fsbtactic-code/marketing-brain-skill / skills/marketing-brain
marketing-brain
FlaggedCreate, improve, research, or audit marketing strategy and execution using a bundled marketing library, verified live primary sources for volatile facts, and clearly labeled agent judgment. Use for full marketing systems, audits, landing pages and CRO, competitor intelligence, brand voice, audience and JTBD research, email sequences, SMM and social content systems, paid ads, headlines, copywriting, positioning, offers, launches, GTM, growth, funnels, retention, pricing, SEO, analytics, attribution, experiments, and consumer or buyer psychology. Trigger for English and Russian requests, including requests for ready-to-use assets.
assets/cards.jsonl
View on GitHub{"application": "Segment ad tests by preexisting firm attributes and organic demand, and avoid applying an average result uniformly across accounts.", "card_id": "card_ad_effects_heterogeneous_by_firm", "checked_at": "2026-07-22", "claim": "A randomized field experiment across restaurants found digital advertising effects varied by firm characteristics, with larger gains for some independent, higher-rated, and higher-organic-traffic businesses.", "conditions": ["Restaurant firms on the studied review platform and the ad package used in the experiment."], "confidence": 0.84, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "strategy-research", "measurement-analytics"], "evidence_class": "B", "limitations": ["Results may not generalize to other industries, platforms, or campaign formats.", "Authors disclosed consulting relationships with technology companies."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1287/mksc.2023.1436", "relation": "supports", "source_id": "src_dai_kim_luca_digital_ads_2023"}], "status": "verified"}
{"application": "Plan ad incrementality tests around detectable effect size, variance, sample size, and cost. Treat observational attribution as vulnerable to targeting selection bias.", "card_id": "card_ad_roi_measurement_noise", "checked_at": "2026-07-22", "claim": "Across 25 large field experiments, Lewis and Rao found advertising ROI estimates were often extremely imprecise and argued that randomized trials add unbiased information while observational methods face serious selection bias.", "conditions": ["Large U.S. retailer and brokerage digital advertising experiments studied in the paper."], "confidence": 0.92, "counterevidence": [], "distribution": "public", "domains": ["measurement-analytics", "paid-acquisition"], "evidence_class": "A", "limitations": ["The exact variance and required sample size depend on business economics and outcome noise.", "The study does not imply that every ad experiment is infeasible."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1093/qje/qjv023", "relation": "supports", "source_id": "src_lewis_rao_ad_returns_2015"}], "status": "verified"}
{"application": "Do not reduce product choices by default. Test assortment size together with decision complexity, preference certainty, and customer context.", "card_id": "card_choice_overload_context_dependent", "checked_at": "2026-07-22", "claim": "A meta-analysis of 63 conditions found the mean choice-overload effect was virtually zero with substantial variation, so more options do not reliably reduce choice or satisfaction in every context.", "conditions": ["The 50 published and unpublished experiments included in the meta-analysis."], "confidence": 0.94, "counterevidence": ["Individual studies found strong overload effects under some conditions."], "distribution": "public", "domains": ["consumer-behavior", "cro-funnels", "offer-messaging-copy"], "evidence_class": "A", "limitations": ["The meta-analysis did not identify sufficient universal boundary conditions.", "Specific product and decision environments can still show overload."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1086/651235", "relation": "supports", "source_id": "src_choice_overload_meta_2010"}], "status": "verified"}
{"application": "Test cost transparency when disclosure can credibly signal openness and the cost story is understandable and favorable to trust.", "card_id": "card_cost_transparency_context", "checked_at": "2026-07-22", "claim": "Mohan, Buell, and John found that disclosing production costs increased trust and purchase interest in their field and online experiments, including a preregistered cafeteria study.", "conditions": ["The product, message, and experimental contexts reported in the paper."], "confidence": 0.83, "counterevidence": [], "distribution": "public", "domains": ["pricing-promotions", "consumer-behavior", "offer-messaging-copy"], "evidence_class": "B", "limitations": ["The effect should not be assumed for all categories, cost structures, or customer segments.", "Disclosed numbers must be truthful and contextually meaningful."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1287/mksc.2019.1200", "relation": "supports", "source_id": "src_cost_transparency_2020"}], "status": "verified"}
