Install and run the Channel Strategy Justification skill

Where to run this

This is an installable AI Skill (Claude Skill format), not a one-off prompt, install it once, then invoke it whenever a channel mix needs a defensible rationale.

Upload first

• ☐Client/brand, market, category and business outcome
• ☐Audience definition and the barrier or behaviour that needs to change
• ☐Campaign duration and available budget or range
• ☐Creative assets available or feasible
• ☐Measurement requirement
• ☐Brand safety, legal or regulatory restrictions
• ☐Any client-preferred or client-rejected channels
• ☐Historical campaign evidence and audience research, if available

Steps

1. 01

   Understand what this skill does before you install it

   Do this

   This is not a media-plan generator, it's the strategy and decision record that should exist before line items, rates or formats get built. It answers eight questions: what behaviour needs to change, what communications tasks media must perform, what channel roles are required, which channels can credibly fill each role, why each recommended channel was picked, why plausible alternatives weren't, what's lost if a channel is removed, and how the recommendation should shift if budget, audience, creative or measurement constraints change. Its core discipline: recommend roles before naming channels, never default to familiar platforms just because they're familiar, and never invent audience, cost, reach or performance evidence. It explicitly connects to the Cross-Channel Reach Investment Planner skill for budget curves and minimum spend, rather than inventing those numbers itself, install both if you need the full planning-to-budget chain.

2. 02

   Install the skill in your AI platform

   ---
   name: channel-strategy-justification
   version: 1.0.0
   description: >
     Builds evidence-led, client-defensible channel recommendations from business outcomes,
     audience behaviour, communications tasks, market conditions, budget, creative and
     measurement realities. Use when a user asks which media channels to select, why a
     channel is needed, why the recommendation resembles a standard media plan, what to
     remove under budget pressure, or how open-web and platform channels should work
     together. Considers the full addressable channel universe rather than defaulting to
     familiar platforms. Never invents audience, cost, reach or performance evidence.
   ---
   
   # How to: Channel Strategy Justification Framework
   
   ## 1. Purpose
   
   Turn a media channel list into a defensible investment thesis. The skill must answer: what behaviour or perception needs to change; what communications tasks must media perform; what channel roles are required; which channels can credibly perform each role; why each recommended channel was selected; why plausible alternatives were not selected; what is lost if a channel is removed; and how the recommendation should change if budget, audience, creative or measurement constraints change.
   
   This is not a media-plan generator. It establishes the strategy and decision record that should exist before detailed allocations, rates, formats or line items are produced.
   
   ## 2. Planning philosophy
   
   Decision sequence: business outcome, then audience barrier or behaviour to change, then communications task, then required media role, then the full candidate-channel universe, then evidence and constraint evaluation, then the recommended channel system, then removal and trade-off analysis, then budget allocation and measurement design. Never begin with a preferred platform, partner, rate card or historically familiar plan.
   
   **Recommend roles before channels.** Bad: "Use YouTube, Meta and TikTok." Better: "Required role: scaled audiovisual reach. Candidate environments: broadcast television, connected television, YouTube, open-web online video, social video and cinema. Selected route: [channel], because [evidence tied to the brief]."
   
   **Familiarity is not evidence.** Do not recommend a channel merely because the client used it previously, the agency commonly buys it, it appears in a competitor's plan, it has a recognisable name, a partner supplied a persuasive sales deck, it has a low headline cost, or it's easy to activate or report. Prior use is only relevant evidence when comparable historical results and limitations are supplied.
   
   **A channel recommendation is a system.** Evaluate how channels work together across reach and incremental audience access, frequency and continuity, attention and message depth, context and mental availability, demand creation and demand capture, physical-world and digital moments, sequencing and creative roles, and measurement and learning. Do not assume more channels automatically improves the plan.
   
   ## 3. Full candidate-channel universe
   
   Consider the full relevant universe before narrowing it. Do not default to search, social and YouTube.
   
   **Open web and programmatic:** standard display, high-impact display and rich media, native advertising, premium publisher direct display, contextual display, open-web in-stream video, open-web outstream/in-read video, premium publisher video, connected television and OTT streaming, FAST (free ad-supported streaming TV) where available, digital audio and music streaming, podcast advertising, digital out-of-home, programmatic DOOH, in-app display and video, mobile web, gaming and in-game advertising, retail media (display/video/off-site/sponsored), commerce and marketplace media, digital screens in transport/retail/office/residential/place-based environments.
   
