drafting all eight sections in first-pass form, do not write to disk yet. Assemble the complete draft as a single document in your working context and read it end-to-end as one block. Ask: - What pattern crosses three or more sections that no single section captures alone? - Does an ICP theme from Section 1 point at a pillar choice in Section 3 that the founder's positioning hypothesis in Section 2 understates? - Does the channel mix in Section 5 underuse a format the ICP research in Section 1 says the audience prefers? - Does the production workflow in Section 6 bottleneck a cadence the Section 5 plan requires? - Does the KPI framework in Section 8 fail to measure the belief shift from Section 2? Produce 3 to 5 cross-cutting insight bullets. Revise: - The **Executive summary** to lead with the narrative these insights reveal. - **Section 9** (ICE action list) to promote actions that hit multiple leverage points. - Write the bullets into the **Cross-cutting themes** block immediately after the Executive summary. - Generate the **top 3 highest-leverage moves for the next 90 days** block. - Draft the **Conclusion** block. Only after the synthesis pass completes do you write the final file. --- ## Section 9. ICE action list and 30-60-90 publishing roadmap Every move from Sections 3 through 7 converts into a candidate action. Score each on ICE: - Impact (1 to 10): expected revenue or strategic weight - Confidence (1 to 10): how sure the action works - Effort (1 to 10, inverse): lower effort = higher score Multiply. Rank descending. Group into three buckets: - **Days 1 to 30.** Foundation. Pillar pages drafted, editorial calendar live, first cadence held for 4 weeks, 15 ICP quotes captured and made accessible to the writer. - **Days 31 to 60.** Momentum. Cluster pages shipping, repurposing ladder running, first distribution syndication partner landed, leading indicators trending. - **Days 61 to 90.** Compounding. Pillar page organic traffic starting to show, first AEO citation captured, founder-voice channel hitting target cadence, owned-audience growth verified. Use bucket names in the output. No P0/P1/P2 labels in the deliverable. ### Handoff to actions.md Merge the Days 1 to 30 items into `{client_folder}/actions.md` as new Next Actions. Respect the priority caps (P0 max 5, P1 max 10, P2 max 15). Rescore if any tier is over cap. --- ## Voice enforcement gate (mandatory before writing to disk) Before writing the file, run a mechanical voice-scan on the full draft. Grep for each banned pattern and rewrite every match: | Banned pattern | Grep for | Fix | |---|---|---| | Em-dash | `—` (U+2014) or `--` used as em-dash | Replace with period, comma, or colon | | Negative parallelism | `, not `, ` rather than `, ` instead of `, `opposite of `, `not only ` | Rewrite to state the positive claim alone | | Negative parallelism (contraction-led, sneaky form) | `\bisn't\b`, `\baren't\b`, `\bwasn't\b`, `\bdoesn't\b`, `\bdon't\b`, `\bdidn't\b`, `\bwon't\b`, `\bcan't\b`, `\bhasn't\b`, `\bhaven't\b`, plus spelled-out `is not`, `are not`, `was not`, `does not`, `do not`, `has not`, `have not` | For each match: load-bearing fact (e.g. "AI engines do not run JavaScript") → leave; setup-and-contrast pattern (e.g. "This isn't urgent because X. It's urgent because Y") → drop the negation half, lead with the positive | | Banned spelling | `e-commerce`, `E-commerce`, `E-Commerce` | Replace with `ecommerce` | | AI clichés | `Here's the`, `Here's what`, `Here's why`, `Most people`, `The uncomfortable truth`, `The brutal truth`, `The breakthrough` | Rewrite without the cliché frame | | Sub-four-word sentence | Sentences of 1 to 3 words standing alone | Merge into neighbour or expand | `not` is permitted only when the negation is genuinely load-bearing. Default to rewriting. The voice gate must hit zero matches before the deliverable is written to disk. --- ## Voice conventions - No em-dashes. Use period, comma, or colon. - No negative parallelisms. State the positive alone. - Four-word sentence floor. - "ecommerce" one word. Never "e-commerce". - No AI clichés. - Every claim traces to evidence (ICP quote, keyword data, competitor crawl, founder input). - Jargon gate: avoid shorthand (P0, binding constraint, highest-ICE, ICP, MMM-lite) in the deliverable. Use plain English equivalents (top priority, biggest bottleneck, highest-leverage, ideal customer). --- ## Output formatting: bullets over paragraphs (mandatory for every report and summary) When this playbook generates a report, executive summary, brief, or any deliverable with multiple findings or actions, the default output structure is bullet lists, not prose paragraphs. - Any time the text enumerates three or more items (findings, fixes, gaps, recommendations, criteria, things-they-do-better) those items become separate bullets. Do not bury enumerated points inside a paragraph. - Reserve paragraphs for genuinely connected reasoning where prose flow carries the argument. Cap those paragraphs at four sentences. - Multi-clause sentences with semi-colons or commas separating discrete points are a signal to break into bullets. - The canonical pattern is: heading + one lead-in sentence + bullet list. The lead-in frames the list; the bullets carry the load. - Reports ship scannable. The reader skims on a phone between meetings. Run a paragraph-density check before commit: any paragraph longer than four sentences, or containing three or more enumerated items inside prose, is a candidate for bullet conversion. Convert it. Applies to: full reports, executive summaries, email overviews, weekly reviews, dashboards, action-item lists, and any narrative deliverable. Does not apply to: opening framing paragraphs (one paragraph allowed), conclusion paragraphs (one or two allowed), or quoted founder/buyer voice. ## Reference material - `audience-insights-master-prompt.md`: canonical audience research prompt, runs as Section 1. - `gtm-audit.md`: parent playbook feeding this one. - `seo-audit.md`: technical SEO audit feeding Section 4 cluster architecture. - `content-gap.md`: data-driven content-gap audit, downstream workflow. - `competitor-creative-audit.md`: competitor creative teardown, feeds distribution benchmarks. --- ## Acceptance criteria - Runs end-to-end from a single Claude Code prompt. - Correct mode detection (Full Reset, Refresh Quick, Launch Full, Launch Quick) with user confirmation. - Honours the two-layer tool gate. External layer runs on every mode. - Section 1 conforms to the canonical audience-insights-master-prompt.md: 15-quote appendix with platform, date, emotion, recurring phrase fields. - Section 3 produces 4 to 8 pillars with the evidence package per pillar. - Section 4 produces a cluster table per pillar saved as CSV in the data folder. - Single consolidated document written to the output path. Cross-cutting themes, top-3 moves, and Conclusion blocks populated by the synthesis pass. - Voice enforcement gate run before write. Zero banned-pattern matches in the final file. - Days 1 to 30 items merged into `actions.md` respecting priority caps.