Contents

Engagement: Local Marketing Playbook, Phases 1 & 2 (18 deliverables)
QA performed: August 27, 2026
Scope: 29 files, ~21,400 lines, ~1.5 MB. Full-corpus grep/parse sweeps plus targeted reads and live verification of external citations.
Verdict: DO NOT SHIP AS-IS. The Phase 2 content is largely sound and unusually well-disciplined about fabrication. The Phase 1 technical track is not internally consistent, and several deliverables were written against site facts that a sibling Phase 1 deliverable had already disproved. Executing the migration plan as written would 301 URLs that do not exist while leaving every real URL unredirected.

Counts: 12 Blockers · 17 Should-fix · 9 Notes · ~256 open client questions

BLOCKERS

Must be resolved before anything ships to the client or to the site.

B1. The URL migration map is built on slugs that do not exist

Files: Phase 1 - Foundation/Site Architecture Review.md §2.4, §4.1 (lines 52–66, 272–283) vs Phase 1 - Foundation/SEO Site Audit.md §4 (lines 395–407, 550–564)

The Site Architecture Review's redirect table lists 11 page-level 301s. Seven source URLs are marked [CONFIRM] — i.e. guessed. The SEO Site Audit, working from an actual crawl of the same site on the same day, lists the real slugs. They do not match:

Four live URLs — /educational-fly-fishing-retreats, /private-fly-fishing-lessons-washington, /washington-casting-camp, /nate — appear nowhere in the migration map at all.

Fix: Rebuild §4.1 of Site Architecture Review.md from the SEO Site Audit's crawled inventory (lines 395–407) as the single source of truth, and delete every [CONFIRM] slug guess. Add rows for the four unmapped live URLs. Do not execute any redirect until this table is reconciled.

B2. Two Phase 1 deliverables prescribe different destinations for the site's most important URL

Files: Site Architecture Review.md §4.1 row 1 (line 275) vs SEO Site Audit.md M3 (lines 570–583)

• Site Architecture: /about-the-science → /method
• SEO Site Audit: /about-the-science → /trout-psychology-method

Every downstream Phase 2 deliverable (126 references) uses /method. Executing both instructions creates a redirect chain on the single highest-value page in the engagement.

Fix: Adopt /method (it is what 126 downstream references assume). Rewrite SEO Site Audit.md M3 steps 1–2 to match, or explicitly mark M3 as superseded.

B3. Conflicting redirect target for /waretreats, and the real retreats page is ignored

Files: SEO Site Audit.md C3 (lines 173–186, 932) vs Site Architecture Review.md §4.1 row 3 (line 277) vs Phase 2 - Content/Service Pages/05 - Group Retreats.md (lines 6–7, 370)

• SEO Site Audit: /waretreats → /educational-fly-fishing-retreats (an existing live page, title "3-Day Fly Fishing Retreats in Washington", SEO Site Audit line 405).
• Site Architecture + the Group Retreats deliverable: /waretreats → a new /trips/group-retreats.

The Group Retreats service page claims to "replace /waretreats" and never mentions /educational-fly-fishing-retreats, which means shipping it creates a duplicate of a page already indexed, with no redirect between them.

Fix: Decide one destination. If /trips/group-retreats is canonical, add /educational-fly-fishing-retreats → /trips/group-retreats 301 to the map and say so in 05 - Group Retreats.md §8.

B4. The two Phase 1 deliverables disagree on whether to migrate URLs at all

Files: SEO Site Audit.md M2 (line 564) vs Site Architecture Review.md §7 (lines 512–609)

SEO Site Audit M2, on the concatenated slugs: "Do not bulk-rename these. The ranking gain is small and every rename costs a redirect and risks losing accrued signal. Apply the correct convention to new pages only."

The Site Architecture Review prescribes a 12-week, four-tranche migration that bulk-renames exactly those slugs.

Fix: The agency must take one position. Reconcile before the client reads both documents in the same folder.

B5. The migration is hard-gated on an issue the client profile ADDENDUM verified as a non-issue

Files: Site Architecture Review.md (lines 81, 521, 615), Keyword Research.md (lines 459, 492, 496, 521), Analytics Setup Guide.md (lines 59, 488, 824, 838), Schema Markup.md (line 1979) vs 00 - Client Profile.md ADDENDUM (line 150) and SEO Site Audit.md §3 (lines 763–776, 967)

The ADDENDUM and the SEO Site Audit both verify: the HTTP 429s did not reproduce — 8 unspaced Googlebot-UA requests all returned 200, response times improved across the burst, and ~30 audit requests produced zero 429s. The ADDENDUM's Revised #1 priority explicitly states "Not the 429."

Four other deliverables still call it the top blocker:
- Site Architecture §7 Phase 0.1: "This is a hard gate — do not proceed until resolved." The entire 12-week plan cannot start.
- Analytics Setup §4: "This is a higher-priority fix than anything else in this document."
- Keyword Research: "Blocker #5 should be checked before anything else in this engagement."