{"application": "When publishing sponsored creator content for U.S. consumers, disclose money, gifts, discounts, employment, personal, or family relationships directly with the endorsement in a form viewers can notice and understand.", "card_id": "card_ftc_material_connection_disclosure", "checked_at": "2026-07-22", "claim": "FTC guidance says influencers should disclose material brand relationships and place the disclosure with the endorsement where it is hard to miss, not only in a profile or behind a More action.", "conditions": ["U.S. advertising context or content reasonably expected to affect U.S. consumers."], "confidence": 0.99, "counterevidence": [], "distribution": "public", "domains": ["creator-community", "law-policy-ethics"], "evidence_class": "A", "limitations": ["Other jurisdictions may impose additional or different requirements.", "This card is operational guidance, not legal advice."], "review_due": "2027-07-22", "source_refs": [{"locator": "Sections: When to Disclose; How to Disclose", "relation": "supports", "source_id": "src_ftc_influencer_disclosures"}], "status": "verified"}
{"application": "Before launching an ad experiment, state one business-linked hypothesis, isolate the changed variable, choose one or two decision metrics, avoid contaminating the base, and log the result.", "card_id": "card_google_ads_experiment_hygiene", "checked_at": "2026-07-22", "claim": "Google Ads recommends a clear hypothesis, testing one variable at a time, selecting success metrics before the test, limiting base-campaign changes, and keeping experiment records.", "conditions": ["Google Ads campaign experiments."], "confidence": 0.97, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "cro-funnels", "measurement-analytics"], "evidence_class": "A", "limitations": ["Platform guidance does not replace power analysis or statistical review.", "Some experiment types can involve multiple coordinated settings."], "review_due": "2026-08-21", "source_refs": [{"locator": "Sections 1 through 3: Set a clear hypothesis; Create your experiment; Analyze results", "relation": "supports", "source_id": "src_google_ads_experiment_practice_2026"}], "status": "verified"}
{"application": "Do not pad articles to a rumored target length or change dates without substantive updates. Optimize structure only after usefulness and originality are established.", "card_id": "card_google_no_word_count_freshness_trick", "checked_at": "2026-07-22", "claim": "Google says it has no preferred word count and warns against changing dates or adding and removing content mainly to make a site appear fresh for rankings.", "conditions": ["Google Search guidance, not a general editorial rule for every product."], "confidence": 0.98, "counterevidence": [], "distribution": "public", "domains": ["seo-geo-ai-search", "content-editorial"], "evidence_class": "A", "limitations": ["Some topics still require more detail than others.", "Substantive updates remain useful."], "review_due": "2026-08-21", "source_refs": [{"locator": "Section: Avoid creating search engine-first content", "relation": "supports", "source_id": "src_google_people_first_content_2025"}], "status": "verified"}
{"application": "Build content around a real audience goal, first-hand expertise, original analysis, and a satisfying answer instead of scaling thin summaries.", "card_id": "card_google_people_first", "checked_at": "2026-07-22", "claim": "Google Search recommends people-first content with original value, clear expertise, a defined audience, and enough substance to help the reader achieve a goal.", "conditions": ["Google Search content guidance, page updated 2025-12-10."], "confidence": 0.98, "counterevidence": [], "distribution": "public", "domains": ["seo-geo-ai-search", "content-editorial"], "evidence_class": "A", "limitations": ["This is qualitative guidance, not a guaranteed ranking formula.", "Ranking uses many systems and factors."], "review_due": "2026-08-21", "source_refs": [{"locator": "Sections: Self-assess your content; Focus on people-first content", "relation": "supports", "source_id": "src_google_people_first_content_2025"}], "status": "verified"}
{"application": "Add accurate bylines, author background, methodology, test details, and disclosure of meaningful automation where readers would benefit from them.", "card_id": "card_google_who_how_why", "checked_at": "2026-07-22", "claim": "Google recommends evaluating content by who created it, how it was produced, and why it exists, with accurate authorship and useful process transparency.", "conditions": ["Content where authorship or production method affects reader trust."], "confidence": 0.98, "counterevidence": [], "distribution": "public", "domains": ["seo-geo-ai-search", "content-editorial", "law-policy-ethics"], "evidence_class": "A", "limitations": ["These are trust and quality principles, not a standalone ranking factor checklist."], "review_due": "2026-08-21", "source_refs": [{"locator": "Section: Ask Who, How, and Why about your content", "relation": "supports", "source_id": "src_google_people_first_content_2025"}], "status": "verified"}
{"application": "Use a treatment and unexposed control design when the decision requires causal incremental conversions rather than attributed conversions alone.", "card_id": "card_incrementality_control_group", "checked_at": "2026-07-22", "claim": "Google defines Conversion Lift as a controlled experiment that separates audiences into treatment and control groups to estimate conversions caused by ad exposure.", "conditions": ["Eligible Google Ads accounts and supported Conversion Lift designs."], "confidence": 0.98, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "measurement-analytics"], "evidence_class": "A", "limitations": ["Availability and implementation requirements vary.", "A noisy or underpowered experiment can still be inconclusive."], "review_due": "2026-08-21", "source_refs": [{"locator": "Sections: About Conversion Lift; How Conversion Lift works", "relation": "supports", "source_id": "src_google_ads_conversion_lift_2026"}], "status": "verified"}
