{
  "agent_role": "Senior Claude B · Skeptic / Contradiction / Outlier Auditor",
  "topic_context_received": "WR Placement-Automation Opportunity 2026 · 146 YouTube transcripts spanning local-SEO tools, GBP optimization, AEO/GEO, manufacturer cert programs, BBB, Yelp Fusion, Apple Maps, Wikidata, Reddit, citation-builder vendors, industry-specific marketing.",
  "source_bundle_ids": ["bundle_00", "bundle_01", "bundle_02", "bundle_03", "bundle_04", "bundle_05", "bundle_06", "bundle_07"],
  "source_video_count_total": 146,
  "challenged_consensus": [
    {
      "consensus_claim_being_challenged": "AI agents reconstruct local-business answers from a small graph of 8-15 external surfaces (the Texas forensic finding)",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_01","bundle_02","bundle_03","bundle_04","bundle_05","bundle_06","bundle_07"],
      "challenge": "The Texas forensic finding is WR's OWN data. The YouTube corpus AGREES with it because AEO/GEO creators are downstream of the same narrative (Reddit/Wikipedia/BBB cited by LLMs). This is parallel claims, not independent corroboration. None of the 146 videos ran a controlled forensic test like WR did. Risk: confirmation bias loop where WR cites creators who cite WR-style reasoning.",
      "failure_mode": "circular_citation",
      "what_would_make_it_solid": "Re-run the Texas forensic in 2 non-Texas verticals (HVAC-Florida and Dental-California) using the SAME methodology before scaling the Pack to 100 customers. If the surface graph changes meaningfully by vertical/region, the 7-platform MVP needs vertical-specific reweighting."
    },
    {
      "consensus_claim_being_challenged": "Placement Pack at $1,497 one-time with 76% margin is defensible",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_01","bundle_02","bundle_03","bundle_04","bundle_05","bundle_06","bundle_07"],
      "challenge": "The $150 labor estimate is wishful. Realistic per-customer Pack delivery includes: GBP claim assistance (often requires video verification call · 30-60 min), Apple BCN walkthrough (15-30 min including D-U-N-S helper), BBB application kit prep (45 min), Yelp claim walkthrough (15-30 min), Wikidata Q-item construction (30 min), Facebook Page audit (15 min), vertical add-on (30-60 min) → 3-5 hours of Wave/Ricky time per Pack. At a real $200/hr labor opportunity cost, COGS is closer to $800-1200, margin closer to 20-50%, not 76%.",
      "failure_mode": "no_underlying_data",
      "what_would_make_it_solid": "Time-track the first 3 pilot Pack deliveries end-to-end · use ACTUAL hours not the optimistic Wave-orchestrated number · re-price if margin <60%."
    },
    {
      "consensus_claim_being_challenged": "Reviews velocity at 40+/month is a consensus target",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_01","bundle_02","bundle_05"],
      "challenge": "40+ reviews/month is achievable for high-volume service businesses (chain pizza, mid-size dental practice) but unrealistic for a small towing operator or solo law practice. The 40/mo number is a creator-doctrine echo chamber · multiple videos use the SAME phrase (Brock Misner's GBP optimization series in particular). Risk: WR promises 40/mo to a customer whose total monthly customer count is 30, and the promise breaks.",
      "failure_mode": "groupthink_among_creators",
      "what_would_make_it_solid": "Set per-vertical realistic review-velocity targets · roofing 10-15/mo · dental 20-30/mo · towing 5-10/mo · scale promise to customer's actual volume."
    },
    {
      "consensus_claim_being_challenged": "BrightLocal/Whitespark/Synup/Yext/Moz Local 'don't ship done-for-you' so WR has clean wedge",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_01","bundle_02","bundle_03","bundle_05"],
      "challenge": "All five vendors have white-label agency tiers where their own partners DO sell done-for-you placement at $497-$2,000 one-time. The 'gap' is not pricing-gap, it's INTEGRATION-gap (none of them combine pixel-faithful site + AI-citation score + placement pack). WR's wedge is the COMBINED stack, not the placement pack alone. If WR sells the Pack standalone (without the Rebuild attach), they compete head-on with white-label citation agencies at saturated $500-1500 prices.",
      "failure_mode": "vendor_marketing",
      "what_would_make_it_solid": "Position Pack as attach-only (Rebuild required) for first 6 months · only break out as standalone after the combined-stack story has 20+ case studies."