   **Walled-garden and platform:** YouTube, Meta platforms, TikTok, LinkedIn, Pinterest, Snapchat, Reddit, X (only where brand/audience/safety/market considerations support it), market-specific social/video/messaging/commerce/super-app platforms, app-store and app-promotion inventory.
   
   **Search and intent:** paid search, Shopping and product listing ads, Performance Max where relevant, organic search as a connected non-paid consideration, retail search, marketplace search, map and local discovery. Search should generally be treated as demand capture unless evidence supports a different role, not a substitute for broad awareness reach.
   
   **Offline and traditional:** linear TV, addressable TV, radio, static OOH, print newspapers and magazines, cinema, direct mail, sponsorships, events and experiential, point-of-sale and shopper media, partnerships and branded content.
   
   **Owned, earned and direct** (consider their connected role even outside paid-media investment): CRM and email, messaging, brand website and app, organic social, PR, influencers and creators, community, loyalty channels. Do not assign paid-media reach or cost claims to owned/earned channels without data.
   
   **Market-specific inventory:** the user may supply local publishers, broadcasters, retailers, platforms, transport networks, super-apps or media owners, include them as candidates without forcing them into a global taxonomy that removes their strategic value.
   
   ## 4. Channel versus platform versus buying route
   
   Keep these levels separate: Channel (e.g. connected television), Environment (e.g. broadcaster streaming, premium app, FAST service), Buying route (direct, private marketplace, curated deal, open auction), Technology or partner (a named DSP, SSP or managed-service provider), Format (e.g. 15-second non-skippable video). Do not treat a DSP, SSP, publisher, reseller or managed-service partner as a channel. Do not reject an entire channel because one buying route is weak, consider a stronger supply path, direct buy, private marketplace, curated package or alternate partner instead.
   
   ## 5. Required inputs
   
   **Critical:** client/brand, market(s), category and product/service, business outcome, campaign objective, audience definition, audience barrier/need-state/behaviour to change, campaign duration and key timing, available budget or planning range, creative assets available or feasible, measurement requirement, material brand safety/legal/regulatory/operational restrictions.
   
   **Strongly recommended:** brand maturity and market position, existing demand level, purchase cycle, priority geography, audience media behaviour/research, historical campaign evidence, reach/frequency targets, required calls to action, client-mandated or client-rejected channels, existing commitments/minimum spends, first-party data availability, landing-page/commerce/conversion readiness, budget flexibility, production budget and localisation needs, attribution or experimental design.
   
   **Evidence inputs accepted:** brand/audience research, platform planning outputs, reach-curve outputs, historical campaign reports, market media-consumption studies, publisher/platform/partner data with source and date, rate cards and inventory availability, creative performance evidence, brand-lift/sales-lift/controlled-test results, approved agency benchmarks. Always record source, market, audience, date, methodology and relevance. Do not treat vendor claims as neutral evidence.
   
   ## 6. Missing-input protocol
   
   If critical inputs are missing: do not invent them, label the work "Draft for strategic review," list missing inputs, provide a provisional role architecture only if it can be supported, do not issue numerical scores/budget splits/reach claims/final channel decisions, and provide the input template from Section 22. If the user supplies client-preferred channels but little else, treat them as candidates, not conclusions.
   
   ## 7. Define the strategic problem
   
   Translate the brief into: business outcome, current audience state, desired audience state, primary barrier, required action or perception change, campaign job. Example barriers: insufficient awareness, weak mental availability, limited understanding, low relevance, lack of trust, low consideration, weak differentiation, low intent, friction in action/purchase, inadequate retention or repeat behaviour. Do not assume the barrier from the objective alone.
   
   ## 8. Convert the problem into communications tasks
   
   Select only tasks justified by the brief: build broad awareness, build/refresh memory structures, communicate a distinctive message, demonstrate a product/experience, explain a complex proposition, establish credibility/trust, reach a specialist/hard-to-find audience, create fame or cultural presence, build local physical presence, generate consideration, encourage content engagement, capture existing demand, create qualified traffic, generate a lead or sale, retarget/re-engage, support a launch/event/seasonal moment, maintain continuity/salience, reach customers at a relevant location/moment. Prioritise as primary, supporting, or optional. Too many primary tasks usually indicates an unfocused brief.
   
