# The answer isn’t the experience. Sparkitecture turns a roofing website into a cinematic, useful experience. Follow Vito through the evidence, change a homeowner outcome and take your blueprint to RoofClaw. Useful destination: Understand Sparkitecture, choose a homeowner outcome, and take a practical experience blueprint into a RoofClaw demo Value promise: Your visitor outcome, cinematic experience plan and next move for a RoofClaw demo. Subject or offer: Explore a RoofClaw configured with the Sparkitecture harness to research, author and build useful cinematic interactive pages. The demo determines business fit and delivery scope. ## A working explanation ### Attention:Mystery — What is left after the answer? A homeowner can ask an answer engine what causes a roof leak. What can your website help them do next? Question: If the answer arrives before the visit, what makes the visit worth taking? Discovery: The missing next step is applying the information to a situation and preparing a useful conversation. Mechanism: Open the clue to distinguish an explanation from a preparation brief. Visitor action: Open the clue Consequence: Identify a task the generic answer does not complete. Portable contribution: Record a useful homeowner task. Next: Follow the evidence behind the question. - Show me the missing next step: An explanation says what leaks can mean. A useful page helps a homeowner organize observations and questions for a qualified inspection. The visit creates something they can use. - Brief contribution: Help a homeowner prepare observations and questions for a qualified inspection. ### Attention:Surprise — A beautiful surface still needs a purpose. Katie Dill's warning about zombie UI asks designers to bring judgment and a point of view. A finished-looking page is the start of the decision. The cinematic surprise here is the useful constraint: motion and polish need a task, a visible consequence and a result worth keeping. Question: What should your cinematic page accomplish? Discovery: A striking scene needs reviewed content, a consequential action and a complete accessible route to become a useful experience. Mechanism: Reveal the work behind the visual and carry the useful requirement into the plan. Visitor action: Reveal the requirement Consequence: Choose the visitor outcome before adding more scenes. Portable contribution: Record the task, meaningful choice and accessible result. Next: Follow the evidence behind the question. - Give the cinema a job: Use a scene to frame the question or reveal a relationship. Let native choices change the result. Plan content review, playback, performance and reduced motion alongside production. - Brief contribution: Every cinematic scene frames a useful question; native choices change the result; review content, playback, performance and access. ### Interest:Education — Answers can arrive before visits. In Pew's March 2025 U.S. browsing panel, people clicked traditional results on 8% of Google visits with an AI summary, versus 15% without. Summary-source links were clicked on 1% of summary visits. Google reports stable aggregate organic clicks. Query-level click declines and aggregate stability can coexist. Question: What has changed in how people click? Discovery: The studies describe particular searches, samples and periods. They inform a design hypothesis to test with your own visitors. Mechanism: Inspect primary findings alongside their methods, scope and the Google counterpoint. Visitor action: Inspect a source finding Consequence: Keep the source scope attached to the finding. Portable contribution: Carry evidence and its limits into the blueprint. Next: Look inside the Sparkitecture method. - Pew Research Center: In Pew's March 2025 U.S. browsing panel, people clicked traditional results on 8% of Google visits with an AI summary, versus 15% without. Summary-source links were clicked on 1% of summary visits. Observational association, one month, Google only. Different queries and recreated results can affect the comparison. This is not a measured traffic loss for roofing websites. - Depends on: Starting information - Brief contribution: In Pew's March 2025 U.S. browsing panel, people clicked traditional results on 8% of Google visits with an AI summary, versus 15% without. Summary-source links were clicked on 1% of summary visits. Source: https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ Limits: Observational association, one month, Google only. Different queries and recreated results can affect the comparison. This is not a measured traffic loss for roofing websites. - Ahrefs: Ahrefs found AI Overview presence correlated with a 58% lower position-one desktop clickthrough rate relative to its modeled counterfactual. Selected keyword sample, desktop only, observational comparison. The 58% is relative to a modeled counterfactual, not 58 percentage points or all website traffic. No roofing conversion outcome was studied. - Depends on: Starting information - Brief contribution: Ahrefs found AI Overview presence correlated with a 58% lower position-one desktop clickthrough rate relative to its modeled counterfactual. Source: https://ahrefs.com/blog/ai-overviews-reduce-clicks-update/ Limits: Selected keyword sample, desktop only, observational comparison. The 58% is relative to a modeled counterfactual, not 58 percentage points or all website traffic. No roofing conversion outcome was studied. - SparkToro / Similarweb: SparkToro's Similarweb-based estimate found 68.01% of U.S. Google browser searches ended without a click in January–April 2026. Excludes Google's mobile search app. Historical comparisons use different data providers and panels. Zero-click can mean a useful answer, reformulation or abandonment; it does not mean a lost customer. - Depends on: Starting information - Brief contribution: SparkToro's Similarweb-based estimate found 68.01% of U.S. Google browser searches ended without a click in January–April 2026. Source: https://sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/ Limits: Excludes Google's mobile search app. Historical comparisons use different data providers and panels. Zero-click can mean a useful answer, reformulation or abandonment; it does not mean a lost customer. - Cloudflare: Google referrals to Cloudflare's news-customer cohort fell approximately 9% in March 2025 compared with January; April was approximately 15% lower. News publishers are not all websites or roofing businesses. AI launches coincided with the changes but this does not isolate causality. Crawl requests and human referred visits measure different things. - Depends on: Starting information - Brief contribution: Google referrals to Cloudflare's news-customer cohort fell approximately 9% in March 2025 compared with January; April was approximately 15% lower. Source: https://blog.cloudflare.com/crawlers-click-ai-bots-training/ Limits: News publishers are not all websites or roofing businesses. AI launches coincided with the changes but this does not isolate causality. Crawl requests and human referred visits measure different things. - Wikimedia Foundation: Wikipedia reported human pageviews down roughly 8% year over year in recent months of 2025, after updating its bot detection. It linked the trend to changes involving AI answers and social media. Classification rules changed, so Wikimedia advises care in interpretation. The explanation is an attribution, not a controlled causal estimate. This is a major informational publisher, not a roofing-company traffic study. - Depends on: Starting information - Brief contribution: Wikipedia reported human pageviews down roughly 8% year over year in recent months of 2025, after updating its bot detection. It linked the trend to changes involving AI answers and social media. Source: https://wikimediafoundation.org/news/2025/10/17/new-user-trends-on-wikipedia/ Limits: Classification rules changed, so Wikimedia advises care in interpretation. The explanation is an attribution, not a controlled causal estimate. This is a major informational publisher, not a roofing-company traffic study. - Google: Google said total organic clicks were relatively stable year over year and click quality had increased, while traffic shifted between sites. It defined quality as users not quickly clicking back. An interested platform's aggregate statement. Per-query CTR can fall while total clicks stay stable as query volume grows. A quality-click proxy does not establish leads, sales or learning. - Depends on: Starting information - Brief contribution: Google said total organic clicks were relatively stable year over year and click quality had increased, while traffic shifted between sites. It defined quality as users not quickly clicking back. Source: https://blog.google/products-and-platforms/products/search/ai-search-driving-more-queries-higher-quality-clicks/ Limits: An interested platform's aggregate statement. Per-query CTR can fall while total clicks stay stable as query volume grows. A quality-click proxy does not establish leads, sales or learning. - Google Search Central: Google recommends existing SEO fundamentals for AI features: crawl access, helpful native text, good page experience and structured data matching visible content. Supporting links need indexing and snippet eligibility. Applies to Google's features. Eligibility does not guarantee crawling, indexing, ranking, AI citation or traffic. Google requires no special AI file or schema. - Depends on: Starting information - Brief contribution: Google recommends existing SEO fundamentals for AI features: crawl access, helpful native text, good page experience and structured data matching visible content. Supporting links need indexing and snippet eligibility. Source: https://developers.google.com/search/docs/appearance/ai-features Limits: Applies to Google's features. Eligibility does not guarantee crawling, indexing, ranking, AI citation or traffic. Google requires no special AI file or schema. - Nielsen Norman Group: Usability research describes calculators and quizzes as exploratory decision aids. Inputs should meaningfully change outputs, and forcing contact details before revealing results can frustrate users. Supports useful interaction design, not a promise that every quiz increases retention or sales. Experienced users can reject generic outputs; decision support is one input to judgment. - Depends on: Starting information - Brief contribution: Usability research describes calculators and quizzes as exploratory decision aids. Inputs should meaningfully change outputs, and forcing contact details before revealing results can frustrate users. Source: https://www.nngroup.com/articles/calculator-expectations/ Limits: Supports useful interaction design, not a promise that every quiz increases retention or sales. Experienced users can reject generic outputs; decision support is one input to judgment. ### Interest:Education — Spark. Value. Education. Realm. Spark invites participation. Value gives the visitor a useful result. Education makes a mechanism understandable. Realm gives the journey its distinctive sense of place. Attention opens a mystery. Interest lets you investigate. Decision changes a choice. Action carries the result onward. Question: What gives a cinematic page substance? Discovery: The environment and the interaction serve the same visitor outcome. Mechanism: Inspect each pillar and connect its design job to the portable result. Visitor action: Open a pillar Consequence: Connect presentation, participation, understanding and takeaway. Portable contribution: Keep the four design jobs in the experience brief. Next: Try those jobs on a homeowner choice. - Spark: Invite a real action: choose a situation, uncover a relationship or test a decision. - Depends on: Starting information - Brief contribution: Spark: invite an action with a visible consequence. - Value: Name the useful result early and carry the visitor's choices into it. - Depends on: spark - Brief contribution: Value: provide a task-specific result the visitor can retain. - Education: Make the explanation and evidence inspectable so the visitor can apply the principle. - Depends on: value - Brief contribution: Education: explain the mechanism and preserve its evidence. - Realm: Use cinema, character and rhythm to frame the discoveries, with an equivalent readable route. - Depends on: education - Brief contribution: Realm: give the journey a distinct cinematic place while preserving access. ### Interest:Delight — An answer explains. An experience helps you act. Choose a homeowner situation. Watch the preparation task change, then carry that choice into your page blueprint. Illustrative preparation scenarios for a qualified inspection. Question: Which homeowner task should your page help with first? Discovery: Different situations produce different observations and inspection questions. Mechanism: Choose a situation to build its applicable preparation task. Visitor action: Select a homeowner situation Consequence: See a different task and next step for the selected situation. Portable contribution: Record the chosen situation and useful preparation output. Next: See how RoofClaw can help assemble the experience. - A leak appeared: Gather when it appeared, affected interior areas and photos taken safely from the ground or indoors. - Tradeoff: A qualified roofer assesses the property; the page prepares the conversation. - Brief contribution: Leak preparation: observations, timing, safe photos and inspection questions. - After a storm: Record storm timing and visible changes from the ground; prepare questions for a qualified inspection. - Tradeoff: Visible observations alone do not establish damage or an insurance decision. - Brief contribution: Storm preparation: timeline, ground-level observations and inspection questions. - Planning a replacement: Record goals, timing, materials to discuss and questions for comparing written scopes. - Tradeoff: A real assessment determines suitability, price and warranty details. - Brief contribution: Replacement preparation: goals, timing and questions for comparing written scopes. ### Interest:Education — Build the journey behind the cinema. The Sparkitecture harness starts with your audience, source material and useful outcome. It turns them into a reviewed creative plan, native choices, scenes and a portable result. Vito's machine illustrates the process; your business supplies the facts and final scope. Question: What must be prepared before the scenes can serve the visitor? Discovery: Creative direction, source review and meaningful interactions belong in the same plan. Mechanism: Inspect the brief-to-result sequence before discussing implementation at a demo. Visitor action: Inspect a production step Consequence: Identify the resources and decisions each step needs. Portable contribution: Record required content, review and visitor output. Next: Follow an original published experience. - Brief and research: Name the visitor task, gather approved resources and scope the factual claims. - Depends on: Starting information - Brief contribution: Prepare audience, visitor task, approved resources and claim support. - Creative direction: Map an answerable opening, investigation, decision and action. Give every scene a purpose. - Depends on: brief - Brief contribution: Map the cinematic AIDA journey and each scene's useful consequence. - Choices and result: Let the visitor change an explanation or output. Preserve the same task with native controls and text. - Depends on: direction - Brief contribution: Build native choices and a portable result alongside the optional films. - Review and evaluate: Inspect claims, access, media playback and the saved artifact. Test the intended task with real visitors. - Depends on: experience - Brief contribution: Review source scope, playback, access and artifacts; evaluate actual visitor outcomes. ### Interest:Education — The page became the story. A technical comparison became a fight-night experience. Tom's Hardware covered the story and linked to it on October 1, 2026. The creator reports that happened within two days. That is evidence of earned editorial attention; retention and conversion still need analytics. Question: How can a technical comparison become a memorable experience? Discovery: The boxing-round framing makes a complex comparison explorable. Dated coverage and direct links establish a specific published example. Mechanism: Follow the original page and the editorial coverage trail. Visitor action: Open the coverage trail Consequence: Retain the original sources and creator-reported timing. Portable contribution: Save the example and its measurement requirements. Next: Choose what the homeowner should accomplish. - Open the original and the coverage: Original experience: https://ccdocs.com/deepseek-lost/ . Coverage: https://www.tomshardware.com/tech-industry/artificial-intelligence/firm-rents-four-nvidia-h200s-to-test-80x-cheaper-deepseek-claim-usd13-200-monthly-gpu-rental-doubles-claude-bill-while-security-flaws-keep-code-offline . Published October 1, 2026. Within two days, according to the creator; exact launch time not independently verified. - Brief contribution: CCdocs received Tom's Hardware coverage. Original: https://ccdocs.com/deepseek-lost/ . Coverage: https://www.tomshardware.com/tech-industry/artificial-intelligence/firm-rents-four-nvidia-h200s-to-test-80x-cheaper-deepseek-claim-usd13-200-monthly-gpu-rental-doubles-claude-bill-while-security-flaws-keep-code-offline . The two-day interval is creator-reported; retention and conversion need their own analytics. ### Decision:Surprise — Choose what your visitor should accomplish. Pick the outcome first. The production and review plan then follows the task. Question: Which result is useful enough to build around? Discovery: The useful constraint is the output: each route requires different reviewed content, production and evaluation. That changes what you build first. Mechanism: Compare differentiated homeowner outcomes and their scope. Visitor action: Choose the first visitor outcome Consequence: Set the output and the evidence needed to evaluate it. Portable contribution: Keep the chosen outcome and its implementation tradeoff. Next: Make the experience blueprint yours. - Prepare for an inspection: Build a cinematic preparation journey that organizes observations and inspection questions. - Tradeoff: Needs reviewed guidance and a mobile-friendly saved brief. Evaluate completed useful preparations and qualified conversations. - Brief contribution: First outcome: inspection preparation; evaluate usable saved briefs and qualified conversations. - Compare written scopes: Help a homeowner organize questions about supplied roofing scopes and what is missing. - Tradeoff: Needs approved sample scopes and clear assessment boundaries. Evaluate whether the visitor can identify questions for the contractor. - Brief contribution: First outcome: organize scope-comparison questions using approved examples; keep property-specific advice with the contractor. - Understand your process: Use a cinematic story to show what happens before, during and after your company's work. - Tradeoff: Needs verified company process, approved media and a useful next-step guide. Evaluate comprehension and the intended next action. - Brief contribution: First outcome: explain the verified roofing process and preserve a useful next-step guide. ### Action:Delight — Your experience starts with one useful outcome. Name your company, the task and the result you want a homeowner to keep. The blueprint preserves your choices, evidence and next move. Question: What should the homeowner be able to take away? Discovery: The portable result turns the cinematic journey into a plan for your business. Mechanism: Collect the inputs that change the plan and export a personal blueprint. Visitor action: Complete and save the blueprint Consequence: Carry a visitor task, scope and evaluation rule into the demo. Portable contribution: Visitor-entered company, outcome, constraints and success criterion. Next: Bring the blueprint to RoofClaw. - Your roofing company: Company name (optional) - What should the homeowner be able to do?: Prepare useful questions before an inspection (needed) - What must the experience respect?: Works on a phone; uses approved company guidance - How will you know it helped?: Usable preparation briefs and verified qualified conversations ### Action:Education — Get a RoofClaw built for Sparkitecture. Explore a RoofClaw configured with the Sparkitecture harness. Bring one homeowner question and your blueprint; use the demo to agree resources, scope and the experience you want to build. Question: What is the next concrete action? Discovery: The demo connects your visitor outcome with an implementation plan. Mechanism: Continue to the existing RoofClaw booking route with your saved blueprint. Visitor action: Book a RoofClaw demo Consequence: Open RoofClaw's demo booking destination. Portable contribution: Next move: bring this blueprint to https://www.roofclaw.com/demo . Next: Take your blueprint with you. ## Blank personal preparation brief Chosen path: Information to gather: Questions and unresolved facts: Acceptable result: Reviewer or owner: Next action and review date: ## Claims, sources and limits Review is an editorial status. A reviewed claim is not independent proof of a measured outcome. ### sparkitecture-definition / editorial / reviewed Sparkitecture combines Spark, Value, Education and Realm in a useful journey: an answerable question, an investigation, a consequential choice and a portable next step. Scope: Our authored design method Limitations: A design proposition, not measured proof of traffic, retention or sales. The cited learning experiments motivate active exploration; they do not test Sparkitecture. - pnas-learning.txt / lines:5-5 - Selected supporting excerpt: Finding: In learning-and-advice tasks, participants using LLM summaries reported shallower learning and wrote advice that was shorter, less original and received lower recipient willingness-to-adopt ratings than advice from participants exploring web links. Some conditions held the information constant or compared tools on the same platform. - Limits: Agent-authored source note. Original: https://pmc.ncbi.nlm.nih.gov/articles/PMC12560091/ . Seven randomized online/laboratory experiments, including four in the main text and three conceptual replications. Topics included gardening, healthier living and financial scams; outcomes included self-report, advice-text analysis and recipient ratings. These tasks do not measure commercial conversion or Sparkitecture. Several learning outcomes were subjective or inferred from advice. The abstract reports 10,462 participants while the introduction gives 10,426; avoid repeating the unresolved total. Results may differ for factual lookup, advanced domains or other learning measures. ### roofing-examples / illustrative / reviewed A homeowner preparation guide can change its checklist when the visitor selects a leak, storm concern or planned replacement. Scope: Authored teaching examples Limitations: No roof diagnosis, price, site assessment, completed customer or conversion result is represented. ### tradeoff / editorial / reviewed Purposeful cinema makes a question, relationship or choice tangible. Give motion a job, make choices change the result, and preserve a complete native text route. Scope: Design recommendation for a bounded cinematic experience Limitations: No universal engagement or conversion improvement is asserted. Media adds production, performance and accessibility requirements; compare actual task success and qualified conversations before claiming an improvement. - nng-animation.txt / lines:5-5 - Selected supporting excerpt: Finding: Page Laubheimer describes restrained animation as useful feedback for recognized actions, state transitions, relationships and available interactions. Motion unrelated to the visitor's task can distract or overwhelm. - Limits: Agent-authored source note. Original: https://www.nngroup.com/articles/animation-purpose-ux/ . Expert usability guidance with interface examples and references to attention research; not a controlled landing-page conversion experiment. Supports purposeful motion, not a claim that animation improves engagement or revenue. Animation supplements clear navigation and should be brief and restrained. A useful reduced-motion route remains necessary. ### roofclaw-offer / editorial / reviewed Explore getting a RoofClaw configured to make Sparkitecture experiences; bring this blueprint to the existing RoofClaw demo to agree scope. Scope: User-requested commercial positioning Limitations: No price, delivery time, live integration or booked appointment is asserted. ### ccdocs-coverage / inferred / reviewed A technical comparison became a fight-night experience. Tom's Hardware covered the story and linked to it on October 1, 2026. The creator reports that happened within two days. That is evidence of earned editorial attention; retention and conversion still need analytics. Scope: Directly inspected editorial coverage and creator-reported timing Limitations: Article date and source link are verified. Launch interval, referral count, dwell time and conversion are not independently established. - case-video.json / lines:1-54 - Selected supporting excerpt: A technical comparison became a fight-night experience. Tom's Hardware covered the story and linked to it on October 1, 2026. The creator reports that happened within two days. That is evidence of earned editorial attention; retention and conversion still need analytics. - Limits: An agent-authored bounded research note links the original Tom's Hardware article; this excerpt is the note, not a quotation of the publisher. ### zombie-ui / editorial / reviewed Katie Dill warns about generic zombie UI in the linked Lenny's Podcast talk. Our response: give every visual a job and every interaction a useful consequence. Scope: Design framing inspired by the verified talk description Limitations: A speaker's viewpoint; no endorsement of RoofClaw or measured Sparkitecture business benefit. Exact original transcript wording was unavailable. ### evidence-pew / measured / reviewed In Pew's March 2025 U.S. browsing panel, people clicked traditional results on 8% of Google visits with an AI summary, versus 15% without. Summary-source links were clicked on 1% of summary visits. Scope: Pew Research Center; 2025-07-22 Limitations: Observational association, one month, Google only. Different queries and recreated results can affect the comparison. This is not a measured traffic loss for roofing websites. - pew.txt / lines:5-5 - Selected supporting excerpt: Finding: In Pew's March 2025 U.S. browsing panel, people clicked traditional results on 8% of Google visits with an AI summary, versus 15% without. Summary-source links were clicked on 1% of summary visits. - Limits: Agent-authored source note. Original: https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ . 900 U.S. adults; 68,879 unique Google searches during March 2025. Search results were recreated April 7–17 to classify AI summaries and links. Observational association, one month, Google only. Different queries and recreated results can affect the comparison. This is not a measured traffic loss for roofing websites. ### evidence-ahrefs / measured / reviewed Ahrefs found AI Overview presence correlated with a 58% lower position-one desktop clickthrough rate relative to its modeled counterfactual. Scope: Ahrefs; 2026-02-04 Limitations: Selected keyword sample, desktop only, observational comparison. The 58% is relative to a modeled counterfactual, not 58 percentage points or all website traffic. No roofing conversion outcome was studied. - ahrefs.txt / lines:5-5 - Selected supporting excerpt: Finding: Ahrefs found AI Overview presence correlated with a 58% lower position-one desktop clickthrough rate relative to its modeled counterfactual. - Limits: Agent-authored source note. Original: https://ahrefs.com/blog/ai-overviews-reduce-clicks-update/ . 300,000 keywords, split between 150,000 with AI Overviews and 150,000 informational queries without. Aggregated Search Console desktop CTR compared December 2023 with December 2025. Selected keyword sample, desktop only, observational comparison. The 58% is relative to a modeled counterfactual, not 58 percentage points or all website traffic. No roofing conversion outcome was studied. ### evidence-sparktoro / measured / reviewed SparkToro's Similarweb-based estimate found 68.01% of U.S. Google browser searches ended without a click in January–April 2026. Scope: SparkToro / Similarweb; 2026-06-08 Limitations: Excludes Google's mobile search app. Historical comparisons use different data providers and panels. Zero-click can mean a useful answer, reformulation or abandonment; it does not mean a lost customer. - sparktoro.txt / lines:5-5 - Selected supporting excerpt: Finding: SparkToro's Similarweb-based estimate found 68.01% of U.S. Google browser searches ended without a click in January–April 2026. - Limits: Agent-authored source note. Original: https://sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/ . Desktop/mobile web clickstream panel; combined estimates assume two-thirds mobile and one-third desktop. Mobile sessions end after ten seconds of inactivity; mobile paid-click share is modeled. Excludes Google's mobile search app. Historical comparisons use different data providers and panels. Zero-click can mean a useful answer, reformulation or abandonment; it does not mean a lost customer. ### evidence-cloudflare / measured / reviewed Google referrals to Cloudflare's news-customer cohort fell approximately 9% in March 2025 compared with January; April was approximately 15% lower. Scope: Cloudflare; 2025-08-29 Limitations: News publishers are not all websites or roofing businesses. AI launches coincided with the changes but this does not isolate causality. Crawl requests and human referred visits measure different things. - cloudflare.txt / lines:5-5 - Selected supporting excerpt: Finding: Google referrals to Cloudflare's news-customer cohort fell approximately 9% in March 2025 compared with January; April was approximately 15% lower. - Limits: Agent-authored source note. Original: https://blog.cloudflare.com/crawlers-click-ai-bots-training/ . Cloudflare network observations of news-related customers across the Americas, Europe and Asia. Fixed customer cohorts were used for crawler-growth comparisons. News publishers are not all websites or roofing businesses. AI launches coincided with the changes but this does not isolate causality. Crawl requests and human referred visits measure different things. ### evidence-wikimedia / measured / reviewed Wikipedia reported human pageviews down roughly 8% year over year in recent months of 2025, after updating its bot detection. It linked the trend to changes involving AI answers and social media. Scope: Wikimedia Foundation; 2025-10-17 Limitations: Classification rules changed, so Wikimedia advises care in interpretation. The explanation is an attribution, not a controlled causal estimate. This is a major informational publisher, not a roofing-company traffic study. - wikimedia.txt / lines:5-5 - Selected supporting excerpt: Finding: Wikipedia reported human pageviews down roughly 8% year over year in recent months of 2025, after updating its bot detection. It linked the trend to changes involving AI answers and social media. - Limits: Agent-authored source note. Original: https://wikimediafoundation.org/news/2025/10/17/new-user-trends-on-wikipedia/ . First-party pageview measurements across Wikipedia language versions; revised bot classification for March–August 2025. Classification rules changed, so Wikimedia advises care in interpretation. The explanation is an attribution, not a controlled causal estimate. This is a major informational publisher, not a roofing-company traffic study. ### evidence-google-traffic / self-reported / reviewed Google said total organic clicks were relatively stable year over year and click quality had increased, while traffic shifted between sites. It defined quality as users not quickly clicking back. Scope: Google; 2025-08-06 Limitations: An interested platform's aggregate statement. Per-query CTR can fall while total clicks stay stable as query volume grows. A quality-click proxy does not establish leads, sales or learning. - google-traffic.txt / lines:5-5 - Selected supporting excerpt: Finding: Google said total organic clicks were relatively stable year over year and click quality had increased, while traffic shifted between sites. It defined quality as users not quickly clicking back. - Limits: Agent-authored source note. Original: https://blog.google/products-and-platforms/products/search/ai-search-driving-more-queries-higher-quality-clicks/ . Google statement based on internal platform data; no underlying reproducible dataset or detailed method supplied in the post. An interested platform's aggregate statement. Per-query CTR can fall while total clicks stay stable as query volume grows. A quality-click proxy does not establish leads, sales or learning. ### evidence-google-guidance / editorial / reviewed Google recommends existing SEO fundamentals for AI features: crawl access, helpful native text, good page experience and structured data matching visible content. Supporting links need indexing and snippet eligibility. Scope: Google Search Central; Live documentation; checked 2026-10-04 UTC Limitations: Applies to Google's features. Eligibility does not guarantee crawling, indexing, ranking, AI citation or traffic. Google requires no special AI file or schema. - google-guidance.txt / lines:5-5 - Selected supporting excerpt: Finding: Google recommends existing SEO fundamentals for AI features: crawl access, helpful native text, good page experience and structured data matching visible content. Supporting links need indexing and snippet eligibility. - Limits: Agent-authored source note. Original: https://developers.google.com/search/docs/appearance/ai-features . Official implementation guidance for Google AI Overviews and AI Mode; not an experiment measuring design performance. Applies to Google's features. Eligibility does not guarantee crawling, indexing, ranking, AI citation or traffic. Google requires no special AI file or schema. ### evidence-nng-calculators / inferred / reviewed Usability research describes calculators and quizzes as exploratory decision aids. Inputs should meaningfully change outputs, and forcing contact details before revealing results can frustrate users. Scope: Nielsen Norman Group; 2024-03-22 Limitations: Supports useful interaction design, not a promise that every quiz increases retention or sales. Experienced users can reject generic outputs; decision support is one input to judgment. - nng-calculators.txt / lines:5-5 - Selected supporting excerpt: Finding: Usability research describes calculators and quizzes as exploratory decision aids. Inputs should meaningfully change outputs, and forcing contact details before revealing results can frustrate users. - Limits: Agent-authored source note. Original: https://www.nngroup.com/articles/calculator-expectations/ . Qualitative user-research observations summarized with examples. The public article does not provide a full sample frame or a quantitative conversion estimate. Supports useful interaction design, not a promise that every quiz increases retention or sales. Experienced users can reject generic outputs; decision support is one input to judgment. ## Continue Book a RoofClaw demo: https://www.roofclaw.com/demo ## Build identity Project: sparkitecture-by-vito Run: 20261004T152057880Z-be8b2115 Content fingerprint: 3c901e88014548e06756f36895565b951609318346b7d00db296a5a1952d587f Mode: public Publication: local artifact; this file does not establish a live deployment. A download records an action, not later use, learning or business value. ## Cinematic composition and draft continuity Nine received Vito films form full-screen chapters: invitation, mirage, evidence, four-pillar inspection, task-flow prototype, the CCdocs coverage trail, a consequential roofing choice, blueprint assembly and the next conversation. Aperture, inspection-line and opening-threshold transitions connect the question to a useful portable result. Native text and controls preserve the same meaning with Read mode, reduced motion or no JavaScript. Film playback and an authored direction do not establish entertainment, learning, traffic or conversion results. The page keeps a local browser draft across example changes, reloads and the update from the published roofing run `20261004T071429432Z-0e82fd49`. Explicit edits stay yours; untouched starters can follow a new example or situation. A saved current draft, including a reset, takes precedence over prior drafts. Browser-storage failure is visible, and both downloads remain available. Recovery is confined to named published predecessors. Original workshop pilots `20261004T075559385Z-480f8ff5` and `20261004T015845082Z-f2c174c4` retain identified original fields and provenance in both exports. Their storage bytes and activity history are not rewritten or imported. Blueprint continuity from the roofing page is identified separately from original workshop-pilot recovery. No predecessor fingerprint is presented as this page’s review, exposure or measured outcome.