{"application": "For noisy lift tests, consider preplanned heterogeneity models only when the strata model is identifiable, documented, and validated against a standard estimator.", "card_id": "card_incrementality_precision_stratification", "checked_at": "2026-07-22", "claim": "Berman and Feit report that latent stratification reduced average-treatment-effect variance by 30% to 60% in five catalog experiments under favorable identification conditions.", "conditions": ["The method, simulations, and catalog experiments reported in the paper."], "confidence": 0.83, "counterevidence": [], "distribution": "public", "domains": ["measurement-analytics", "paid-acquisition"], "evidence_class": "B", "limitations": ["Benefits depend on treatment-effect differences and model identification.", "This is a specialized estimator, not a replacement for randomization or power planning."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1287/mksc.2022.0297", "relation": "supports", "source_id": "src_berman_feit_incrementality_2024"}], "status": "verified"}
{"application": "When evaluating short-form video ads, test off-platform outcomes where possible and do not assume platform-reported conversions capture the full effect.", "card_id": "card_short_video_ad_spillover", "checked_at": "2026-07-22", "claim": "A 2025 randomized ByteDance field experiment found substantial off-platform conversion spillover, which materially changed campaign cost-per-conversion estimates in the studied automobile campaign.", "conditions": ["The ByteDance randomized experiment and advertiser context reported in the paper."], "confidence": 0.82, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "organic-smm", "measurement-analytics"], "evidence_class": "B", "limitations": ["One platform and campaign context cannot establish a universal spillover multiplier.", "Cross-platform measurement requires lawful data access and careful identity handling."], "review_due": "2027-07-22", "source_refs": [{"locator": "Abstract, DOI 10.1287/mksc.2023.0575", "relation": "supports", "source_id": "src_bytedance_ad_spillover_2025"}], "status": "verified"}
{"application": "Draft Telegram Ads copy within the 160-character limit as a single clear text block with one destination.", "card_id": "card_telegram_ad_format", "checked_at": "2026-07-22", "claim": "Telegram Ads currently limits ad text to 160 characters including spaces and prohibits line breaks, bullet points, and numbered lists in the ad text.", "conditions": ["Official Telegram Ad Platform."], "confidence": 0.99, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "offer-messaging-copy"], "evidence_class": "A", "limitations": ["Third-party Telegram ad products may have different formats.", "The official policy can change without notice."], "review_due": "2026-07-29", "source_refs": [{"locator": "Sections 1 and 2: Ad format; Editorial requirements", "relation": "supports", "source_id": "src_telegram_ads_guidelines_2026"}], "status": "verified"}
{"application": "Before moderation, verify that the landing page, bot, or channel is active, complete, directly relevant, and matches the ad language.", "card_id": "card_telegram_destination_quality", "checked_at": "2026-07-22", "claim": "Telegram requires ad destinations to be functional, active, relevant to the ad, and language-aligned; inaccessible, redirect-only, or mismatched destinations can be rejected.", "conditions": ["Official Telegram Ad Platform destination review."], "confidence": 0.99, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "cro-funnels", "law-policy-ethics"], "evidence_class": "A", "limitations": ["Moderation decisions can include additional contextual checks."], "review_due": "2026-07-29", "source_refs": [{"locator": "Sections 4.1 through 4.4: Destination requirements", "relation": "supports", "source_id": "src_telegram_ads_guidelines_2026"}], "status": "verified"}
{"application": "Replace absolute superiority and guaranteed-outcome wording with specific, supportable claims and clear product context.", "card_id": "card_telegram_no_unsupported_absolutes", "checked_at": "2026-07-22", "claim": "Telegram prohibits deceptive claims, clickbait, exaggerated comparisons, and unsupported absolute or superlative claims in ads.", "conditions": ["Official Telegram Ad Platform content review."], "confidence": 0.99, "counterevidence": [], "distribution": "public", "domains": ["paid-acquisition", "offer-messaging-copy", "law-policy-ethics"], "evidence_class": "A", "limitations": ["A supported comparison may still require additional substantiation and rights checks."], "review_due": "2026-07-29", "source_refs": [{"locator": "Section 5.4: Deceptive, misleading, or predatory advertising", "relation": "supports", "source_id": "src_telegram_ads_guidelines_2026"}], "status": "verified"}
{"application": "Test emotionally activating content against usefulness and audience fit instead of applying a simplistic positive-versus-negative rule.", "card_id": "card_viral_content_arousal", "checked_at": "2026-07-22", "claim": "Berger and Milkman found that social transmission was associated with physiological arousal: high-arousal positive and negative emotions were more viral, while low-arousal sadness was less viral in their studied contexts.", "conditions": ["New York Times article data and the experiments reported in the paper."], "confidence": 0.82, "counterevidence": [], "distribution": "public", "domains": ["consumer-behavior", "content-editorial", "organic-smm"], "evidence_class": "B", "limitations": ["The observational dataset covered one publisher and period.", "Platform algorithms and audiences differ.", "The result does not justify manipulative emotional content."], "review_due": "2028-07-22", "source_refs": [{"locator": "Abstract and experiments, DOI 10.1509/jmr.10.0353", "relation": "supports", "source_id": "src_berger_milkman_viral_2012"}], "status": "verified"}
