- referenceMigration objection-handling map — sourced answers to every common client fear about migrating off WordPress
- referenceCandid migration pricing bands 2026: $3.5-7K brochure / $8-18K 50-page+100-post / $20-45K 200-page content-heavy; ACF/Elementor/WooCommerce multipliers
- referenceLive-preview state of the art (May 2026): Sanity Presentation most mature; Storyblok Visual Editor pioneered iframe pattern; TinaCMS only Git-based with true visual editing
- referenceCMS workflow decision matrix — Sanity/Storyblok (weekly editors, visual preview); Keystatic (monthly, technical-adjacent); Markdown-in-Git (engineers only)
- referenceContent extraction decision tree — WP REST API default, WXR XML fallback, direct DB only for hidden postmeta
- referenceMigration hidden-killers catalogue — the seven failure modes that appear in every post-mortem
- referenceRedirect map deliverable spec — CSV of old→new URL with notes; sources are XML sitemap + Screaming Frog + GSC + Ahrefs/Semrush
- referenceSame-domain CMS migration with proper 1:1 redirects: 2-4 weeks of crawl turbulence; 4-8 weeks to full stability. The 523-day stat is domain-to-domain.
- referenceWhen NOT to migrate a client off WordPress — the honest counter-cases Candid Creative will refuse
- referenceNext.js suffered a 10 percentage-point CWV drop when FID was replaced by INP (March 2024) — "modern stack" is not synonymous with "fast"
- referenceManaged WordPress host TTFB benchmarks (HostingStep 2025, 34 hosts, 1,960+ days of 24/7 monitoring)
- referenceA well-optimized WordPress + Gutenberg + Kadence/GeneratePress site on managed hosting can plausibly match an Astro-on-Cloudflare site in CWV
- referenceWordPress CWV gap causal decomposition (May 2026): ~40-50% hosting/TTFB, ~20-30% page builders, ~15-20% plugins, ~5-10% core, ~5-10% theme
- referenceAI search citation likelihood — direct evidence is thin; indirect evidence (overlap with organic top 10) is moderate. Speed is not a known direct lever.
- reference$5K WordPress refresh vs $20K custom build — performance alone does not justify the $15K delta
- referenceContentsquare 2023 (n=997 customer sites): Good INP retail users converted 2.5% vs 2.0% for Poor (25% relative lift); +45% on desktop
- referenceGoogle neural-net (Think with Google, 2018): mobile load 1→3s = +32% bounce probability; 1→5s = +90%; 1→10s = +123%
- referenceAkamai/SOASTA (2017, ~10B visits): 100ms delay = -2.4% desktop / -7.1% smartphone conversions; 2s delay = -36.5% desktop / -26.2% mobile
- referenceThe Economic Times (2021): LCP 4.5s→2.5s, CLS 0.25→0.09 → -43% bounce site-wide
- referenceRenault (2021): 1s LCP improvement → -14 ppt bounce rate, +13% conversions (regression analysis, multi-month, millions of sessions)
- referenceredBus (2023): INP improved ~72% on search page → +7% sales (6-7.2% CR uplift across operating countries)
- referenceRakuten 24 (2022 A/B test): CWV optimization → +33.13% conversion rate, +53.37% revenue per visitor, -35.12% exit rate
- referenceRAG on a marketing site only pays off above ~50 pages of substantive content AND when visitor questions don't map cleanly to navigation
- referenceAI-assisted dev productivity 2026: published evidence ranges from 21-28% speedup (GovTech) to 19% slowdown (METR RCT) — honest ceiling is 1.3-1.7× for fluent users
- referenceFor sub-100k pageviews/mo SMB sites, Cloudflare Pages + Workers is default; flip only when Postgres connection pooling or existing Vercel contract
- referencescheduler.yield() is Limited Availability — Chrome 129+, Edge, Firefox; Safari has not shipped. Use scheduler-polyfill.