    },
    {
      "consensus_claim_being_challenged": "Yelp Fusion API + Yelp for Business flow lets WR include Yelp in the Pack",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_05","bundle_06"],
      "challenge": "Yelp Fusion API is READ-ONLY. There is NO Yelp listing-write API for third parties. Yelp's terms of service explicitly prohibit automated posting on behalf of clients AND Yelp aggressively removes review-requesting behaviors. Including Yelp in the Pack risks (a) Yelp suppressing the customer's reviews if WR review-request campaign triggers Yelp's filter, (b) ToS violation if WR posts as the customer. Yelp traffic for service businesses has also been collapsing since 2020 — adding Yelp may be performative not impactful.",
      "failure_mode": "outdated_tactic",
      "what_would_make_it_solid": "Verify Yelp's 2026 ToS directly · confirm policy on agency claim assistance vs posting · consider DROPPING Yelp from the MVP-7 in favor of Foursquare + Bing Places."
    },
    {
      "consensus_claim_being_challenged": "Apple Business Connect verification is low-friction now",
      "supporting_bundle_ids_in_corpus": ["bundle_03","bundle_05"],
      "challenge": "The 'phone-bill verification' fallback is mentioned in videos from 2023-2024. As of 2026, Apple has tightened verification for non-US businesses and added a behind-the-scenes manual review queue for high-volume claims. The friction-reduction claim is based on consumer-anecdote videos, not Apple platform documentation. A WR customer using a virtual mailbox or shared address may STILL be stuck.",
      "failure_mode": "outdated_tactic",
      "what_would_make_it_solid": "Run live Apple BCN setup on a fresh WR-side test business · document actual 2026-05 friction · update Pack runbook."
    },
    {
      "consensus_claim_being_challenged": "Wikidata is fully automatable at near-zero marginal cost via Quickstatements",
      "supporting_bundle_ids_in_corpus": ["bundle_01","bundle_02"],
      "challenge": "Wikidata's community has been pushing back on commercial paid-editor activity since 2024. Q-items for local service businesses (notability-low entities) are being challenged for deletion at increasing rates. A WR-created Wikidata Q-item for 'Bob's Roofing in Salisbury MD' has a non-trivial chance (rough estimate 15-30%) of being nominated for deletion within 90 days unless the entity has external citations supporting notability. The 'near-zero marginal cost' ignores the rework cost when deletion is fought.",
      "failure_mode": "coach_speculation",
      "what_would_make_it_solid": "Test Quickstatements creation for 5 WR-pilot customers · monitor for deletion-nomination over 90 days · only productize if survival rate >75%."
    },
    {
      "consensus_claim_being_challenged": "Reddit brand-helper-account model is productizable",
      "supporting_bundle_ids_in_corpus": ["bundle_01","bundle_05","bundle_06","bundle_07"],
      "challenge": "Reddit's site-wide promo policy enforces a 1:9 self-promo ratio AND many service-vertical subreddits ban 'who's the best [vertical] in [town]' threads outright. A brand-helper account that comments on local service threads (which is the productizable pattern) is exactly the kind of behavior that gets shadowbanned. WR Reddit doctrine memory already flagged this. The 19 corpus videos cite the tactic but only 2-3 show actual conversion data. The remaining 16 are 'do this'-style coaching without proof.",
      "failure_mode": "coach_speculation",
      "what_would_make_it_solid": "Run Reddit helper account in 3 verticals for 90 days from one Wave-managed identity · track actual reply-to-customer conversion · only include in Pack if reply rate >2% AND no account suspension."
    },
    {
      "consensus_claim_being_challenged": "AI-Citation Monitor at $79/mo recurring is a high-margin add-on",
      "supporting_bundle_ids_in_corpus": ["bundle_03","bundle_05","bundle_07"],
      "challenge": "Pricing competition is the actual risk, not API cost. Profound is enterprise but Peec and Bluefish are racing to $99/mo small-business tiers. Search Atlas already offers AI-visibility tracking bundled in $149/mo. The $79/mo monitor competes head-on with already-shipped products. WR's differentiation must be 'tied to placement remediation' not just 'we score you' · otherwise it's just another vendor in a crowded category.",
      "failure_mode": "saturated_market",
      "what_would_make_it_solid": "Price $79/mo only if it includes auto-remediation (when score drops, WR fires the re-placement playbook free for that customer). Otherwise raise to $129/mo with quarterly placement-refresh service · or KILL the monitor and roll it into Marina recurring."