Fix: Demote to a Search Console crawl-stats spot-check in all four files. Remove the Phase 0 hard gate in Site Architecture Review.md line 521 and the risk-register row at line 615.

B6. Three deliverables prescribe a 302→301 fix for a redirect that is already a 301

Files: Analytics Setup Guide.md (lines 56, 481, 862), Site Architecture Review.md (lines 27, 80, 291, 536), Keyword Research.md (lines 398, 491, 521) vs 00 - Client Profile.md ADDENDUM (line 149), Citation Audit Report.md (lines 34–36), SEO Site Audit.md (line 764), GBP Optimization Report.md (line 70)

Three independent verifications (direct HTTP header inspection, root and deep paths) confirm streamsidecovenco.com returns HTTP 301. Three deliverables still describe it as a 302 and schedule a fix.

Knock-on: Site Architecture Review.md line 27 states the scope as "12 URL changes (11 page-level 301s + 1 domain-level redirect-type fix)." The domain fix is a no-op, so the real scope is 11.

Fix: Delete the 302 language and the remediation tasks from all three files; correct the scope count in the Site Architecture executive summary.

B7. The GBP Q&A pack publishes service commitments every other deliverable holds for sign-off

File: Phase 1 - Foundation/GBP Optimization Report.md §13 (lines 741–835)

§13 supplies 20 ready-to-post Q&A answers with posting instructions ("post 3–5 per week"). Several make binding commitments that other deliverables explicitly gate:

This is the single largest overclaiming exposure in the package, and it is the one most likely to be published first — GBP Q&A takes minutes to post and is publicly visible immediately.

Fix: Mark §13 Q3, Q13, Q16, Q17, Q18, Q19, Q20 and Q22 (pricing) as ⛔ HOLD — Kristen sign-off required, matching the FAQ deliverable's own convention. Only Q1, Q2 and the purely factual river/season answers should be cleared for posting now.

B8. Prices sourced from a stale test booking page are published as current rates

Files: GBP Optimization Report.md line 773 vs Citation Audit Report.md lines 183–184 vs Service Pages/02 - Virtual Fly Casting Lessons.md line 369

The GBP Q&A publishes "a private education day has been listed at $750 for an individual, with a split option around $350 per person" and "$250 for four weeks of unlimited video analysis."

• The $750/$350 figures come from the Picktime page that the Citation Audit itself flags as carrying test data ("Sample Service", "Puget Sound School - Test Class" ($150)) under the legacy business name and an old Gmail address.
• Service Pages/02 line 369 states flatly: "The $250.00 / month price is the one confirmed price in this content set … Everything else on every other page stays price-free until the rate card is confirmed."
• The GBP wording "$250 for four weeks" also contradicts the confirmed product shape, which Service Pages/02 line 175 defines as "a subscription, not a per-session fee."

Fix: Strip $750/$350 from GBP Optimization Report.md line 773 and replace with a pointer to the booking page. Change "$250 for four weeks" to "$250 per month, a subscription."

B9. Blog publish dates precede the pages they publish to

Files: Phase 1 - Foundation/Content Calendar.md (lines 174–273, 1020, 1026) vs Site Architecture Review.md §7 (lines 515–600)

The Site Architecture plan builds /journal + the four category pages in Phase 1, Weeks 1–2, and moves /about-the-science → /method in Tranche D, Week 9. No calendar date is anchored to "Week 0" anywhere in the document.

• Post #1 publishes Tue Sep 1, 2026 — five days after the engagement start date on every deliverable (Aug 27), i.e. before /journal exists.
• Site Architecture Phase 1.7 says "Publish the first 2 posts." The Content Calendar publishes three in September.
• Post #5, the cornerstone, publishes Oct 20 and carries an explicit DEPENDENCY: /method and /method/think-like-a-trout must be live before this publishes. /method does not exist until Week 9 — roughly two weeks after Post #5's publish date.

Fix: Anchor Week 0 to a real calendar date in Site Architecture Review.md §7, then re-time the Content Calendar against it. Either pull /method forward out of Tranche D, or slip Post #5.

B10. Every CTA button in the five service pages points at a URL with no redirect planned

Files: all five Phase 2 - Content/Service Pages/*.md; Site Architecture Review.md §4.1

The service pages use /book (29 references) and /gift-certificates (21 references) as primary and secondary CTAs. The live slugs are /booknow and /giftcertificates. /giftcertificates appears nowhere in the redirect map, and /book is only reachable via a redirect row whose source is wrong (/book-and-contact, per B1).

Fix: Add /booknow → /book 301 and /giftcertificates → /gift-certificates 301 to §4.1, or change the service-page CTAs to the live slugs and defer the rename.

B11. The Review System prescribes self-serving review schema that Google's guidelines prohibit

File: Phase 1 - Foundation/Review Generation System.md line 1115

"Schema markup — Review and AggregateRating on the LocalBusiness schema once you're past ~10 reviews. This is what gets the star rating into search results."

Google's structured-data policy disallows self-serving reviews for LocalBusiness and Organization; the markup is ignored at best and is a manual-action risk at worst. Schema Markup.md correctly ships no aggregateRating anywhere in its 13 blocks — so the two Phase 1 deliverables contradict each other, and only one is right.