   ## 9. Define required media roles
   
   Translate tasks into roles before naming channels. Possible roles: scaled reach, incremental reach, efficient frequency, high-attention storytelling, demonstration and education, contextual relevance, cultural participation, physical-world fame, location and moment proximity, specialist-audience access, trust transfer through premium environments, social proof and participation, content discovery, demand capture, commerce conversion, retargeting, sequential messaging, continuity and recency, measurement or learning cell. Every recommended channel must have one primary role, supporting roles are allowed, avoid duplicating roles without a clear reason.
   
   ## 10. Build the candidate set
   
   For each required role: generate candidates from the full universe in Section 3, include at least one credible open-web or non-walled-garden candidate where available, include offline or physical-world media where the objective and market make them relevant, include a no-channel or owned-channel alternative where paid media isn't essential, remove only channels that fail an explicit feasibility gate, and record the reason for removal. Do not pad the candidate list with obviously unsuitable channels solely to imply rigour.
   
   ## 11. Feasibility gates
   
   A candidate must pass all mandatory gates that apply: available in the target market, can reach the defined audience with acceptable confidence, supports the required communications task, compatible with available/producible creative, viable within the campaign duration, viable within budget/minimum-spend constraints, meets legal/regulatory/client-policy requirements, meets brand-safety/suitability requirements, can be measured sufficiently for its assigned role, can be activated operationally by the required deadline, has an acceptable supply/transparency/verification route. If a candidate fails, state whether the failure is fatal, resolvable, or unknown and requiring validation.
   
   ## 12. Evidence hierarchy
   
   In order of weight: (1) comparable, audited or controlled client outcome evidence; (2) comparable first-party historical campaign evidence; (3) independent market and audience research; (4) platform or publisher planning data specific to the market and audience; (5) approved agency benchmarks and operational evidence; (6) vendor research with transparent methodology; (7) expert judgement, clearly labelled. A lower tier is not automatically invalid, it carries more uncertainty and must not be presented as independent proof.
   
   ## 13. Evaluation dimensions
   
   Evaluate every serious candidate against the same relevant dimensions: **Strategic fit** (business outcome, audience barrier, communications task, funnel role, message depth). **Audience fit** (availability, quality/relevance, scale potential, incremental audience potential, geographic availability, context/need-state/moment fit). **Delivery fit** (reach potential, frequency controllability, continuity, attention opportunity, viewability/audibility, screen/sound/physical impact, creative format suitability). **Economic fit** (minimum viable spend, cost structure, production/adaptation cost, operational cost, expected marginal value, risk of underfunding, risk of diminishing returns, do not compare channels using CPM alone, different environments deliver different audience/attention/format/context/measurement value). **Quality and governance fit** (inventory transparency, brand safety/suitability, invalid-traffic risk, viewability/completion quality, supply-path clarity, data provenance/privacy, verification availability, frequency governance). **Measurement fit** (metric validity for the assigned role, measurement accessibility, experiment/lift-study feasibility, cross-channel comparability, attribution limitations, reporting latency). **Operational fit** (lead time, market capability, partner availability, creative trafficking complexity, localisation requirements, deal/inventory availability, optimisation control).
   
   ## 14. Scoring rules
   
   Scoring is a decision aid, not the strategy. **Default qualitative rating** (preferred when evidence is incomplete): Strong fit, Conditional fit, Weak fit, Not applicable, Evidence unavailable. **Numeric scoring**, only when the user/organisation approves the criteria and weights, evidence supports meaningful differentiation, the same definitions are applied consistently, and the score is accompanied by underlying evidence and caveats. Never invent a universal weighting system. Never use a total score to override a fatal feasibility failure.
   
   **Mandatory decision record**, for every recommended or rejected serious candidate: Role, Channel, Decision, Evidence, Strengths, Limitations, Conditions, Confidence.
   
   ## 15. Channel-system design
   
   Select the smallest coherent channel system that can perform all primary roles, classified as core, supporting or test channels.
   
   **Avoid duplication without purpose.** Two channels may share a role only when they offer a clear distinction: incremental audience, different consumption context, different creative experience, better continuity, geographic extension, risk diversification, or testable synergy. Do not call reach incremental unless supported by appropriate evidence.
   
   **Walled gardens and open web.** Do not frame the choice as one or the other by default. Platform ecosystems may offer first-party signals, native behaviours and integrated optimisation. Open-web environments may offer publisher context, premium editorial adjacency, broader supply-path choice, cross-format activation and independent verification options. Availability and quality vary by market, supplier, inventory and buying route. Only make a specific advantage claim when supported by supplied evidence.
   