- referenceTailwind v4 (January 2025) — Oxide engine; CSS-first @theme config; ~3.5× faster full builds, ~8× faster incrementals
- referenceSpeculation Rules with eagerness:"moderate" is the 2026 default for instant navigation; ~82% LCP improvement in CoreDash field data
- referenceCross-document View Transitions are Limited Availability — Chrome/Edge 126+, Safari 18.2+, Firefox not yet shipped
- referenceSame-document View Transitions became Baseline Newly Available on October 14, 2025 (Firefox 144)
- referenceReact Compiler 1.0 went stable October 7, 2025; battle-tested on Meta apps and fully production-ready
- referenceNext.js 16 (October 2025) made Partial Prerendering production via "Cache Components" with dynamic-by-default semantics
- referenceAstro 6 went stable in March 2026; removes legacy APIs (getEntryBySlug, getDataEntryById, Astro.glob, Astro.canonicalURL)
- referenceCloudflare acquired The Astro Technology Company on January 16, 2026 — strategic alignment for Candid's default stack
- referenceResearch brief: The Candid Creative WordPress Migration Playbook (piece 19)
- referenceResearch brief: WordPress + Page Builders vs Modern Custom Stacks — sourced performance comparison (piece 18)
- referenceResearch brief: Website Performance & Revenue — defensible evidence for KW small-business owners (piece 17)
- referenceResearch brief: Candid Creative 2026 Build-Standards — web stack decision framework for SMB marketing sites & lightweight apps (piece 16)
- referenceClarification: the SEJ 892-migration study explicitly measures domain-to-domain moves, NOT same-domain CMS swaps
- referenceLow-risk cutover pattern for same-domain CMS migration — 2-week pre-flight, DNS TTL 300s, monitor 48h, keep WP firewalled 30 days
- referenceFeature-parity replacements for common WordPress plugins (forms, SEO, search, comments, commerce, membership, newsletter, analytics)
- referenceBricks Builder produces the cleanest output of the WordPress visual builders; recommend it only when client absolutely cannot work in Gutenberg
- referenceWordPress page-builder market share, 2025 Web Almanac: Elementor 43% (down from 56% in 2024); Gutenberg 18% (up from ~12%); WPBakery 13%; Divi 10%
- referenceWordPress median mobile JS payload is 528 KB — smaller than Wix (1,462 KB) and Squarespace (1,314 KB), yet WordPress performs worse in the field
- referencePlatform mobile CWV pass rates, November 2025 CrUX Tech Report (SEJ analysis, Dec 2025)
- referenceCore Web Vitals is a tiebreaker-class Google ranking factor, not a heavy lever — independent studies (Perficient, AWR, Backlinko) corroborate
- referenceCANDID REFERENCE: how the 15-brief foundation roadmap connects — the throughline from strategic frame to editorial layer
- referenceCANDID REFERENCE: 9-step link-rot mitigation plan — archive on capture, verbatim quote, persistent IDs, quarterly check
- referenceCANDID REFERENCE: retraction/correction playbook — 4 magnitudes from typo to fundamental retraction (adapted from COPE)
- referenceCANDID REFERENCE: "what to source" checklist — Must / Should / Doesn't need
- referenceCANDID REFERENCE: 8-pattern citation library — inline links, hover footnotes, end-of-section, archive pairs, schema
- referenceCANDID REFERENCE: 7-label confidence taxonomy — Verified / Industry-consensus / Single-source / Estimated / Author's view / Contested / Stale
- referenceCOPE Retraction Guidelines (Aug 2025): "The purpose of retraction is to correct the literature and ensure its integrity, not to punish the authors"
- referenceTrust Project / Reach Plc UK: trust in The Mirror jumped 8% after adding Trust Indicators
- referenceGoogle retired ClaimReview rich results in June 2025; schema persists for non-Google AI ingestion
- referenceLink rot: NYT external links 1996-2019 show ~15-year half-life; 13% of "live" links no longer point to original content
- referenceSimhi et al. (Technion/Oxford/Hebrew U, Feb 2025): "models can hallucinate with high certainty even when they have the correct knowledge"
- referenceFTC reasonable-basis doctrine: advertisers must possess "the amount and type of substantiation the ad actually communicates"