    },
    {
      "consensus_claim_being_challenged": "Manufacturer-cert programs (GAF Master Elite, OC Platinum) are NOT productizable",
      "supporting_bundle_ids_in_corpus": ["bundle_00","bundle_03","bundle_06"],
      "challenge": "The corpus says they're invitation-only and multi-year, but they ALSO show that social-media-attention accelerates invitation. There IS a productizable angle WR is missing: 'manufacturer-attention campaign' as a $297-997 add-on that creates the social-media trail GAF/OC reps watch. The senior consensus mentions this in passing as a singleton, but it deserves elevation as a real product line, not just a side-note.",
      "failure_mode": "cherry_picked_anecdote",
      "what_would_make_it_solid": "Spec the 'Manufacturer Attention Campaign' as a separate product line · 90-day social-media program targeting GAF/OC reps in customer's state · price $497-997 · separate from base Pack."
    }
  ],
  "weak_or_overstated_findings": [
    {
      "finding": "'47.9% of ChatGPT local citations come from Wikipedia' style stats",
      "in_corpus_strength": "appears_strong",
      "actual_evidence_strength": "weak",
      "why_weak": "Multiple videos cite this kind of stat but none link to the underlying research. The actual study (Profound's 2024 white paper or similar) had a narrow query set that may not generalize to service-vertical queries.",
      "source_video_ids": ["xYhsEA_RfZk", "qDniSowMxh0", "OCfS0XCcct4"]
    },
    {
      "finding": "'Becoming GAF Master Elite takes 5+ years'",
      "in_corpus_strength": "appears_strong",
      "actual_evidence_strength": "weak",
      "why_weak": "One first-person operator narrative · creators echo it · but the actual GAF program documentation may have changed since the source video was recorded (2021). 2026 requirements need direct verification.",
      "source_video_ids": ["-b6aaqNJchY"]
    },
    {
      "finding": "'Reddit pages often outrank real business websites'",
      "in_corpus_strength": "appears_strong",
      "actual_evidence_strength": "weak",
      "why_weak": "Anecdotal. True for some queries (especially commercial intent + brand mentions), false for many. Generalizing to 'always rank for local service queries' overstates the pattern.",
      "source_video_ids": ["qV8E793Y_xk"]
    },
    {
      "finding": "Yext at $4-10K/yr enterprise price",
      "in_corpus_strength": "appears_strong",
      "actual_evidence_strength": "weak",
      "why_weak": "Yext SMB tier exists at $199-499/mo · enterprise figures are top-end · using top-end to anchor 'we are cheaper' is misleading marketing not strategy.",
      "source_video_ids": ["u8j7vIJRMUU", "COTaR1IZeSM"]
    },
    {
      "finding": "BBB Accreditation generates leads / increases revenue",
      "in_corpus_strength": "appears_medium",
      "actual_evidence_strength": "weak",
      "why_weak": "Only one BBB-affiliate testimonial in the corpus · plus BBB's own promo videos · zero independent measurement of accreditation lift vs cost.",
      "source_video_ids": ["MxY1Ntszdcs", "bllUat2exos"]
    }
  ],
  "contradictions": [
    {
      "topic": "Yelp's value for service businesses in 2026",
      "claim_A": {"text": "Yelp Fusion API is alive, claim flow works, Yelp is in the AI-citation graph", "video_ids": ["mn6aj3JitVo", "OuYU7n4wp1A"], "credibility": "coach"},
      "claim_B": {"text": "Yelp traffic for service businesses has collapsed, ToS restricts agency posting, may not be worth the include", "video_ids": ["p81So0yofSY"], "credibility": "operator"},
      "skeptic_take": "Operator claim B is more credible. Yelp's relative weight in 2026 LLM citations is shrinking · keep in Pack but flag low effort, not high impact.",
      "wave_should_investigate": "Pull Yelp citation frequency from WR's own forensic data · if <10% in service-vertical queries, demote to optional add-on."
    },
    {
      "topic": "AI-search tool category maturity",
      "claim_A": {"text": "Profound + Peec + Bluefish are early-stage and beatable", "video_ids": ["FTQsWzpeSSc", "8qvir5dxSyo", "a0Kpni9XEdc"], "credibility": "vendor"},
      "claim_B": {"text": "Semrush + Search Atlas already shipped AI-visibility bundled into existing $149+/mo SEO subscriptions · SMB is already served", "video_ids": ["XuK_duOuoEk", "2iUY6d5W-9o"], "credibility": "vendor"},
      "skeptic_take": "Both true · the AI-monitor SMB market segment is already crowded · WR's $79/mo plus auto-remediation is the only differentiation",
      "wave_should_investigate": "Manually pull pricing pages for Semrush AI Visibility Toolkit + Search Atlas + Peec + Profound · confirm Q3 2026 pricing before launch."
    },
    {
      "topic": "Whether NAP consistency matters",
      "claim_A": {"text": "NAP consistency across 300+ directories is foundational, must-have", "video_ids": ["HcXkQipbsDM", "J5rYupg2l0c", "BS1J3_4XCMo"], "credibility": "coach"},
      "claim_B": {"text": "Whitespark's own studies show NAP consistency is a debated ranking factor, NOT a top-5 driver in 2024-2025", "video_ids": ["J5rYupg2l0c"], "credibility": "vendor"},
      "skeptic_take": "Consensus overstates NAP-consistency. Modern LLMs and Google use entity resolution that tolerates moderate inconsistency. NAP perfection is hygiene, not the ranking lever.",
      "wave_should_investigate": "Position NAP consistency as table-stakes not a marketed Pack feature · don't oversell."
    },
    {
      "topic": "Whether AI Overview / SGE actually drives revenue",
      "claim_A": {"text": "Google AI Overviews on 50%+ of local queries, must optimize for", "video_ids": ["sAMpdORISW0", "Qg8ZaaHo72M", "W014GTY5sfE"], "credibility": "coach"},
      "claim_B": {"text": "AI Overview click-through rates are 30-50% LOWER than blue-link clicks · being CITED doesn't equal getting customers", "video_ids": ["NNn4jdlzTog"], "credibility": "operator"},
      "skeptic_take": "Citation is necessary but insufficient. Track downstream conversion · WR Pack ROI must be measured in customer-acquisition, not just citation-rate.",
      "wave_should_investigate": "Add customer-acquisition KPI to monthly Pack reporting · not just citation score."
    }
  ],
  "ignored_outliers": [
    {
      "outlier_claim": "Apple Business Connect now lets you publish 'Showcases' (mini-promos) that surface in Maps · richer than just NAP",
      "source_video_id": "V0Jj2xUNA2s",
      "why_ignored_by_consensus": "Buried in walkthrough not pitched as differentiated feature",
      "why_worth_re_elevating": "Showcases are a free, under-used Maps-citation surface · WR can include 1-2 Showcase setups per customer in Pack at marginal cost",
      "operator_test_hypothesis": "Customers with 3+ Showcases get cited by Siri 2x more often than customers with bare ABC profile"
    },
    {
      "outlier_claim": "GBP suspension is reaching record highs in 2025-2026 due to AI-detection of fake reviews · adding 40 reviews/month MAY TRIGGER suspension",
      "source_video_id": "t_Radsgg-jU",
      "why_ignored_by_consensus": "Consensus emphasizes 40/mo target without flagging suspension risk",
      "why_worth_re_elevating": "Real risk to the Pack's headline value-add · review-velocity campaign needs throttling + pacing",
      "operator_test_hypothesis": "Pacing 8-12 reviews/wk consistently beats 40-in-month-1-then-zero · less suspension risk · same total"
    },
    {
      "outlier_claim": "Lighthouse 100/100/100/100 scores DROP without a real-world performance win because LLMs don't read Lighthouse scores",
      "source_video_id": "n-_lURRus4w",
      "why_ignored_by_consensus": "WR culture has been Lighthouse-perfectionist · this view challenges it",
      "why_worth_re_elevating": "If true, WR's 100-score Pillar 11 doctrine should be reframed: perfect scores are operator hygiene, NOT a citation lever",
      "operator_test_hypothesis": "Measure LLM citation rate of WR customers with 100/100/100/100 vs 85+ scores · if equal, deprioritize the perfectionism · reallocate hours to placement"
    },
    {
      "outlier_claim": "BBB rating may not actually flow into LLM citations · LLMs cite BBB URLs but use the URL's CONTEXT, not the rating score",
      "source_video_id": "MxY1Ntszdcs",
      "why_ignored_by_consensus": "BBB-pitch videos all conflate citation with rating-as-trust-signal",
      "why_worth_re_elevating": "If LLMs cite BBB regardless of rating, then a customer with B-rating gets same citation as A+ · BBB application support is the product, not 'getting an A+'",
      "operator_test_hypothesis": "Cross-query LLM citations of 5 BBB-listed customers across rating ranges · if citation rate equal, BBB application support is the entire product"
    },
    {
      "outlier_claim": "ChatGPT/Perplexity citations on 'X near me' queries are sourced via Browse tools, not the model's parametric memory · the citation surface is dynamic",
      "source_video_id": "NNn4jdlzTog",
      "why_ignored_by_consensus": "Consensus treats LLM citation graph as static",
      "why_worth_re_elevating": "If browse-dynamic, the surfaces that matter are whichever Bing+Google rank highly TODAY for the customer's town/vertical · re-rerunning the forensic monthly catches drift",
      "operator_test_hypothesis": "Monitor LLM browse-fetched URLs for 5 service queries over 60 days · log the URLs · see how the surface set rotates"
    },
    {
      "outlier_claim": "Local Service Ads (LSA / Google Guaranteed) + Google Business Profile cross-feed each other · LSA-verified businesses get GBP priority",
      "source_video_id": "VoumtFXyy3k",
      "why_ignored_by_consensus": "LSA was deprioritized as 'separate product' but it directly improves GBP standing",
      "why_worth_re_elevating": "LSA verification could be a Pack add-on for verticals where Google Guaranteed exists (HVAC, plumbing, locksmith, lawyers in some states)",
      "operator_test_hypothesis": "Include LSA-verification-support in HVAC vertical add-on · measure 30-day GBP rank lift"
    }
  ],
  "singletons_that_may_matter": [
    {
      "insight": "Bypass video verification via aged-account + citation-consistency · workaround for the 2024+ GBP video-call gate",
      "source_video_id": "t_Radsgg-jU",
      "source_bundle_id": "bundle_05",
      "specificity": "high",
      "evidence_strength_in_source": "medium",
      "why_might_be_gold": "The video-verification gate is the #1 onboarding blocker · workaround is operator-grade specific",
      "smell_test_pass_or_fail": "pass",
      "smell_test_reasoning": "Specific enough to test · also matches WR memory pattern about GBP verification stalls"
    },
    {
      "insight": "Cartesia voice + Marina answer-agent reused as inbound-handler during 60-day review-velocity push",
      "source_video_id": "vbofkN89sCg",
      "source_bundle_id": "bundle_06",
      "specificity": "high",
      "evidence_strength_in_source": "weak",
      "why_might_be_gold": "Cross-product synergy WR already paid for · raises Pack attach rate without raising COGS",
      "smell_test_pass_or_fail": "pass",
      "smell_test_reasoning": "Marina is shipped · Cartesia voice locked · only integration work"
    },
    {
      "insight": "MyGBPro $40-99/mo bundles GBP audit + NAP consistency + reviews automation · is the closest competitor to the Pack delivery vehicle",
      "source_video_id": "4XftBZJSK78",
      "source_bundle_id": "bundle_00",
      "specificity": "high",
      "evidence_strength_in_source": "medium",
      "why_might_be_gold": "If MyGBPro can be the white-label execution layer for WR's $1,497 Pack, it eliminates 60-80% of the delivery labor",
      "smell_test_pass_or_fail": "inconclusive",
      "smell_test_reasoning": "Pricing/ToS depend on validating MyGBPro's white-label/agency terms · worth a 30-min vendor call before MVP"
    },
    {
      "insight": "Salient Process saves BBB $1.5M/yr by automating accreditation licensing compliance · BBB has internal API surface but not public",
      "source_video_id": "gpv4esOq67A",
      "source_bundle_id": "bundle_04",
      "specificity": "medium",
      "evidence_strength_in_source": "weak",
      "why_might_be_gold": "Suggests BBB has private API surface · agency-partner program may exist · worth direct outreach to BBB partner team",
      "smell_test_pass_or_fail": "inconclusive",
      "smell_test_reasoning": "Promising lead but unverified · low-cost lookup before committing in MVP"
    },
    {
      "insight": "Foursquare data feeds Apple Maps, Bing Places, Snap Maps, Uber, several other downstream surfaces · 1 claim = 5+ derivative citations",
      "source_video_id": "cib1Mnl2Kfc",
      "source_bundle_id": "bundle_07",
      "specificity": "medium",
      "evidence_strength_in_source": "medium",
      "why_might_be_gold": "Foursquare punches above weight as upstream-of-many-listings · should be in MVP-7 instead of (or alongside) Yelp",
      "smell_test_pass_or_fail": "pass",
      "smell_test_reasoning": "Free, fast claim, broad downstream amplification · low-risk high-leverage inclusion"
    },
    {
      "insight": "Nextdoor lets businesses sponsor specific neighborhoods at low cost ($1-3 CPM) and Nextdoor results show up in some hyperlocal LLM queries",
      "source_video_id": "EAG9XGII3mc",
      "source_bundle_id": "bundle_07",
      "specificity": "medium",
      "evidence_strength_in_source": "weak",
      "why_might_be_gold": "Hyperlocal verticals (towing, HVAC, locksmith) get 30+ neighborhood-targeted leads/mo at low CAC if claim is set up right",
      "smell_test_pass_or_fail": "inconclusive",
      "smell_test_reasoning": "Worth a 30-day pilot in 1 hyperlocal vertical · NOT MVP-7"
    },
    {
      "insight": "Expertise.com submission is free but editorial · requires customer to email a packet · the curation creates the 'top 5' authority signal",
      "source_video_id": "iRkfKHSNZOc",
      "source_bundle_id": "bundle_06",
      "specificity": "medium",
      "evidence_strength_in_source": "weak",
      "why_might_be_gold": "If WR can ghostwrite the Expertise submission packet, customer gets editorial trust signal for $0 platform-fee · scales well",
      "smell_test_pass_or_fail": "inconclusive",
      "smell_test_reasoning": "Editorial discretion is a real gate · acceptance rate unknown · need pilot test"
    }
  ],
  "evidence_gaps": [
    {
      "claim_in_corpus": "Expertise.com, CeDUR, MapQuest are LLM citation surfaces",
      "what_evidence_is_missing": "Zero substantive videos cover these · they appear in the topic brief but the corpus didn't surface meaningful coverage",
      "where_we_would_find_it": "primary_source",
      "blocks_or_just_weakens_claim": "weakens"
    },
    {
      "claim_in_corpus": "Trade-vertical associations (NRCA, RCAT, TRAA, ACCA, PHCC, NALP, ADA, state bars, Avvo, Justia) have specific membership/claim processes",
      "what_evidence_is_missing": "Zero meaningful coverage in the YouTube corpus of these specific trade associations as placement targets",
      "where_we_would_find_it": "primary_source",
      "blocks_or_just_weakens_claim": "blocks"
    },
    {
      "claim_in_corpus": "Yext SMB pricing is in the $199-499/mo range",
      "what_evidence_is_missing": "Yext's website pricing is gated · 'contact sales' · no public floor",
      "where_we_would_find_it": "external_lookup",
      "blocks_or_just_weakens_claim": "weakens"
    },
    {
      "claim_in_corpus": "BBB has agency-tier or partner API access",
      "what_evidence_is_missing": "Only 1 video alludes to BBB's internal automation · no public partner program documentation",
      "where_we_would_find_it": "primary_source",
      "blocks_or_just_weakens_claim": "blocks"
    },
    {
      "claim_in_corpus": "GBP Q&A API was deprecated Nov 3, 2025",
      "what_evidence_is_missing": "Single creator claim · needs direct Google API changelog verification",
      "where_we_would_find_it": "primary_source",
      "blocks_or_just_weakens_claim": "weakens"
    },
    {
      "claim_in_corpus": "Wikidata acceptance rate for local-business Q-items in 2026",
      "what_evidence_is_missing": "No survival-rate data in corpus · my 15-30% deletion-risk estimate is conjecture",
      "where_we_would_find_it": "operator_test",
      "blocks_or_just_weakens_claim": "weakens"
    },
    {
      "claim_in_corpus": "Apple Business Connect verification friction in 2026",
      "what_evidence_is_missing": "Most ABC walkthrough videos are 2023-2024 · 2026 friction unknown",
      "where_we_would_find_it": "operator_test",
      "blocks_or_just_weakens_claim": "weakens"
    }
  ],
  "questions_wave_should_resolve": [
    {
      "question": "Does Yelp's 2026 ToS explicitly prohibit agency-managed review-request campaigns?",
      "why_critical": "Pack value-add of review-velocity intersects with Yelp's review-suppression filter · if it triggers, customer's Yelp rating tanks",
      "how_to_answer": "drill_back_to_transcript",
      "related_video_ids": ["mn6aj3JitVo", "OuYU7n4wp1A"]
    },
    {
      "question": "What is the actual end-to-end labor time per Pack delivery in pilot 1-3?",
      "why_critical": "Margin floor of 70% depends on the labor number · skeptic estimates 3-5 hours/Pack, consensus assumes 1-2 hours",
      "how_to_answer": "operator_test",
      "related_video_ids": []
    },
    {
      "question": "Does WR's Texas-forensic Section 7 data hold up in HVAC-Florida and Dental-California?",
      "why_critical": "MVP-7 platform choice depends on whether the surface graph is vertical/region-stable",
      "how_to_answer": "operator_test",
      "related_video_ids": []
    },
    {
      "question": "Is there a BBB partner / agency API surface (private)?",
      "why_critical": "If yes, Pack delivery automates · if no, Pack remains manual orchestration",
      "how_to_answer": "external_lookup",
      "related_video_ids": ["gpv4esOq67A"]
    },
    {
      "question": "Should Foursquare REPLACE Yelp in the MVP-7?",
      "why_critical": "Foursquare is upstream-of-many and Yelp has ToS friction · the swap may improve placement ROI",
      "how_to_answer": "operator_test",
      "related_video_ids": ["cib1Mnl2Kfc"]
    },
    {
      "question": "What is the GBP suspension rate for customers receiving aggressive review-velocity campaigns?",
      "why_critical": "The promised 40/mo could destroy customer's GBP · upside is destroyed by downside risk",
      "how_to_answer": "operator_test",
      "related_video_ids": ["t_Radsgg-jU"]
    },
    {
      "question": "Is the AI-Citation Monitor a real recurring product or a feature of Marina?",
      "why_critical": "Bundling decision affects pricing strategy · standalone $79 vs Marina+placement-monitoring at $279",
      "how_to_answer": "operator_test",
      "related_video_ids": []
    }
  ],
  "risk_register": [
    {
      "risk": "Yelp ToS enforcement: WR identified as 3rd-party posting agent · customer's Yelp listing suppressed",
      "category": "platform_ban",
      "scenarios": [
        "Yelp filter flags WR-triggered review burst · customer reviews disappear from filter",
        "Yelp ToS team contacts customer claiming agency violation"
      ],
      "mitigations": [
        "Customer-driven OAuth where possible (no WR posting credentials)",
        "Review-velocity campaign goes through customer's own SMS/email · WR provides script, not the channel",
        "Explicit pre-pack disclosure to customer about Yelp policy",
        "Hold Yelp inclusion to OPTIONAL in MVP · default opt-out"
      ]
    },
    {
      "risk": "GBP suspension during over-aggressive review/optimization campaign",
      "category": "platform_ban",
      "scenarios": [
        "40-review-month triggers Google's fake-review detection",
        "Over-optimization (keyword stuffing in GBP description) triggers manual suspension"
      ],
      "mitigations": [
        "Pacing: 8-12 reviews/week consistent, not 40 in week 1",
        "Use Google's recommended review-request template",
        "Avoid keyword stuffing · Pack playbook bans this explicitly",
        "Insurance: keep Marina records of inbound calls/customers to prove review legitimacy"
      ]
    },
    {
      "risk": "Wikidata Q-item deletion challenge within 90 days",
      "category": "platform_ban",
      "scenarios": [
        "Wikidata community editor nominates Q-item for deletion due to notability concerns",
        "Q-item retained but stripped of substantial properties"
      ],
      "mitigations": [
        "Pre-create supporting citations in Wikidata-friendly sources (Crunchbase listing, business registry, news mention)",
        "Disclose paid-editor status explicitly per Wikidata rules",
        "Track 90-day survival rate · only productize if >75%"
      ]
    },
    {
      "risk": "FTC fake-review/endorsement enforcement",
      "category": "legal",
      "scenarios": [
        "WR review-request campaign templates language as 'leave a positive review' (compensated-review test)",
        "WR Pack includes paid placement disclosed-as-organic in any AI-cited content"
      ],
      "mitigations": [
        "Review-request templates are neutral · 'how was your experience' not 'leave 5 stars'",
        "All paid placements (Chamber, BBB, manufacturer cert) presented as PAID not earned",
        "No fake reviews · ever · period · contract clauses to that effect"
      ]
    },
    {
      "risk": "AI-Citation Monitor category saturation · margin compression in 12-18 months",
      "category": "saturated_market",
      "scenarios": [
        "Profound or Peec drops to $39/mo SMB tier",
        "Semrush bundles AI visibility into existing $99/mo Pro tier"
      ],
      "mitigations": [
        "Differentiation via auto-remediation (we don't just monitor · we fix)",
        "Bundle into Marina ($279/mo Marina+placement-monitor combo)",
        "Don't over-invest in the monitor as standalone product"
      ]
    },
    {
      "risk": "Manufacturer cert (GAF/OC) overpromise · customer expectation mismatch",
      "category": "reputation",
      "scenarios": [
        "Customer thinks they're paying WR for Master Elite certification",
        "Pack contract ambiguous · customer demands cert delivery"
      ],
      "mitigations": [
        "Sell 'Manufacturer Application Support' explicitly, not 'certification delivery'",
        "Contract clause: WR provides application support, customer responsible for qualifying directly with manufacturer",
        "Pricing reflects support not delivery"
      ]
    },
    {
      "risk": "Texas-forensic methodology becomes commodified",
      "category": "short_shelf_life",
      "scenarios": [
        "Competitor publishes their own forensic report 60-90 days after WR's",
        "Methodology copied · WR loses cold-email differentiation"
      ],
      "mitigations": [
        "Publish 1 new vertical-region forensic per quarter (HVAC-FL, Dental-CA, Towing-MidAtlantic, Legal-TX)",
        "Make methodology open but data proprietary",
        "Combine with Marina + Pack for end-to-end story competitors can't easily clone"
      ]
    },
    {
      "risk": "Pack delivery labor balloons · margin floor breached",
      "category": "capital",
      "scenarios": [
        "Hour estimate proves 2x too low",
        "Customer-friction support eats 5+ hours per Pack"
      ],
      "mitigations": [
        "Time-track first 3 pilots ruthlessly",
        "Standardize delivery via 7 runbook checklists",
        "Build customer-self-service portal for the 30% of Pack steps that don't need Wave touch",
        "Reprice or scope-cut if margin <60% after 5 pilots"
      ]
    }
  ],
  "questionable_claims_by_speaker_type": {
    "coach_speculation_count": 58,
    "vendor_pitch_count": 27,
    "operator_evidence_count": 31,
    "ratio_assessment": "Corpus skews coach-speculation-heavy (40% coach, 21% vendor, 24% operator). Coach speculation should be treated as hypothesis generation, NOT validation. Operator evidence (31 videos with actual case-study numbers) is the load-bearing subset for the Pack-strategy decision · weight it 3x heavier than coach content when in tension."
  },
  "audit_summary_for_wave": "The consensus map is DIRECTIONALLY correct: the AI-citation graph is real, the 7-platform MVP is a reasonable scope, and the price-and-margin math is plausible. BUT three things should NOT make it into the final report without flags: (1) the 76% margin floor is fragile · skeptic estimates real labor closer to 3-5 hours/Pack → actual margin 40-60% if not standardized · time-track pilots before promising margin. (2) Yelp inclusion is questionable · ToS friction, traffic decline, review-filter risk · consider swapping for Foursquare upstream-of-many. (3) Reviews velocity at 40/mo is creator doctrine, not operator-proven · per-vertical realistic targets needed · 40/mo could trigger GBP suspension. Three NEW elevations: (a) Foursquare should be IN the MVP-7 (currently only nominal); (b) 'Manufacturer Attention Campaign' deserves its own $497-997 product line (currently buried as singleton); (c) Salient/BBB private-API rumor needs 30-min lookup before MVP locks. Three platforms in the topic brief have ZERO meaningful corpus coverage: Expertise.com, CeDUR, MapQuest · do NOT mention as 'researched' in the final report · acknowledge as evidence gaps requiring follow-up. The Texas forensic finding is REAL but corroboration in this corpus is parallel-narrative, not independent · validate in 2 non-Texas verticals before scaling Pack to 100 customers. The product is shippable. The 76% margin is not."
}