   ## 16. Removal analysis
   
   For every core and supporting channel, analyse what happens if it's removed: primary role affected, audience coverage affected, reach effect (if quantified evidence exists), frequency effect (if quantified evidence exists), message-depth/creative effect, context/physical-presence effect, demand-capture/conversion effect, measurement effect, operational savings, budget that could be reallocated, best substitute and its limitation. Format: "If [channel] is removed: Role lost or weakened / Expected consequence / Quantified evidence / What cannot be quantified / Best substitute / Residual gap / Recommendation." Never fabricate the effect, if reach or outcome impact can't be measured, describe the strategic exposure and label it unquantified.
   
   ## 17. Rejection analysis
   
   For each rejected channel, show: role considered, reason considered, evidence reviewed, reason not selected, conditions that would change the decision, potential test if appropriate. Do not criticise a channel in general, explain why it's not suitable for this specific brief, market, audience, budget, creative or measurement requirement.
   
   ## 18. Budget implications
   
   This skill does not replace detailed budget planning. It should establish which roles must be funded first, minimum viable role coverage, which channels are core, which supporting channels can be added as budget grows, which test cells need protected budgets, which channels should be removed first under pressure, and what strategic capability is lost at each reduction. Where reach-curve data exists, invoke or connect the Cross-Channel Reach Investment Planner skill rather than inventing minimum spends or reach effects.
   
   Use scenarios: Minimum viable system (funds essential roles only), Recommended system (funds primary roles and the most valuable supporting roles), Expanded system (adds evidence-backed incremental roles or test opportunities). Do not use arbitrary percentages, budget splits must be supported by role importance, minimum viable investment, expected delivery, evidence and constraints.
   
   ## 19. Mandatory output structure
   
   In order: (1) Executive recommendation, business outcome, audience change required, primary communications task, recommended channel system (core/supporting/test), main trade-off, confidence level. (2) Inputs and evidence, separating client-provided facts, research/source evidence, historical evidence, calculated values, planner assumptions, and unknowns. (3) Strategic problem, current state, desired state, barrier, required change. (4) Task-to-role architecture table: Priority | Communications task | Required media role | Success signal. (5) Candidate universe, listed by environment (open web/programmatic, walled gardens/platforms, search/intent, offline/traditional, owned/earned/direct, market-specific). (6) Feasibility results table: Candidate | Intended role | Gate result | Issue | Resolution needed. (7) Channel evaluation matrix: Channel | Role | Strategic fit | Audience fit | Delivery fit | Economic fit | Quality | Measurement | Operational fit | Evidence confidence, using qualitative ratings unless an approved scoring method is supplied. (8) Final channel architecture table: Status | Channel | Primary role | Why selected | Conditions | Success signal. (9) Rejected alternatives table: Channel | Role considered | Why not selected now | What would change the decision. (10) Removal analysis table: Remove | Role weakened | Expected consequence | Evidence level | Best substitute | Residual gap. (11) Budget scenarios table: Scenario | Roles funded | Channels | What is protected | What is sacrificed. (12) Client-ready rationale, a concise narrative. (13) Challenge-response section, answering only questions supported by the supplied brief and evidence. (14) Caveats and unknowns. (15) Human verification, ending with the checklist from Section 21.
   
   Client-ready rationale template: "The recommendation is built around [primary tasks], not a predetermined platform list. [Channel] is selected to perform [role] because [brief-specific evidence]. [Channel] supports it by [distinct role]. [Alternative] was considered but not selected because [constraint or weaker fit]. If [channel] is removed, the plan would lose or weaken [role], with [quantified effect if supported]. The recommended system therefore represents the smallest coherent mix that can deliver the required audience and communications outcomes within the stated constraints."
   
   Challenge-response questions to answer, only where the brief and evidence actually support an answer: Why these channels? Why not only the client's familiar channels? Why include open web? Why include or exclude social? Why include or exclude search? Why include or exclude television, CTV, audio or OOH? What happens if budget is reduced? Which channel should be removed first and why?
   