- referenceUS Intelligence Community: high/moderate/low confidence taxonomy (ICD 203/206, 2007 NIE Iran convention)
- referenceWikipedia: "The threshold for inclusion is verifiability, not truth" — four claim types always need inline citations
- referenceSPJ Code of Ethics (2014): "Identify sources clearly. The public is entitled to as much information as possible to judge reliability and motivations"
- referenceReuters Handbook of Journalism: "A named source is always preferable to an unnamed source. Anonymous sources are the weakest sources"
- referenceBBC Editorial Guidelines §3.2.2: "All BBC output...must be well sourced, based on sound evidence, thoroughly tested"
- referenceEdelman 2025 Trust Barometer (n=33,000, 28 countries): 7 in 10 believe government/business/journalists deliberately mislead them
- referenceResearch brief: Confidence Levels, Sources, and Dated Claims — why every statement on a credible site should be verifiable (piece 15 of 15)
- referenceReference: research-first tooling stack 2026 — Obsidian + Git + Quartz/Astro + AI overlay (RAG)
- referenceReference: 8 named exemplars of research-first/docs-as-product methodology — Stripe, Twilio, Anthropic, Gwern, Matuschak, Appleton, BAM, Construction Physics
- referenceReference: research-first cadence by operation type — 1 piece/month to bimonthly print
- referenceReference framework: audience layering — foundation research vs public article vs marketing page (3-column comparison)
- referenceReference framework: Research-first workflow — 5 stages (Capture → Foundation → Synthesis → Article → Marketing page → Maintain)
- referenceDevon Zuegel: epistemic-status labels are "a hack in order to publish half-baked ideas I'd otherwise not feel comfortable sharing"
- referenceHubSpot 2015 (n=13,500+): companies posting 16+ blog posts/month got "almost 3.5× more inbound traffic" — dated but still cited
- referenceCasey Newton / Platformer retreat from daily cadence (April 2026): "More scoops, less aggregation and analysis" — AI commoditizes daily synthesis
- referenceResearch brief: Research Before Pages — methodology for KB-backed websites (piece 14 of 15)
- referenceIllustration: what "modeling the actual business" means for Boucher & Jones fuel distribution (in-development, NOT delivered)
- referenceReference: operational data map by industry — what gets generated, stranded, and unlockable for 5 service verticals
- referenceReference: the 5-stage Data Maturity Curve — from stranded data to data as product
- referenceReference: minimum viable data stack for a $1M-$10M Canadian service business (2026, C$100-C$500/month)
- referenceClio Legal Trends: AI adoption among legal professionals jumped 19% (2023) → 79% (2024) → 93% mid-sized firms (2025); 64% mid-sized offer flat-fee
- referenceCFIB / Sage / Payworks 2025: only 10% of Canadian SMBs have digital tools fully integrated; $1.60 ROI per dollar invested
- referenceQuebec Law 25 (fully in force Sept 22, 2024): data portability + fines up to C$25M / 4% of worldwide turnover
- referenceOPC vs Brinks Home (PIPEDA Findings #2024-002, Mar 28 2024): inadequate safeguards left customer data accessible for 10 weeks
- referenceCanadian privacy 2026: PIPEDA still governs; Bill C-27 died on the Order Paper Jan 6, 2025 — no fines, only findings
- referenceTunguz: vertical SaaS median sales efficiency 0.97 vs horizontal 0.66 (n=54, 2015 — flag age)
- referenceReference: vertical SaaS data portability comparison (ServiceTitan / Jobber / Housecall Pro / Clio / Karbon / Tekmetric)
- referenceServiceTitan: "Open Data Pledge" promises CSV export — but practitioner reports cite $24k-$39k exit contract buyouts (flag for verification)
- referenceServiceTitan IPO Dec 12, 2024: $8.9B closing cap, $685M revenue, $62B GTV, >95% gross retention, >110% NDR
- referenceMuleSoft 2025 Connectivity Benchmark: organizations use 897 applications on average; only 29% are integrated
- referenceCRM project failure rate: 47% (Forrester) to 50% (Gartner) to 55% (Johnny Grow 2025) to 70% (industry aggregators)
- referenceResearch brief: The Dataset is the Product — when a service business should own its data (piece 12 of 15)
- referenceReference: underexploited Canadian open data by industry — highest-leverage starts for KW SMB clients
- referenceReference: compliance-grade attribution checklist by open-data source
- referenceProfound (Aug 2024-Jun 2025, 680M citations): only 11% domain overlap between ChatGPT and Perplexity; 13.7% between AI Overviews and AI Mode
- referenceReference: open-data ingestion stack for a 1-3 person SMB operation (2026) — under $50/mo realistic
- referenceMotherDuck pricing 2026: Lite ($25/mo) removed; Business moved to $250/mo between Dec 2025 and Feb 2026
- referenceCarfax: from 10,000 records faxed in 1986 to 35B+ records across 151,000+ sources — sold to S&P Global Mobility 2022
- referenceHelloSafe: "Canada's leading insurance/financial comparison platform" — built on StatCan + OSFI data
- referenceLocal Logic (Montreal): 100B+ data points — "largest location dataset in real estate"
- referenceThe Climate Corporation (Bayer Crop Science since 2018): field-level overlay on NOAA + NWS + USGS + NRCS + NASA
- referenceFlightAware: FAA + 45-country ANSP feeds + 30,000+ user-hosted ADS-B receivers + Aireon — the crowdsourced moat
- referenceCherre: property knowledge graph powering management of $3.3T AUM globally
- referenceATTOM Data: 500M+ real estate/loan transactions, 2,690+ counties, 20-step Enterprise Data Management Program
- referenceZillow: 110M-home "living database" built on Census/ACS + 3,000 county assessors + USPS + MLS feeds
- referenceMeta v Bright Data (Jan 2024, N.D. Cal.): Facebook/Instagram terms don't bar logged-off scraping of public data
- referencehiQ v LinkedIn (9th Cir. Apr 2022): scraping publicly accessible data likely doesn't violate CFAA — but hiQ still settled for $500K
- referenceInternational open-data licences (2026): UK OGL v3, OECD CC BY 4.0, Eurostat, World Bank
- referenceNOAA/NWS: information on NWS web servers is in the public domain — no attribution required, provided "as is"
- referenceU.S. EIA APIv2: free registered-key access to petroleum/electricity/gas/coal/STEO/AEO data; WPSR releases 10:30 AM ET Wednesdays
- referenceOntario Open Data Catalogue: 2,948 datasets under OGL-Ontario v1.0
- referenceCanada Energy Regulator: pipeline throughput/capacity/tolls + Market Snapshots, all under OGL-Canada
- referenceECCC/MSC Open Data: free anonymous access to weather/climate/water via OGC-compliant GeoMet APIs
- referenceStatistics Canada Open Licence: explicitly permits "use, reproduce, publish, freely distribute, or sell value-added products"
- referenceOGL-Canada v2.0: worldwide royalty-free perpetual licence for commercial use
- referenceResearch brief: Public data as a private moat — building proprietary intelligence from government open data (piece 11 of 15)
- referenceCandid Creative page-builder transition roadmap: 4 stages over 12 months
- referenceReference: alternative-stack recommendations by use case and budget (Candid 6-tier framework)
- referenceReference: what page builders cost a small-business site — 10 categories ranked by long-term impact
- referenceAstro on Cloudflare Pages: ~$0/month vs Next.js SSR ~$20-200/month (Bobes 2026 benchmark)
- referencePronto Marketing: Elementor FAQ Accordion + Element Caching breaks FAQPage JSON-LD ("Missing field text")
- referenceElementor does not generate schema markup by default — Schema Pilot: "None of the layouts or widgets add structured data"
- referenceDOJ Interim Final Rule 2026-07663: ADA Title II compliance dates extended to Apr 26 2027 / Apr 26 2028
- referenceElementor GitHub Issue #11779 (open): "Elementor is not fully compliant out of the box and breaks many accessibility rules"
- referenceElementor 3.26 (late 2024): removed legacy Schemes API + enabled Element Caching by default — risky for shortcode-dependent themes
- referenceElementor 3.24.1 (August 2024): widespread reports of broken sites + entire designs lost in update
- referenceBricks CVE-2024-25600: unauthenticated RCE (CVSS 10) — exploited in the wild ~24 hours after patch release
- referenceBricks Builder retired $199 unlimited-sites lifetime license in January 2024
- referenceElementor Pro Nov 2023: removed ACF/Pods integration from Essential plan for new customers
- referenceGeneratePress official: "You cannot convert Elementor's code to the code required by the Block Editor"
- referenceWPPoland: Elementor → Gutenberg rebuilds yield ~40% mobile site speed increase (multi-client benchmark)
- referenceAnubiz Host: Elementor adds 500KB+ of CSS/JS to every page; simple pages can have 2000+ DOM elements
- referenceElementor own admission: Flexbox Containers ship 40% less HTML output than sections/columns
- referenceWeb Almanac 2024: WordPress mobile CWV pass rate jumped 28% (2023) → 40% (2024); median mobile Lighthouse Perf 38
- referenceW3Techs (April 2026): Elementor on 18.6% of WordPress sites, 13.2% of all websites
- referenceResearch brief: The Case Against Page Builders (piece 10 of 15)
- referenceGoogle WebP Compression Study: WebP is 25-34% smaller than JPEG at equivalent SSIM (avg ~30%)
- referenceReference: 2026 performance budget — SMB marketing site (Candid default targets)
- referenceReference: the 5 most common CLS killers — and the fixes
- referenceReference: the 8 most common INP killers — ranked by observed frequency
- referenceReference: the 8 most common LCP killers — ranked by observed frequency
- referenceDan Taylor (SE Land, Jan 13 2026, n=107,352): CWV is a gate for AI citation, not a growth lever
- referenceChrome Aurora team: Google Tag Manager with 18 tags increases Total Blocking Time ~20×
- referenceAstro vs Next.js (eastondev benchmark, Dec 2025): ~40% faster, ~90% less JS for static content
- referenceBBC: lost 10% of users for every additional second the site took to load
- referenceWalmart (2012, foundational): 100ms improvement = 1% incremental revenue; 1s improvement = 2% conversion increase
- referencePinterest PWA rebuild (2017, foundational): 60% faster → +40% conversions, +40% time spent, +44% ad revenue
- referenceDeloitte/Google "Milliseconds Make Millions" (2020): 0.1s improvement → 8.4% retail conversion lift, 10.1% travel
- referenceVodafone A/B test (2021): 31% LCP improvement → 8% more sales, 15% better lead-rate, 11% better cart-rate
- referenceWeb Almanac 2025: median mobile page weight crossed 2,559 KB in July 2025 (+7.8% YoY)
- referenceWeb Almanac 2025: 48% mobile / 56% desktop origins pass all 3 CWV; LCP is the bottleneck
- referenceREFUTED: "March 2026 LCP tightened to 2.0s / INP elevated" — no such Google announcement exists
- reference2026 CWV thresholds (confirmed unchanged): LCP ≤ 2.5s, INP ≤ 200ms, CLS ≤ 0.1 at p75 field data
- referenceResearch brief: Page Speed as a Moat — why CWV separates the agencies from the freelancers (piece 9 of 15)
- referenceHubSpot (Pamela Vaughan): historical-optimization refresh of old posts lifts organic traffic by avg 106%
- referenceAhrefs (14B-page study): 96.55% of pages get zero Google traffic — supersedes older 90.63% figure
- referenceReference framework: 10-year cost model — rebuild-every-3-years vs foundation-first (Canadian SMB, CAD)
- referenceReference framework: which website dimensions decay vs compound over 10 years (12-dimension matrix)
- referenceGDS: GOV.UK Design System pages download 2x faster than non-system pages with half the code
- referenceLong-lived sites worth studying: Daring Fireball, Berkshire Hathaway, Craigslist, Pinboard, GOV.UK, Stack Overflow, Wikipedia
- referenceDrupal 7 EOL January 5, 2025 — ~291,386 sites still on D7 in Sept 2024; migration is effectively a rebuild
- referenceDivi 5 official release Feb 26, 2026; Divi 4 → 5 is one-way migration; rollback gets harder over time
- referenceEuropean Accessibility Act enforcement began June 28, 2025 — penalties up to €100k (Germany), 4-5% revenue (France/Italy)
- referenceFTC (Jan 3, 2025): accessiBe ordered to pay $1M for deceptive claims its AI overlay could make sites WCAG-compliant
- referenceAccessibility: 95% of sites fail basic WCAG; 40% of new federal ADA filings are pro se (AI-assisted)
- referenceUsableNet 2025: 5,000+ digital accessibility lawsuits filed; 46% of federal cases involve repeat defendants
- referenceZittrain et al. (Harvard Law 2014): 50% of URLs in U.S. Supreme Court opinions suffer reference rot
- referenceWeb Almanac 2024: median desktop page weight — WordPress 2,252 KB, Wix 2,560, Squarespace 3,323; 90th pct crosses 8 MB
- referenceVerizon DBIR 2025: 88% of SMB breaches involved ransomware vs 39% of enterprise; median ransom $115k
- referenceSucuri 2023: 39.1% of CMS apps outdated at point of infection (down from 50.58% in 2022)
- referenceWordfence 2024: 54 billion malicious requests blocked, ~325-350k sites infected on any given day
- referencePatchstack 2024: 1,614 plugins and themes removed from .org repo for unpatched security issues
- referencePatchstack 2024: 4,166 new vulnerabilities, 96% in plugins, 4% in themes, only 7 in core
- referenceOrbit Media: average website lifespan across Inc 5000 is 2y4mo; under continuous care, 6y4mo
- referenceResearch brief: Built to Last — why most SMB sites rebuild every 3-4 years (piece 5 of 15)
- referenceBoucher & Jones IA recommendation: Pattern A + hub-and-spoke, NOT enterprise matrix
- referenceReference: 5 URL structure patterns for multi-vertical IA (A through E), with field-observed usage
- referenceReference: which IA pattern to use, by business shape (decision matrix)
- referenceNN/g: split buttons unreliable on touch (fat-finger conflicts) — use sequential menus or accordions on mobile
- referenceField survey (May 2026): the /for-<audience>/ URL pattern was observed on ZERO of 8 multi-vertical service sites
- referenceHUB International: parallel /industries/ + /products/ with industry sub-pages where complexity warrants
- referenceFastenal: NO /industries/ IA at all — product+service-first nav. The steel-manned alternative.
- referenceDentons: explicit Industries × Practices matrix at /find-your-dentons-team/ — "organized around your business agenda"
- referenceBDO USA: mirrors Crowe — /industries/<slug>/ pattern with separate /services/ namespace
- referenceCrowe LLP: cleanest /industries/<slug>/ URL pattern in the sample (Audit/Tax/Advisory/Consulting × industries)
- referenceE.H. Wolf & Sons: closest direct B&J analog — "Industries We Serve" + 16 flat-root vertical pages
- referenceRelaDyne: /products-solutions/<industry>/ namespace, ~19 verticals, multi-axis IA ("By Industry / Product / Brand / Service")
- referenceAubrey Yung: "audience captures relevance; areaServed captures coverage"
- referenceSchema.org canonical pattern: one Service entity with audience: { type: Audience, name: ... } + areaServed
- referenceSchema.org: Service.audience accepts Audience | PeopleAudience | BusinessAudience with audienceType as free-text
- referenceZuko: multi-step forms only outperform single-page above ~7 fields; below that, neutral or worse
- referenceNuminam (2024): 11-field B2B SaaS form 0.96% → 8.1% via multi-step (+743% lift)
- referenceVenture Harbour: multi-step forms up to 300% better than single-page — correctly attributed (NOT CXL/ConversionXL)
- referenceFormstack 2014: multi-step forms 13.9% vs 4.5% single-step (n=450k+ accounts) — directional, not RCT
- referenceGartner: 77% of B2B buyers say their latest purchase was "very complex or difficult" (CSO Update 2019)
- referenceGartner: typical B2B buying group is 6-10 decision-makers; ~95% revisit earlier decisions (CSO Update, 2019)
- referenceGartner: real B2B buying journey looks "like a big bowl of spaghetti" (pull quote)
- referenceGartner: B2B buyers complete "six buying jobs" in non-linear "looping" order
- referenceJohn Mueller on internal linking: "one of the biggest things that you can do on a website"
- referenceGoogle Search Central: faceted navigation creates combinatorial duplicate URLs; not search-friendly by default
- referenceMueller on near-duplicate vertical/location pages: fold them into one stronger page unless each has something unique
- referenceJohn Mueller: Google does NOT penalize duplicate content for localized websites — it consolidates
- referenceInformation Architecture: For the Web and Beyond, 4th ed. (Rosenfeld/Morville/Arango, O'Reilly 2015) — the canonical IA reference
- referenceNN/g: use labels with strong information scent; avoid vague verbs and conversational tone in nav (Apr 16 2023)
- referenceNN/g: "On large screens, don't cover the entire screen when megamenus are open" (Apr 30 2023)
- referenceNN/g mega-menu hover protocol: wait 0.5s before showing, then display within 0.1s
- referenceJakob Nielsen: "Mega menus show everything at a glance, so users can see rather than try to remember"
- referenceNN/g: five recurring problems with role-based navigation (Laubheimer, 2022)
- referenceNN/g: audience-based navigation "will often degrade usability" (Laubheimer, Aug 14 2022)
- referenceResearch brief: Information architecture for service businesses with multiple verticals (piece 6 of 15)
- referenceCounter-argument: a $200/mo managed platform buys you a larger security team than most agencies have
- referenceOwnership Checklist: what an SMB must be able to walk away with at agency separation
- referenceCounter-example: well-planned migrations can produce 3-5× traffic gains (BrightEdge / Numen case studies)
- referenceCoalition Tech "Client ABC" BigCommerce migration: 28,000 → 13,400 ranking keywords in 10 months
- referenceUK retailer lost ~£3.8M in first month after botched migration — IT consultants rejected redirect mapping
- reference"9 of 10 web migrations fail" — single-sourced (Numen Technology); use 523-day SEJ stat instead
- referenceCanadian Bill C-15 (tabled Nov 4, 2025) — proposes data-mobility framework for federal PIPEDA
- referenceQuebec Law 25: data portability effective Sept 22, 2024; penalties up to C$25M / 4% of worldwide turnover
- referenceEU Data Act (Regulation 2023/2854): SaaS switching procedures effective Sept 2025; switching fees abolished by Sept 2027
- referenceGDPR Article 20: portability is narrow — only user-provided data, only under consent/contract, excludes derived data
- referenceGoogle Workspace: cannot transfer file ownership to external accounts; client separation requires IMAP+DNS migration
- referenceACF custom fields don't survive WordPress's native XML export — image IDs + serialized arrays break
- referenceElementor: no built-in "deactivate but retain content" option — open feature request since 2018
- referenceDivi 4 stored content as proprietary [et_pb_*] shortcodes — orphan text on theme deactivation (Divi 5 fixes this)
- referenceAgency-as-Registrant "domain hostage" pattern — practitioner-documented but rarely litigated
- referenceICANN: listed Registrant is the legal owner of a domain — admin/technical contact is NOT ownership
- referenceSquarespace acquired Google Domains for $180M (Sept 7, 2023) — ~10M SMB domains migrated automatically
- referenceShopify checkout customization rebuild: $2k–$10k per shop, 4–8 weeks for complex Scripts
- referenceShopify Plus: ~47,130 distinct merchants on 70,696 domains (Store Leads 2026); ~31% of Shopify's MRR
- referenceShopify checkout.liquid sunset (Aug 2024 → Aug 2025 → June 30 2026) — unmigrated customizations will be DELETED
- referenceSquarespace 7.1: no XML export at all. 7.0 export omits images, products, custom CSS, video, audio, drafts, non-blog pages
- referenceWix officially: "no export is possible" — confirmed in writing as a structural property
- referenceWebflow code export: CMS / accounts / e-commerce / localized content NOT included; forms and search don't work
- referenceBlackRock marks down Automattic shares 67.4% to $27.74 (June 30, 2025) — financial signal of dispute cost
- referenceACF → SCF forced fork (Oct 12, 2024) — first unilateral plugin takeover in 21-year WordPress history
- referenceWP Engine vs Automattic timeline (Sept 2024 → ongoing) — open-source has centralized choke points
- referenceIndieWeb POSSE: "Publish on Own Site, Syndicate Elsewhere" (Tantek Çelik, June 2012)
- referenceSolid project (Tim Berners-Lee) — data-pod architecture; ODI took stewardship Oct 2024
- referenceDoctorow's prescription: "right of exit" requires interoperability, not just open source
- referenceDoctorow: "enshittification" — the three-phase decay pattern of platforms (Word of the Year 2023 + 2024)
- referenceResearch brief: Owning your stack — why agency-managed platforms cost more than they save (piece 4 of 15)
- referencePatchstack 2026: 91% of new WordPress vulnerabilities are in plugins; only 6 CVEs in core
- referenceAstro Content Collections (stable since 2.0, modernized in Content Layer API late 2024): Zod-validated frontmatter, build-time TypeScript types
- referenceOpenAlex (271M works) and Semantic Scholar (214M+ works) — open scholarly citation graphs at scale
- referenceOur World in Data: CC-BY licensing, per-indicator JSON/CSV endpoints, full GitHub provenance
- referenceStanford Encyclopedia of Philosophy: every entry has permanent dated archived editions
- referenceOtterlyAI: /llms.txt sees 0.1% of AI-bot traffic — performs worse than average content page
- referenceWikipedia is the #1 cited domain in Google AI Mode (11.22%); YouTube #2 at 9.51%
- referenceWikipedia verifiability policy: all challenged material must carry an inline citation to a reliable published source
- referenceStripe docs as a first-class product — Markdoc framework, documentation in performance reviews
- referenceAndy Matuschak: evergreen notes — atomic, concept-oriented, densely linked, accreting over time
- referenceDiátaxis (Daniele Procida): four documentation types — tutorials, how-to, reference, explanation
- referenceINP officially replaced FID as a Core Web Vital on March 12, 2024
- referenceMediafly Feb 2022: 13 min vs 8.5 min engagement on interactive vs static — but the measurement was B2B sales decks, not marketing websites
- referenceElementor sites show pre-optimization median mobile LCP of 3.8–5.2s — well above Google's 2.5s threshold
- referenceWordPress market share (May 2026): 43.2% of web, 60.4% of CMS — peaked at 65.2% in 2022
- referenceExtractability: a quotable paragraph leads with the answer, is 40-60 words, lives under semantic HTML, and names entities concretely
- referenceAI crawlers (GPTBot, ClaudeBot, PerplexityBot) generally do not execute JavaScript — client-side React/Vue without SSR is invisible
- referenceOtterlyAI (Sept 2025): community platforms (Reddit, Quora) capture 52.5% of AI citations vs 47.5% brand domains
- referenceWebMCP + UCP: schema = nouns, WebMCP = verbs, UCP = wallet (Chrome 146 / Google I/O 2026)
- referenceDigital Applied 5K-site audit (Apr 2026): only 22% of schema implementations pass Google Rich Results Test (r=+0.34 with AI citation)
- referenceJSON-LD adoption: 41% of pages (Web Almanac 2024); 62M domains, +37% YoY (Schema App / W3Techs)
- referenceSuganthan: schema has three lives — index-time, training-time, query-time
- referenceQuery fan-out: Google AI Overviews issue multiple sub-queries; pages get cited across queries they never targeted
- referenceAI citation overlap with Google top-10 has collapsed — from 76% (Jul 2025) to ~8-38% (2026)
- referenceSchanbacher (SSRN 2025): FAQPage and Product schema strongly predict ChatGPT visibility (single-domain peer-reviewed)
- referenceBrightEdge (Feb 2026): AI Overviews now appear on 48% of tracked queries, up from ~30% a year prior
- referenceGoogle (May 2026): "Structured data isn't required for generative AI search"
- referenceGoogle Search Central (May 15, 2026): "Optimizing for generative AI is still SEO"
- referenceWhitespark 2026: AI Search Visibility added as a formal local ranking category for the first time
- referenceAhrefs (2025, 17M citations): AI-cited content is 25.7% fresher than traditional organic results
- referenceSeer (Sept 2025): brands cited inside AI Overviews earn 35% more organic CTR and 91% more paid CTR
- referenceSeer Interactive (Oct 2025): 65% of AI bot hits target content under 1 year old; 89% under 3 years
- referenceAhrefs (April 2026): adding schema to 1,885 pages produced no AI-citation lift; AI Overviews showed -4.6%
- referenceGEO finding: Lower-ranked pages (organic rank ~5) gain 115.1% AI visibility when given GEO treatments
- referenceGEO finding: Keyword stuffing is the only tested tactic that hurts AI visibility (-8% to -10%)
- referenceGEO finding: Cite Sources lifts visibility +28%
- referenceGEO finding: Statistics Addition is the #2 tactic at +31%
- referenceGEO finding: Quotation Addition is the top-performing tactic at +41% on Position-Adjusted Word Count
- referencePrinceton GEO paper (Aggarwal et al., KDD '24) — the foundational generative engine optimization study
- referenceResearch brief: The knowledge-base-backed website (piece 3 of 15)
- referenceResearch brief: What makes a marketing site do something (piece on brochure vs platform)
- referenceResearch brief: Structured content as a competitive advantage (piece 2 of 15)