Fix: Delete line 1115's schema recommendation and replace with a note that Google review stars come from the GBP listing, not from on-site markup.

B12. /waretreats is recorded as a verified live page in the IA

File: Site Architecture Review.md §2.4 line 91 (| /waretreats | Group Retreats | Verified? **Yes** | Business value: Medium |)

Verified 404 per 00 - Client Profile.md ADDENDUM line 151 and SEO Site Audit.md line 173. The IA assigns medium business value and migration planning to a page that does not exist.

Fix: Change to 404 — not live and remove it from the business-value column.

SHOULD-FIX

S1. The email-platform recommendation ignores a Mailchimp install its own Phase 1 audit found

Files: Phase 2 - Content/Email Marketing System.md §2 (lines 88–122, 120) vs SEO Site Audit.md (lines 672–678, 857, 952)

The Email Marketing System recommends Kit (ConvertKit), describes the list as "growing from roughly nothing," and files [CONFIRM: does an email list exist today, how large, and where does it live?]. The SEO Site Audit already answered part of that: chimpstatic.com (a Mailchimp popup) is live in the site <head>, with a Meta Pixel alongside it. Analytics Setup Guide.md line 863 also lists the platform as unknown.

An existing Mailchimp account with an accumulated list changes the recommendation materially (migration cost, list warm-up, existing subscriber consent).
Fix: Cross-reference the SEO Site Audit finding in Email Marketing System.md §2 and reframe the recommendation as "migrate from Mailchimp to Kit, or stay," with the list export as the deciding input.

S2. Known collision — Email System supersedes the Review System on platform and timing

Files: Email Marketing System.md lines 116, 1879 vs Review Generation System.md lines 33, 56, 148

Documented by the authors and correctly reasoned, but the Review Generation System has not been edited to reflect it: it still names "Squarespace Campaigns / email tool" and a $20/mo tool (Mailchimp, Brevo), and still states "0 to roughly 8–15 reviews in the first three weeks." The Email System changes that to Day 8–29 / four to five weeks.
Fix: Edit Review Generation System.md lines 148 and 56 in place. Superseding language in one file is not a fix when both files ship in the same folder.

S3. NAP canonical format conflicts with every other deliverable

File: Citation Audit Report.md lines 88–91

The Citation Audit declares the canonical format (206) 659-5501 and lists 206-659-5501 as a variant to avoid. Every other deliverable — GBP report body copy, all five service pages, the FAQ, the Social Calendar, the Email System, the Analytics guide, and the client profile itself — uses 206-659-5501. That is 60+ instances against the declared canon.

The digits are correct everywhere; only the format is inconsistent. But citation consistency is the entire point of that deliverable.
Fix: Change the Citation Audit's canonical to 206-659-5501 to match the corpus, or reformat the corpus. The Citation Audit is the lower-cost edit.

S4. Schema phone format violates the rule the schema deliverable itself sets

Files: Schema Markup.md line 1922 vs Service Pages/01–05 (lines 310, 289, 290, 279, 288) and SEO Site Audit.md line 88

Schema Markup.md lists "Phone as 206-659-5501 → Should be E.164: +12066595501" as a common failure mode, and uses +12066595501 in all four of its own occurrences. All five service pages and the SEO Site Audit's sample block use "+1-206-659-5501".
Fix: Normalise the six blocks to +12066595501.

S5. "Roughly 40 years across the team" is published copy resting on an unanswered question

Files: Service Pages/01 lines 165, 218; Service Pages/05 line 201; FAQ Content and Schema.md line 432; GBP Optimization Report.md lines 781, 907; Citation Audit Report.md line 602 vs Team Bios.md lines 253–254

Kristen has 25 years. The "~40 years" figure therefore implicitly asserts Nate has ~15. Team Bios.md carries [NATE: years fly fishing] and [NATE: years guiding / instructing] as open questions, and 00 - Client Profile.md line 33 records the 40-year figure without a [CONFIRM] flag despite being an inference.

Seven public-facing instances of a team credential nobody has verified.
Fix: Hold the "40 years" line out of all published copy until Nate answers Team Bios items. It is trivial to answer and trivial to remove.

S6. Two competing FAQPage schemas for the same /faq page

Files: Schema Markup.md §8 (line 1662, 12 entities) vs FAQ Content and Schema.md (line 1119ff) and faq-schema.json (25 entities)

Both are FAQPage blocks targeting /faq. The 25-entity version is publish-ready; the 12-entity version is stub-filled ({{ANSWER — CONFIRM WITH KRISTEN}}). Neither file says which supersedes. Only one can ship.

Note also that the 12-entity block contains "Do you teach neurodivergent anglers?" — a question the FAQ deliverable puts on hold (see B7).
Fix: Mark Schema Markup.md §8 as superseded by faq-schema.json, using the same explicit language Team Bios.md line 361 uses when it supersedes Schema Markup.md §2.

S7. Structured data broadcasts the exact entity signal the strategy exists to suppress

Files: Schema Markup.md lines 162, 493–495; Team Bios.md lines 494–496

The Revised #1 priority in the ADDENDUM is entity disambiguation away from clinical psychology. The Organization and Person knowsAbout arrays declare, in machine-readable form:
- "Neurodiversity-informed outdoor instruction" (Organization)
- "Neurodiversity-informed instruction", "Autism spectrum education", "Mental health and the outdoors" (Person)

Two problems. First, "Mental health and the outdoors" in a knowledge-graph node for a business named Trout Psychology actively feeds the misclassification the GBP report calls "the single highest-risk field on the profile." Second, "Autism spectrum education" ships a neurodivergence claim in structured data while FAQ Content and Schema.md, Service Pages/03, Blog Post 04 and Blog Post 05 all gate that claim behind Kristen's sign-off.
Fix: Remove "Mental health and the outdoors" and "Autism spectrum education" from both knowsAbout arrays. Retain "Behavior analysis" and "Applied behavior analysis" — those are credentialed and unambiguous. Gate the neurodiversity strings on the same sign-off as the prose.

S8. Blog Post 04's title tag and meta description work against the disambiguation goal

File: Blog Posts/04 - fly-fishing-and-the-nervous-system.md lines 20–21

Title tag: Fly Fishing and Mental Health: The Science | Trout Psychology
Meta description: …the honest limits of "fishing is my therapy."

The post's body handles this exceptionally well (line 95: "Trout Psychology is fly fishing instruction informed by behavior analysis. It is not therapy. It is not clinical treatment. It is not a mental-health service, and I am not your clinician."). But the title tag and meta are what Google indexes and what an AI answer engine quotes, and they pair the brand name with "Mental Health" and "therapy" with no disclaimer in range.
Fix: Retitle to something like Why a Day on the River Resets You: The Science | Trout Psychology and rewrite the meta to lead with attention restoration rather than "therapy." Keep the body exactly as written.

S9. Service Pages 03 and 04 cite the wrong migration tranche

Files: Service Pages/03 line 5, Service Pages/04 line 5 vs Site Architecture Review.md §7 (lines 565–585)

Both headers say Tranche C. Site Architecture puts /casting-camp and /private-education in Tranche B (Weeks 6–7); Tranche C is /casting-lessons alone in Week 8, explicitly "Alone because this is the $250/mo product." Following the service pages would break the isolation that Tranche C exists to provide.
Fix: Change both to Tranche B.

S10. YouTube ownership is resolved in one deliverable and still open in two others

Files: Video Strategy.md §0.2 (lines 30–40) vs Social Media Calendar.md lines 96, 1565 and Schema Markup.md lines 86–88

Video Strategy.md verified it properly — handle @Troutpsychology held, channel ID UCJap_s1ZkyiXBliN8Eun6Dg, display name "Kristen Bufe", created July 31 2025, zero public videos, RSS empty. It also established that the existing "Thinking like a Trout!" video belongs to an unaffiliated third party (@ricksimon277).

Social Media Calendar.md and Schema Markup.md still carry [CONFIRM channel ownership], and the Social Calendar warns "If it does not, that is a Phase 1 problem." It isn't. It's answered.
Fix: Close the [CONFIRM] in both files and cite Video Strategy.md §0.2. Fill {{YOUTUBE_CHANNEL_URL}} in Schema Markup.md with the verified channel.

S11. Customer-facing templates hardcode an email address that may not exist

Files: Email Marketing System.md lines 6, 1462, 1529, 1855; Review Generation System.md lines 297, 766, 778 vs Citation Audit Report.md line 118

kristen@troutpsychology.com is the send-from identity in the email system and appears inside customer-facing review-response copy. The Citation Audit lists creating it as an open [CHECK] item and "item 1 in §14 because it blocks the build" — the business currently runs on streamsidecovenco@gmail.com.
Fix: Flag every instance as dependent on Citation Audit §14 item 1, or template it as {{SEND_FROM_ADDRESS}} until the mailbox exists.

S12. Content Calendar publishes traffic forecasts the Keyword Research forbids

Files: Content Calendar.md §7 (lines 960–970) vs Keyword Research.md §1 (line 17)

Keyword Research's data-honesty statement: "These bands are estimates, not measurements… Do not put them in a client-facing forecast, a traffic model, or a revenue projection."

The Content Calendar publishes an unqualified forecast table: 150 → 600 → 1,500 → 2,500 blog organic sessions/month, 180 keywords ranking top 20, +800 email subscribers. No methodology, no estimate label, no baseline (two rows read [CONFIRM]).
Fix: Label the table DIRECTIONAL TARGETS — NOT A FORECAST and add the derivation, or cut the session and keyword rows until GSC data exists.

S13. The GBP Q&A protocol instructs vote manipulation

File: GBP Optimization Report.md line 749

"Then upvote the most important answers from a few accounts; the highest-upvoted Q&A is the one that displays by default."

Coordinating upvotes across accounts to control what displays is fake-engagement under Google's Maps user-contributed-content policy, on a profile the same document says is at elevated suspension risk (line 48, line 141).
Fix: Delete the upvote instruction. Seeding a question from a personal account and answering as owner is standard practice; the upvote ring is not.

S14. Secondary GBP categories disagree between two Phase 1 deliverables

Files: Review Generation System.md line 31 vs GBP Optimization Report.md §6.3 (lines 291–300)

Review System: "use Tour Operator and Sports School as secondaries" — two categories.
GBP report: five to seven, ranked, with Fishing camp as the #1 secondary (it covers retreats, Casting Camp, Streamer Stomp and the schools) — omitted entirely from the Review System's list.
Fix: Point Review Generation System.md line 31 at GBP Optimization Report.md §6.3 rather than restating a partial list.

S15. ZIP 98922 is canonical in one deliverable and unconfirmed in two others

Files: Citation Audit Report.md line 75 vs GBP Optimization Report.md line 76 and Schema Markup.md line 89

The Citation Audit prints ZIP: 98922 in its canonical NAP block with no caveat. The GBP report flags it [CHECK] — 98922 is the Cle Elum ZIP, but the business's actual mailing ZIP is unconfirmed. Schema Markup flags {{POSTAL_CODE}} the same way. Building citations against an unconfirmed ZIP creates exactly the NAP inconsistency the deliverable exists to prevent.
Fix: Flag it in the Citation Audit's canonical block and gate citation building on the answer.

S16. /method/neurodiversity-informed-instruction is linked from four deliverables but scheduled nowhere

Files: Service Pages/03 lines 30, 229, 237; FAQ Content and Schema.md lines 577, 1337; Blog Posts/04 line 159; Blog Posts/05 lines 232, 320

Twenty references to a page that has no build slot in Site Architecture Review.md §7, no entry in the Content Calendar.md, and no copy anywhere in the package. All four references correctly gate on it being "live and approved," which means four deliverables ship with a dead dependency.
Fix: Either add it to the Phase 1.6 /method/* spoke build with a ⛔ HOLD status, or state plainly in each reference that it is Phase 3 scope.

S17. Site Architecture's own URL count is internally inconsistent

File: Site Architecture Review.md line 20 vs lines 52–66 and SEO Site Audit.md lines 228, 791

Line 20: "Three different conventions across five known URLs." The nav tree immediately below lists 14. The SEO Site Audit reports the XML sitemap contains 22 URLs. The Site Architecture Review marks everything outside its five verified entries as "Verified? No — Unknown until crawled" — but a crawl had already been done, in the sibling deliverable, on the same day.
Fix: Replace §2.4 with the SEO Site Audit's crawled inventory. This is the root cause of B1.

NOTES

N1. External citations — verified clean

Spot-checked all thirteen scholarly and agency citations across the six blog posts. Every one is real, correctly attributed, and says roughly what the posts claim:

All twelve URLs resolve; the 403s returned by Wiley, Science, PNAS, OUP, Taylor & Francis, ScienceDirect and Company of Biologists are standard publisher bot-blocks, not dead links. No fabricated citations found. This is the strongest part of the package.

N2. The USBR flip-flop source is a 2008 press release, and the posts don't say so

Files: Blog Posts/01 line 190, Blog Posts/02 lines 45, 152

usbr.gov/newsroomold/newsrelease/detail.cfm?RecordID=24341 is real and does contain the quoted purpose statement and the cfs figures (Cle Elum to 200–250 cfs by ~Sept 15, Rimrock to ~2,000 cfs by mid-September, ~45-day operation) — but it is dated August 24, 2008. Post 02 does caution readers that "exact dates and flow targets change every single year" (line 49), which is the right instinct; it just doesn't disclose that the numbers are eighteen years old.
Fix: Add "(2008 operating year)" to the figures in Post 02 line 45 and to the source table.

N3. JSON-LD — all 30 blocks parse

Extracted and strict-parsed every <script type="application/ld+json"> block and every fenced json block across all 29 files. 30 blocks, 30 parse cleanly. faq-schema.json parses. The only regex hit for an unclosed script tag was Schema Markup.md:1910, which is prose describing the rule, not markup — a false positive. Schema Markup.md's self-reported claim of "All 13 blocks parse cleanly" is accurate.

However: 14 of the 30 blocks contain live {{PLACEHOLDER}} tokens (see the dependency register below). Schema Markup.md line 1912 warns correctly that "substituting values is exactly when JSON breaks." Nothing in the package requires a re-validation step after substitution. Fix: add a mandatory Rich Results Test gate to the launch checklist.

N4. No fabricated ratings, review counts, follower counts, traffic figures or testimonials

Searched the full corpus for aggregateRating, ratingValue, reviewCount, follower/subscriber counts, Domain Authority, traffic estimates and quoted testimonials. Zero fabrications found. The discipline here is genuinely good and worth preserving:
- Competitor Intelligence.md line 11 opens with an explicit statement that no traffic, DA, backlink or follower numbers appear because none are observable without a paid tool; the two review counts it does cite (Yelp 50, Birdeye 526) carry source URLs.
- Social Media Calendar.md line 16 refuses to publish engagement benchmarks without Insights access and labels every posting time a hypothesis.
- Social Media Calendar.md line 1476: "Writing a plausible-sounding testimonial is not an option, ever."
- Keyword Research.md §1 refuses to invent search volumes.
- Analytics Setup Guide.md line 157 labels the $1,500 LTV a placeholder and schedules its replacement.

N5. No invented street address anywhere

Grepped for streetAddress, US street patterns and ZIP codes across all 29 files. No fabricated address. Schema Markup.md line 55 handles it correctly and explicitly: "Inventing one would be a fabricated business fact and would also break Google Business Profile consistency later." The only ZIP that appears is 98922 (see S15).

N6. Business name is consistent; only the YouTube handle varies in case

Trout Psychology is used consistently across 600+ instances. Trout Psychology LLC and Trout Psychology Fly Fishing appear only inside do-not-use lists in the GBP and Citation deliverables — correct usage. One case variant: @Troutpsychology (Video Strategy.md line 32) vs @troutpsychology (25 instances elsewhere). Handles are case-insensitive; harmless, but pick one for display.

N7. British spellings in publish-ready American copy

FAQ Content and Schema.md (43 instances) and faq-schema.json (9) use behavioural, specialises, practises, licence, organiser. These are customer-facing and go into structured data for a Washington State business. Video Strategy.md (38) and Competitor Content Gap Analysis.md (13) are internal and matter less. Fix: Americanise the FAQ deliverable and the JSON before publish.

N8. Blog Post 04 carries no crisis resource line

The disclaimer at line 95 is excellent and the FAQ at line 143 correctly directs readers to a qualified clinician. For a post that will rank on "can fly fishing help with anxiety," a single line naming 988 (US Suicide & Crisis Lifeline) costs nothing and closes the last gap. Not a legal requirement; a reasonable duty of care.

N9. Calendar dates check out

Programmatically validated every weekday/date pair across all deliverables — 30 numeric (Tue 9/17) and all long-form (Tue Sep 1, 2026) pairs. Zero mismatches. The six blog posts' slot numbers, titles, slugs and publish dates match Content Calendar.md §5 exactly. The internal consistency of the editorial calendar is good; only its relationship to the migration schedule is broken (B9).

UNRESOLVED DEPENDENCY REGISTER

~256 distinct open questions. Deduplicated by text; grouped by owner and by what they block.

By owner

KRISTEN — commercial and policy facts (149 [CONFIRM] + 21 [KRISTEN] = ~170)

Tier 1 — blocks publication of finished copy (approx. 45 items)
- The entire rate card. Guided full day / half day / shared "split with a stranger" / walk-and-wade; private education 1-, 2- and 3-day; Casting Camp; retreats; schools; Streamer Stomp. Currently {{PRICE_FULL_DAY}}, {{PRICE_SHARED}}, {{PRICE_PRIVATE_ED}}, {{PRICE_RETREAT}}, {{PRICE_CAMP}}, {{PRICE_SCHOOL}}, {{PRICE_STOMP}}. Blocks: all 5 service pages, 7 Service schema blocks, 6 FAQ answers, GBP Q&A Q22, the entire email system's booking copy. $250/mo virtual is the only confirmed price in the package.
- Cancellation, deposit, weather and refund policy. FAQ lines 250, 264, 952; Email System lines 626, 1204; Service Pages/02 line 163, /04 line 170, /05 line 167; Schema Markup.md FAQ stub.
- What gear is provided vs. brought. FAQ 189, 492; Email System 656, 669; Service Pages/01 188, /04 178; Schema Markup.md FAQ stub; GBP Q3/Q19 (currently published as fact — see B7).
- Virtual subscription mechanics. Video submissions per month, live-session count and length, turnaround time, annotation platform, minimum commitment, cancellation terms. Service Pages/02 lines 108, 116, 175, 179, 371; FAQ 647, 661, 704; Email System 905, 911, 988.
- Casting Camp and Streamer Stomp definitions. Length, dates, group size, venue, on-water vs. grass. FAQ 766, 780; Service Pages/04 88, 349. Streamer Stomp is the brand's signature program and currently has no defined product shape anywhere in the package.
- Retreat logistics. 2027 dates, lodging and meals, non-angler partners, group-size flexibility, invoicing. Service Pages/05 87, 139, 156, 167, 175, 187.
- Which booking tool is actually in use. Gates the Kit/Zapier automation and the Picktime cleanup. Email System 103; Social Calendar 1639.

Tier 2 — professional and legal territory, agency has deliberately drafted nothing (approx. 12 items)
- The five accessibility / neurodiversity answers. FAQ D1–D5 (lines 562, 585, 603, 617); Service Pages/03 line 144; Blog 04 line 159; Blog 05 lines 232, 320. The FAQ deliverable is explicit: "Any specific accommodation named here becomes a commitment." See B7 — the GBP Q&A has already breached this hold.
- Women-only programs — offered as a bookable product, or not? FAQ 548; Service Pages/04 151, /05 136, 251.
- Legal registrant — Trout Psychology or Streamside Coven Co.? {{LEGAL_NAME}}; Client Profile line 16; GBP line 77; Keyword Research line 534. Blocks GBP verification.
- Mailing address and ZIP for GBP verification (hidden from the public profile), plus a CAN-SPAM postal address for email. GBP 76, 78; Email System 1391, 1951; {{POSTAL_CODE}}.

Tier 3 — narrative inserts, only Kristen can write (7 items)
[KRISTEN:] first-person story blocks in all six blog posts (lines 66, 47, 119, 77, 37, 98) plus the Video Strategy script inserts (436, 618, 814, 872). Each is 100–150 words. The blog posts cannot publish without them.

Tier 4 — biography and credentials (Team Bios, 6 items)
- {{UNIVERSITY_NAME}} — the B.S. institution. Team Bios.md line 69 calls it "the highest-value single missing fact in the whole engagement."
- Casting/guiding certifications — FFI CCI, guide licence, WFR, swiftwater, CPR (line 74).
- Phonetic pronunciation of "Bufe" (line 207).
- Whether anyone beyond Kristen and Nate is on the team this season (line 755).

NATE — 19 items, all in Team Bios.md

Years fly fishing; years guiding/instructing; certifications and licences; background narrative (2–5 sentences, three separate variants at lines 235, 267, 323); hometown and current base; teaching specialism; one non-fishing fact; Instagram handle or an explicit "don't create one"; LinkedIn (optional); what he wants his channel known for; whether he's being approached for media.

Blocks: the Person schema for Nate ({{NATE_PHOTO_URL}}, {{NATE_INSTAGRAM}}, {{NATE_BIO_PAGE_URL}}), the /about/nate-tamarack page, Blog Post 06's co-byline, and — per S5 — the "~40 years across the team" claim currently live in seven pieces of public copy.

AGENCY — 66 {{PLACEHOLDER}} tokens + 1 [NEEDS SOURCE]

Assets to produce or collect (23 tokens): {{LOGO_URL}}/{{LOGO_WIDTH}}/{{LOGO_HEIGHT}}, {{HERO_IMAGE_URL}}, {{KRISTEN_PHOTO_URL}}, {{NATE_PHOTO_URL}}, {{METHOD_IMAGE_URL}}, {{TRIP_IMAGE_URL}}, {{VIRTUAL_IMAGE_URL}}, {{PRIVATE_ED_IMAGE_URL}}, {{RETREAT_IMAGE_URL}}, {{CAMP_IMAGE_URL}}, {{SCHOOLS_IMAGE_URL}}, {{STOMP_IMAGE_URL}}.

URLs that depend on the migration being settled (17 tokens): {{METHOD_PAGE_URL}}, {{GUIDED_TRIPS_URL}}, {{VIRTUAL_URL}}, {{PRIVATE_ED_URL}}, {{RETREAT_URL}}, {{CAMP_URL}}, {{SCHOOLS_URL}}, {{STOMP_URL}}, {{BOOKING_URL}}, {{SUBSCRIBE_URL}}, {{FAQ_PAGE_URL}}, {{ABOUT_PAGE_URL}}, {{KRISTEN_BIO_PAGE_URL}}, {{NATE_BIO_PAGE_URL}}. Every one of these is blocked by B1–B4.

Verifiable externally (6): {{FLY_FISH_JOURNAL_URL}} and {{PODCAST_EPISODE_URL}} — both already resolved in Citation Audit Report.md lines 492–493 and just need copying across. {{YOUTUBE_CHANNEL_URL}} — resolved in Video Strategy.md §0.2 (see S10). {{LATITUDE}}/{{LONGITUDE}}/{{SERVICE_RADIUS_M}} — derivable.

[NEEDS SOURCE] (1): Blog Posts/01 line 195 — conditional. Only triggers if Kristen wants a specific Yakima water-temperature figure stated as a number rather than a range; the USGS gauge is already named as the source. Correctly handled.

RECOMMENDED SEQUENCE

1. Reconcile the URL truth. B1, B2, B3, B4, B12, S17 are one problem with one fix: make the SEO Site Audit's crawl the single source of truth and rebuild the Site Architecture map from it. Nothing downstream — schema, service pages, CTAs, the content calendar — can be finalised until this is settled.
2. Purge the stale technical facts. B5, B6. Four files still treat a disproved 429 and an already-correct 301 as blockers. Fifteen minutes of editing.
3. Pull back the overclaims. B7, B8, S7, S8, B11, S13. Gate the GBP Q&A pack, strip the Picktime prices, clean the knowsAbout arrays, retitle Blog 04.
4. Send Kristen the Tier 1 + Tier 2 list (~57 items) as a single structured request. Everything else queues behind it.
5. Send Nate his 19 items. They are quick and they unblock a public credential claim already in circulation.
6. Re-validate every JSON-LD block after placeholder substitution (N3), then ship.

REMEDIATION LOG — Aug 27, 2026

Scope of this pass: the seven remediation items assigned against Blockers B1–B10 and B12, plus S3, S9, S12 and S17 where they were the same edit. B11 was not in scope and remains open.

Method: surgical in-place edits (python3 exact-string replacement) — no document was rewritten from scratch and no document was truncated. Line counts before → after are recorded per file below; no file lost a line, and the only file with net movement of existing content (GBP Optimization Report.md) was verified afterwards to still contain all 21 Q&A entries and every draft answer.

Blockers — disposition

Should-fix items closed incidentally

• S3 (phone format) — Citation Audit Report.md §2.1 now scopes (206) 659-5501 to citation and directory submissions specifically, and adds an explicit three-format / three-context note stating that 206-659-5501 in body copy and tel: links is correct and is NOT an inconsistency, and that +12066595501 is correct in JSON-LD. It also explicitly warns against a corpus-wide find-and-replace, which would break the E.164 schema values.
• S9 (wrong tranche in Service Pages 03 and 04) — both corrected to Tranche B, and both now name the real legacy slug (/private-fly-fishing-lessons-washington, /washington-casting-camp) instead of the invented one, with a note to preserve the keyword/geo-modifier the old slug carried. Service Page 02 corrected to /castinglessons, Tranche C, Week 9.
• S12 (unlabelled traffic forecast) — Content Calendar.md §7 is now headed "DIRECTIONAL TARGETS — NOT A FORECAST" with a stated derivation for the 2,500 sessions/month and 180-keywords figures (a decelerating compounding assumption, not a bottom-up volume model), an explicit instruction not to use it in any client-facing forecast or revenue projection, a per-row Status column separating the two controllable rows from the six unvalidated ones, and a replace-with-GSC-data trigger at 90 days.
• S17 (internal URL-count inconsistency) — resolved by the §2.4 rebuild; the "five known URLs" line now cites the 22-URL crawl.

Still open — genuinely requires Kristen

These are not oversights in this pass. They are decisions only the client can make, and each is now blocked visibly rather than silently answered:

1. The seven held GBP answers (§13.2). Pricing, women-only programs, weather/reschedule policy, neurodivergent instruction, accessibility, gear list, gift-certificate scope. Seven independent sign-offs — partial approval is fine. The neurodivergent (Q17) and accessibility (Q18) drafts were rewritten to match FAQ Content and Schema.md D2/D3: unbounded accommodation promises ("we will build the day around it", "we can often find an approach that works") were downgraded to possible to discuss, the boat/moving-water constraint was named explicitly, and D2's required non-clinical boundary line was added. They are still held.
2. The rate card. $250/month remains the only confirmed price in the package. Until Kristen supplies real figures, §13.2 Q6 cannot post.
3. The Picktime page cleanup. The $750/$350 provenance problem is upstream of the GBP text — the booking page still carries test services under the legacy business name (Citation Audit Report.md). Cleaning the GBP answer does not clean the page it points at.
4. Group Retreats placement — /trips vs /learn. Must be decided before Tranche B (Week 7); redirecting twice costs signal.
5. Peak-season dates for the Yakima, which lock the Tranche D timing and every seasonal date in the Content Calendar.

Known follow-on work created by this pass

• Social Media Calendar.md is now out of date and is flagged, not fixed. It was built against the old blog dates. A warning block was added at the head of its blog-carrying schedule stating the corrected dates (#1 Sep 22, #2 Sep 29, #3 Oct 6, #4 Oct 13) and instructing that blog-dependent slots shift forward three weeks for #1–#2 and one week for #3–#4, and that nothing should be scheduled from that document until the shift is applied line-by-line. Roughly 1,850 lines of dated slots; it is mechanical but it is a separate pass.
• QA numbering discrepancy, recorded for the record. B7 above cites "Q13, Q16, Q17, Q18, Q19, Q20 and Q22." The GBP document contains 21 questions, not 22, and the numbering does not line up: women-only programs is Q9 (not Q13 — Q13 is the fishing-licence answer, which is factual and cleared), and pricing is Q6 (there is no Q22). The seven held questions were selected by topic, which is unambiguous, and are Q6, Q9, Q16, Q17, Q18, Q19, Q20. The QA report's line references for §13 also run ~30 lines ahead of the file.
• GBP Optimization Report.md §13 cleared Q3 carries one required edit, flagged inline: the general "we provide the gear" statement is safe and cleared, but no itemised included/not-included list may be added to it until the gear split is confirmed. The itemised version is Q19 and is held.
• S13 (the Q&A upvote-ring instruction) was left in place — it is a Should-fix, not a blocker, and it was outside this pass. It sits three paragraphs above the new §13 warning and should be deleted before ship; coordinating upvotes across accounts is fake engagement under Google's Maps policy, on a profile the same document says is at elevated suspension risk.

Validation performed after editing

Remediation performed August 27, 2026. Eleven of twelve blockers closed; B11 remains open and needs one edit with no client input; B8 is quarantined rather than deleted by design. Five items genuinely require Kristen and are listed above.