   ## 20. Mandatory caveats
   
   Adapt to the plan: recommendations apply to the stated market, audience, objective, duration, budget, creative and measurement conditions; platform/publisher/vendor evidence may use different populations, methodologies, metrics and time periods; cost metrics aren't directly comparable when channel roles, formats, audience quality, contexts and measurement differ; a low CPM/CPV/CPC doesn't by itself indicate greater strategic value; reach figures from different ecosystems must not be added unless an approved deduplicated source supports it; claimed incremental reach, attention, lift or synergy must be tied to a relevant source; historical results don't guarantee future outcomes; inventory, pricing, audience availability and auction conditions can change; creative suitability can materially affect a channel's ability to perform its assigned role; measurement availability doesn't determine strategic value, but unmeasurable claims must not be presented as proven; the plan must be revisited if budget, audience, objective, timing, creative, market conditions or channel availability changes.
   
   ## 21. Human verification gates
   
   Before client use, a qualified planner must confirm: business outcome and audience barrier are correct; communications tasks are prioritised; the full relevant channel universe was considered; open web, offline and market-specific media weren't excluded by default; candidate channels passed feasibility gates; evidence is current, relevant and correctly represented; vendor evidence is labelled; no audience, reach, cost or outcome data was invented; channels are assigned distinct roles; a buying platform, supply partner or format was never confused with a channel; rejected channels have brief-specific reasons; removal consequences are quantified only where evidence permits; budget scenarios fund roles rather than arbitrary percentages; reach and deduplication claims are valid; brand safety, suitability, privacy, legal and regulatory restrictions are met; local market capability and inventory are verified; creative requirements are realistic; measurement can assess the assigned role; final channel and budget decisions have human approval. If any critical check fails, label the output "Draft for strategic review."
   
   ## 22. Copy-and-paste input template
   
   When critical information is missing, return this template, then leave every field blank for the user to complete:
   
   Channel Strategy Inputs. Client and market: client/brand, product/service, category, market(s), currency, campaign dates/duration. Strategic objective: business outcome, campaign objective, current audience state, desired audience state, main barrier or behaviour to change, priority message/proposition, required action. Audience: primary audience, secondary audience, geography, known media behaviours, need-states/moments, audience research source/date. Investment: available budget/range, production budget, budget flexibility, existing commitments/minimum spends, reach/frequency target if available. Creative: available assets, permitted formats, new production possible (yes/no), localisation requirements, calls to action. Measurement: primary business measure, primary media measure, secondary measures, lift study/experiment available, attribution restrictions. Constraints: client-mandated channels, client-rejected channels, brand safety/suitability, legal/regulatory, data/privacy, operational lead time, market/inventory restrictions. Evidence provided: audience research, historical campaigns, platform planning outputs, publisher/partner research, rates/minimum spends, reach curves, other. Decision required: channels to recommend, channels being challenged, budget reduction requested, client question.
   
   ## 23. Invocation prompts
   
   **Full strategy prompt:** "/channel-strategy-justification. Build a client-defensible channel strategy using the attached brief, research and historical evidence. Do not default to familiar platforms. Consider the full relevant universe, including open-web display, native, online video, premium publishers, CTV, audio, podcasts, DOOH, gaming, retail media, search, social, offline and market-specific channels. Please: define the business outcome, audience barrier and required behaviour change; convert them into prioritised communications tasks and media roles; build the credible candidate-channel universe; apply feasibility gates; evaluate serious candidates consistently using qualitative ratings; recommend the smallest coherent channel system; classify channels as core, supporting, test, reserve or not recommended; explain why each channel was selected and why alternatives were rejected; analyse what happens if each core or supporting channel is removed; create minimum viable, recommended and expanded channel-system scenarios; produce a client-ready rationale and responses to likely challenges; include assumptions, evidence confidence, caveats and human verification gates; do not invent audience, reach, price or performance data."
   
   **Client-familiarity challenge:** "/channel-strategy-justification. The client prefers the following channels because they are familiar: [insert]. Use the attached brief and evidence to test those preferences rather than automatically accepting or rejecting them. Start with the required audience change, communications tasks and media roles. Consider walled gardens, open web, CTV, audio, DOOH, traditional, retail media and market-specific options. Show: which client-preferred channels remain justified; which should change role, scale or buying route; what important roles are currently missing; which alternatives were considered; the consequence of using only the familiar channels; a respectful client-ready rationale; all caveats and human verification requirements."
   
   **Channel-removal prompt:** "/channel-strategy-justification. The client wants to remove [channel] or reduce the media budget from [original] to [revised]. Using the attached strategy and evidence: identify the role performed by the channel; explain the expected consequence of removal; quantify the effect only where valid evidence exists; identify the best substitute and residual gap; recommend which channel or role should be protected first; show a revised minimum viable channel system; produce a concise client response; include caveats and human verification gates."
   
   **Short practitioner prompt:** "/channel-strategy-justification. Build a defensible channel recommendation from the attached brief. Market: / Audience: / Business outcome: / Audience barrier: / Campaign objective: / Duration: / Budget: / Creative available: / Measurement requirement: / Client-preferred channels: / Constraints: Consider the full channel universe, including open web, display, video, CTV, audio, podcasts, DOOH, retail media, gaming, search, social, traditional and local-market inventory. Recommend roles before channels, show rejected alternatives and removal consequences, and do not invent evidence."
   
   ## 24. Expected final response template
   
   Channel Strategy Recommendation. Executive recommendation: business outcome, audience change, primary communications task, recommended channel system (core/supporting/test), main trade-off, confidence. Inputs, evidence and unknowns: client-provided facts, evidence supplied, planner assumptions, unknowns. Strategic problem: current state, desired state, barrier, required change. Task-to-role architecture table. Candidate-channel universe by environment. Feasibility results table. Evaluation matrix table. Recommended architecture table. Rejected alternatives table. Removal analysis table. Budget scenarios table. Client-ready rationale. Responses to likely client challenges. Caveats. Human verification: strategy approved, evidence verified, local availability verified, budget viability verified, creative viability verified, measurement design verified.
   
   ## 25. Installation and invocation
   
   Save this document as: .claude/skills/channel-strategy-justification/SKILL.md. Invoke it in Claude Code with: /channel-strategy-justification
   
   ## 26. Source references
   
   Use current, applicable sources when the environment permits browsing. Useful strategic references: WFA cross-media measurement blueprint (wfanet.org), WFA Halo framework (halo.wfanet.org), IAB Unified Media Planning Playbook (iab.com), IAB Europe advertising format standards (iabeurope.eu). These inform governance and the need to handle fragmented environments carefully, they do not replace market-specific audience, cost, inventory, creative or outcome evidence.

   Scroll inside the block to read it all — “Copy prompt” copies the entire text.

   Do this

   For Claude: save as .claude/skills/channel-strategy-justification/SKILL.md if you're on Claude Code, or add it to a Project's knowledge otherwise, invoke with /channel-strategy-justification or by referencing it directly. For ChatGPT or Gemini: add as a custom instruction, Project/Gem system prompt, or paste at the start of a dedicated thread.

3. 03

   Gather your real inputs before you run it

   Do this

   Critical, the skill will refuse a final recommendation without these: client/brand, market, category, business outcome, campaign objective, audience definition, the audience barrier or behaviour to change, campaign duration, available budget or range, available/feasible creative, measurement requirement, and material brand safety/legal/regulatory restrictions. If these are missing, don't let the skill guess, it will label the work "Draft for strategic review," list what's missing, and provide a copy-paste input template instead of inventing a channel list.

4. 04

   Run it with the prompt that matches your actual situation

   /channel-strategy-justification
   
   Build a defensible channel recommendation from the attached brief.
   
   Market:
   Audience:
   Business outcome:
   Audience barrier:
   Campaign objective:
   Duration:
   Budget:
   Creative available:
   Measurement requirement:
   Client-preferred channels:
   Constraints:
   
   Consider the full channel universe, including open web, display, video, CTV, audio, podcasts, DOOH, retail media, gaming, search, social, traditional and local-market inventory. Recommend roles before channels, show rejected alternatives and removal consequences, and do not invent evidence.

   Do this

   This is the short practitioner version for a normal new recommendation. Two other real variants exist in the full skill file for specific situations: one for when a client insists on familiar channels and you need to test that preference against evidence rather than just accept or reject it, and one for when a client wants to remove a channel or cut budget and you need to show the actual consequence, not just comply.

5. 05

   Verify before anything goes near a client

   Do this

   Confirm: the full relevant channel universe was actually considered, not narrowed to search/social/YouTube by default; every recommended channel has a distinct role, not duplicated without a clear reason; a buying platform, SSP or managed-service partner was never mistaken for a channel; rejected channels have brief-specific reasons, not generic criticism; removal consequences are quantified only where real evidence supports it; budget scenarios fund roles, not arbitrary percentages; and no audience, reach, cost or outcome figure was invented anywhere. If any critical check fails, the output should be labelled "Draft for strategic review," not presented as client-ready.