{"application": "Design YouTube topics, packaging, and series around a defined audience and evaluate click, watch, and satisfaction signals instead of chasing a single algorithm trick.", "card_id": "card_youtube_audience_signals", "checked_at": "2026-07-22", "claim": "YouTube says recommendations are personalized using viewer preferences and content performance, including watch behavior, search, engagement, dismissals, and satisfaction signals.", "conditions": ["Applies to YouTube recommendations as described in the official 2026 help page."], "confidence": 0.97, "counterevidence": [], "distribution": "public", "domains": ["organic-smm", "content-editorial"], "evidence_class": "A", "limitations": ["YouTube does not disclose full ranking weights.", "Signals and product surfaces can change."], "review_due": "2026-08-21", "source_refs": [{"locator": "Sections: The algorithm follows the audience; Recommendations are personalized", "relation": "supports", "source_id": "src_youtube_recommendations_2026"}], "status": "verified"}
{"application": "Choose publish time for early audience availability, Live, and Premiere needs, but do not treat it as a long-term ranking lever for regular uploads.", "card_id": "card_youtube_publish_time", "checked_at": "2026-07-22", "claim": "YouTube states that publish time is not known to affect a video's long-term performance, although timing can matter for early viewership, Live, and Premiere formats.", "conditions": ["Regular YouTube uploads under the current recommendation system."], "confidence": 0.97, "counterevidence": [], "distribution": "public", "domains": ["organic-smm", "content-editorial"], "evidence_class": "A", "limitations": ["Audience timing can still affect initial response.", "The statement does not apply identically to live formats."], "review_due": "2026-08-21", "source_refs": [{"locator": "Section: What does this mean for your channel growth strategy?", "relation": "supports", "source_id": "src_youtube_recommendations_2026"}, {"locator": "FAQ: When is the best time to publish videos?", "relation": "supports", "source_id": "src_youtube_performance_faq_2026"}], "status": "verified"}
{"application": "Select a sustainable publishing cadence and judge each upload by audience response rather than using daily volume as a default rule.", "card_id": "card_youtube_quality_over_frequency", "checked_at": "2026-07-22", "claim": "YouTube recommends consistent quality over high upload frequency and reports no observed correlation between growth in views and the time between uploads.", "conditions": ["Channel growth under the current official guidance."], "confidence": 0.95, "counterevidence": [], "distribution": "public", "domains": ["organic-smm", "content-editorial"], "evidence_class": "A", "limitations": ["A lower cadence can still reduce the number of opportunities to reach viewers.", "Audience expectations vary by format and niche."], "review_due": "2026-08-21", "source_refs": [{"locator": "Section: Quality over quantity", "relation": "supports", "source_id": "src_youtube_recommendations_2026"}, {"locator": "FAQ: Do I need to upload daily or at least once a week?", "relation": "supports", "source_id": "src_youtube_performance_faq_2026"}], "status": "verified"}
{"application": "Write accurate titles and descriptions that match the search intent and video content, then improve actual engagement for that query.", "card_id": "card_youtube_search_ranking", "checked_at": "2026-07-22", "claim": "YouTube Search ranks videos using how well the title, description, and video content match a query, together with engagement for that search. Results are not simply ordered by total views.", "conditions": ["YouTube Search, not Home or Suggested."], "confidence": 0.97, "counterevidence": [], "distribution": "public", "domains": ["organic-smm", "seo-geo-ai-search"], "evidence_class": "A", "limitations": ["YouTube does not publish exact weights.", "Eligibility and policy constraints also apply."], "review_due": "2026-08-21", "source_refs": [{"locator": "FAQ: How are videos ranked in Search?", "relation": "supports", "source_id": "src_youtube_performance_faq_2026"}], "status": "verified"}
{"application": "Spend optimization effort on audience fit, content, title, and thumbnail before adding extensive tag lists.", "card_id": "card_youtube_tags_low_priority", "checked_at": "2026-07-22", "claim": "YouTube describes tags as not important for discovery and says they are primarily used to correct common spelling mistakes.", "conditions": ["Current YouTube discovery guidance."], "confidence": 0.97, "counterevidence": [], "distribution": "public", "domains": ["organic-smm"], "evidence_class": "A", "limitations": ["Tags can still help with spelling variants.", "This does not remove the need for accurate metadata."], "review_due": "2026-08-21", "source_refs": [{"locator": "FAQ: How important are Tags?", "relation": "supports", "source_id": "src_youtube_performance_faq_2026"}], "status": "verified"}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |