## 1. Mandate, Scope, and Definition of Done **Outcome.** If this analysis is acted on, a visitor arriving from the paid-search ad "Project management that doesn't slow you down" should see, without scrolling, a first screen that continues the ad's promise and offers a visible route to the free-trial signup. A larger share of that ad's visitors should then reach and complete the signup form. The number to move is the current 2.1% conversion rate (A9), re-measured under the checks in Section 9. Any element that stalls visitors before the signup path is identified from real data, not assumed. **Stakeholders.** No individuals were named in the inputs. The requester records a name against each role before Section 12 is used. | Role | Responsibility | Name | |---|---|---| | Page owner | Owns the free-trial page and its content | Not supplied | | Copy/design approver | Approves any copy or above-the-fold design change; because the template is shared (A5), this includes the owners of the four other product lines | Not supplied | | Metric owner | Owns the free-trial signup conversion number and its definition | Not supplied | **Requirements.** | ID | Requirement | Source | |---|---|---| | R1 | Drive free-trial signups as the single primary goal of the page | Supplied: PRIMARY_GOAL (A16) | | R2 | Score the page as it currently exists, element by element, before any fix is proposed | Analysis mandate | | R3 | Judge message match against the ad headline "Project management that doesn't slow you down" specifically | Supplied: TRAFFIC_SOURCE (A18) | | R4 | Hard constraint: the page template is shared across 4 other product lines and cannot be restructured | Supplied (hard constraint): A5 | | R5 | Hard constraint: only copy and above-the-fold elements may change | Supplied (hard constraint): A6 | | R6 | Gap: the signup form sits below the fold while the heatmap shows almost no scroll past the hero video, so a route to the form for non-scrolling visitors must be assessed | Gap revealed by A4, A11 | | R7 | Gap: the reason for the 58% bounce is unknown and must be established, not assumed | Gap revealed by A10, A15 | | R8 | Gap: the 2.1% conversion rate, 58% bounce rate, and heatmap have no stated definition, period, sample size, or tracking verification | Gap revealed by A9–A12 | | R9 | Gap: no page-speed data (including the hero video's effect), device split, or mobile-experience data was supplied | Gap revealed by A13, A14 | | R10 | Gap: headline wording, any above-the-fold CTA, trust signals, form fields, and required disclosures are not described | Gap revealed by A7, A19 | | R11 | Gap: it is unknown whether copy and above-the-fold edits stay confined to this page or propagate to the 4 other product lines | Gap revealed by A8 | | R12 | Fixes are recommended, not authorized; nothing ships without Section 12 approval | Analysis mandate | **Explicit exclusions.** - The ad's own targeting: keywords, bids, ad copy, and audience settings. - The four other product-line pages that share the template, except to confirm they are not changed (R4, R11). - The template's structure, including form position and block order, which is locked by A5 and A6. - Structural redesign of the three feature blocks. Their copy is in scope but low priority, because the heatmap suggests few visitors reach them (A11). - Post-signup onboarding, pricing, and the trial product itself. - Visitors from traffic sources other than the paid-search ad (see Section 5). **Solution shape.** The following are directly observable from the supplied description (no URL was included, so observation is limited to what was described) or were supplied as data: - the hero video, the three feature blocks, and the form below the fold (A2–A4); - the conversion rate, bounce rate, and heatmap finding (A9–A11). The following require real data or direct page review to confirm: an above-the-fold CTA (A7), edit scope across the shared template (A8), data quality (A12), page speed (A13), device split (A14), the cause of bounce (A15), and the on-page headline's match to the ad (A19). The page owner closes the page-review gaps, the metric owner closes the data-definition gaps, and the developer closes the speed and template-scope gaps. **Completion criteria.** - **Scorecard (R1–R3, R6):** PASS when every score in Section 4 cites a named element on a dated capture of the live page or a real number, and every unscored row names its reason. A reviewer checks this against the capture. - **Constraints (R4, R5, R11):** PASS when a diff of every shipped change shows only copy and above-the-fold edits, and the four other product-line pages render identically before and after. - **Data (R7–R9):** PASS when each Section 7 item shows a dated source (analytics export, speed-test report, or user-test notes) and a resolved status. - **Page inventory (R10):** PASS when an inventory of headline, CTAs, trust signals, form fields, and disclosures is recorded and the rows marked "Not scored" are scored. - **Fixes and approvals (R12):** PASS when every shipped fix has a completed Section 12 row with approver and date, and the Section 11 post-change check is recorded. - **Metric (R1):** PASS when the conversion rate is re-measured on the same segment over a window of the same length, with Section 9 checks C13–C15 passing. A scorecard alone is NOT completion (→ Section 13). ## 2. Analyst Instructions ``` Analyze the real page as it currently exists, not a remembered or assumed version of it. Load or review the actual page and current analytics first. Check for a prior analysis of this same page before re-deriving findings that already exist. Resolve every <VALIDATE> item using real analytics, a real page-speed test, or real user testing — not internal guesswork. Produce actual scores grounded in actual page elements and actual data; do not stop at a plausible-sounding critique. Run the data-quality checks defined in Section 9 before trusting any performance number in the scorecard. Stop treating a tracking gap as a visitor-behavior finding; report a concrete blocker when progress needs data access, tooling, or testing outside your authority. Do not present a suspected issue as a confirmed cause of underperformance, invent a bounce rate or scroll depth, or claim a speed test was run when it wasn't. Continue safe, low-cost analysis of directly observable page elements while a fix-critical validation is pending. Ship a page change or present findings as confirmed only within approved authority. Report progress using actual data and actual before/after comparisons, and clearly separate observed, data- supported, and confirmed findings. Operational defaults (which page-speed tool, how the scorecard is formatted) may be chosen and documented autonomously. Do not autonomously invent analytics results or claim a test was run when it was not. ``` In particular: do not assume that the hero video, or the below-the-fold position of the signup form, is the reason 58% of visitors leave — that is the `<VALIDATE: bounce-reason>` validation item, not an analysis default. ## 3. Page and Performance Assumption Ledger Confidence reflects how far the claim is established now: High (supplied and unambiguous), Medium (supplied but a definition or context is missing), Low (unverified). **3a. Claims directly observable from the page (as described)** | ID | Claim | Source | Confidence | Consequence if wrong | |---|---|---|---|---| | A1 | The page is the free-trial page of the project management tool | Supplied | High | Analysis targets the wrong page; all rows void | | A2 | The page includes a hero video at the top | Supplied | High | F3 and F4 lose their basis | | A3 | The page includes three feature blocks | Supplied | High | Feature-block exclusions in Section 1 need revisiting | | A4 | The signup form is below the fold | Supplied | High | F1's rationale and the form-friction score fall away | | A5 | The template is shared across 4 other product lines and cannot be restructured | Supplied (hard constraint) | High | A restructuring fix would affect four other product lines | | A6 | Only copy and above-the-fold elements may change | Supplied (hard constraint) | High | Fix eligibility in Section 6 must be redone | | A7 | No signup CTA or route to the form exists above the fold | Unverified — `<VALIDATE: above-fold-cta>` | Low | F1 may be redundant; the CTA score changes; part of the 2.1% may come from an existing hero CTA | | A8 | Copy and above-the-fold edits made for this page stay confined to this page and do not alter the four other product-line pages | Unverified — `<VALIDATE: template-scope>` | Low | A shipped change alters other product lines, breaching A5 | **3b. Claims about current performance** | ID | Claim | Source | Confidence | Consequence if wrong | |---|---|---|---|---| | A9 | The conversion rate is 2.1% | Supplied | Medium (definition, period, and denominator not stated) | The baseline for every before/after comparison shifts | | A10 | The bounce rate is 58% | Supplied | Medium (definition and tool not stated) | The size of the leak is misjudged | | A11 | The heatmap shows almost no scroll past the hero video | Supplied | Medium (tool, period, and device mix not stated) | The below-the-fold reachability finding weakens | | A12 | The figures in A9–A11 use consistent definitions, cover an adequate sample and period, and the signup event fires correctly | Unverified — `<VALIDATE: conversion-baseline>` | Low | Data-supported scores downgrade to observed-only; all measurement is unreliable | | A13 | The hero video adds meaningful load time before the page is usable | Unverified — `<VALIDATE: page-speed>` | Low | F3 is unjustified; the speed row stays unscored | | A14 | A material share of paid-search visitors arrive on mobile | Unverified — `<VALIDATE: mobile-share>` | Low | Mobile fixes are over- or under-weighted | | A15 | Visitors leave because the first screen shows no immediate path to signup or does not continue the ad's promise | Unverified — `<VALIDATE: bounce-reason>` | Low | Suspected only; fixes aimed at it may miss the actual cause | **3c. Claims about message match with the traffic source** | ID | Claim | Source | Confidence | Consequence if wrong | |---|---|---|---|---| | A16 | The primary goal is free-trial signups | Supplied | High | The whole prioritization changes | | A17 | Traffic is mostly paid search; the remainder is not described | Supplied | Medium | Segment scoping in Section 5 changes | | A18 | The ad headline is "Project management that doesn't slow you down" (description lines, keywords, and final URL not supplied) | Supplied | High | The message-match evaluation is misdirected | | A19 | The above-the-fold copy repeats or clearly continues the ad's promise | Unverified — `<VALIDATE: headline-match>` | Low | The message-match score and F2 change | **Least certain element.** The hero video is the element whose impact on the primary goal is least certain. It could be helping (explaining the product), hurting (load time, a barrier to the form), or neutral, and no supplied data on plays, autoplay behavior, load weight, or engagement separates these. Section 9's checks C9 (speed-test) and C6 (scroll and heatmap pull) probe it first. ## 4. Element-by-Element Scorecard Scores run 1–5 (5 is best). "Not scored" means the described page contains no evidence to rate the element, and no rating is invented (absence is not asserted either). Those rows are scored after the page review in Section 8, task T1. | Element | Score 1–5 | Evidence | Confidence | |---|---|---|---| | Headline and value-proposition clarity | Not scored | No headline or subhead wording was described; its match to the ad is unresolved (A19, `<VALIDATE: headline-match>`) | n/a | | Call-to-action prominence and clarity | 2 | As described, the only conversion entry point is the below-fold signup form (A4); no above-the-fold CTA is described (A7, `<VALIDATE: above-fold-cta>`); the heatmap shows almost no scroll past the hero video (A11) | observed (observed-only; rests on an unresolved `<VALIDATE>`) | | Trust signals (social proof, security, credibility markers) | Not scored | None described; presence or absence is not asserted | n/a | | Form or purchase-flow friction | 1 (reachability only) | The signup form sits below the fold (A4) while almost no visitors scroll past the hero video (A11). Field count, labels, and length are not described, so field-level friction is not scored | data-supported (reachability); cause of low conversion not confirmed | | Page load speed | Not scored | No speed test exists; the hero video (A2) is only a plausible factor (A13, `<VALIDATE: page-speed>`) | n/a | | Mobile experience | Not scored | Device split unknown (A14, `<VALIDATE: mobile-share>`); no mobile rendering described | n/a | | Message match with the traffic source | 2 | The ad promises a tool that "doesn't slow you down" (A18), but the path to signup requires scrolling past a hero video (A4) that the heatmap shows visitors mostly do not pass (A11). Wording-level match is not assessed (A19, `<VALIDATE: headline-match>`) | observed (observed-only; rests on an unresolved `<VALIDATE>`) | | Hero video (above-the-fold anchor element; added row) | Not scored | Autoplay, length, poster, captions, and engagement are not described (A2, A13) | n/a | **Lowest-scoring element most directly tied to the primary goal:** form or purchase-flow friction (reachability), score 1. It is the only score resting on supplied data rather than an unresolved `<VALIDATE>` item, and it is a best-practice-based, not confirmed, cause of underperformance (Section 9, C7). **Evidence rule.** This scorecard is a draft diagnosis for review, not a confirmed report. A stakeholder-facing verdict is gated by Section 12, not by finishing this section. ## 5. Visitor Segment and Message-Match Contracts **Segment contract: paid-search clickers on the ad "Project management that doesn't slow you down"** | Field | Content | |---|---| | Name | Paid-search visitors who clicked the ad with this headline (A17, A18) | | What they expected on arrival | Inferred from the ad text alone: a free-trial page for a project management tool that continues the "doesn't slow you down" promise and lets them start quickly. What visitors actually expected is unmeasured (`<VALIDATE: bounce-reason>`) | | Where the page currently matches | Topical match: the page is a free-trial page for a project management tool (A1) | | Where the page currently mismatches | Plausible mismatch only: the signup form is below a hero video most visitors do not scroll past (A4, A11), and the first-screen wording versus the ad is unknown (A19, `<VALIDATE: headline-match>`). Not a confirmed cause of the 58% bounce | | Owner | Page owner, with the copywriter for wording and the designer for the first screen (names not supplied) | **Segments not analyzed separately (each omission is a deliberate decision):** - Visitors from sources other than paid search (organic, direct, email, social): the supplied context says only "mostly paid search" and does not describe or distinguish them. - Paid-search visitors split by keyword, search query, device, geography, or new versus returning: no such breakdown was supplied. Device split is queued as `<VALIDATE: mobile-share>`. - Visitors from other ads or ad variants: only one headline was supplied. **Internal vs. external claim separation.** Only a confirmed finding on message match (one that has passed Section 9's confirmed-finding checks and Section 12 approval) may be shared outside the immediate team, for example with the ad team or an agency. The expectation inference above, the "plausible mismatch" statement, and every observed-only score stay internal working notes. ## 6. Prioritization Architecture and Fix Backlog **Framework.** Impact, Confidence, and Ease are each scored 1–5 (5 is best) with a one-line reason. Total is the sum (maximum 15). A score marked (stand-in) rests on a fix-critical item in Section 7 and must not be treated as a confirmed high-impact rating until that item resolves. | ID | Fix | Impact | Confidence | Ease | Total | Decision | |---|---|---|---|---|---|---| | F1 | Add a primary free-trial CTA in the hero area that jumps to the existing form (label set by the copywriter to match the actual offer) | 5 (stand-in): closes the gap between the hero and the below-fold form for visitors who do not scroll (A4, A11) | 3: heatmap supports the reachability issue (A11), but the cause of bounce (A15) and any existing CTA (A7) are unconfirmed | 4: above-the-fold element, permitted by A6; needs an anchor target on the existing form and template-scope confirmation | 12 | Top priority (provisional) | | F2 | Rewrite the hero headline and subhead to carry the ad's promise, using only claims the product supports | 4 (stand-in): directly targets message match with A18 | 2: current wording is unknown (A19) | 5: copy-only change | 11 | Next; draft in parallel | | F4 | Rebalance the hero so the video shares the first screen with a headline and CTA instead of occupying it alone | 4 (stand-in): the heatmap shows scroll stops at the hero video (A11) | 2: hero size, autoplay, and layout are not described | 3: above-the-fold element, but the hero is part of a shared template, so R4 needs checking | 9 | Deprioritized: follows F1 and F2 results | | F3 | Change hero video load behavior (poster image, deferred load, compressed source) | 3: only helps if load time contributes to bounce (A13) | 2: no speed test exists | 3: needs a developer; hosting setup unknown | 8 | Deprioritized: run the speed test first | | F5 | Add an above-the-fold trust element from real, verifiable assets (customer logos, a genuine rating, or a factual line about trial terms) | 3: trust effect is not evidenced here | 2: whether any trust signals already exist is unknown | 3: needs approved assets and legal-checked claims | 8 | Deprioritized: needs the page inventory | | F6 | Adjust below-fold form field labels or count | 3: friction at field level is unknown | 1: fields are not described | 2: the form belongs to the shared template; field removal is likely structural | 6 | Deprioritized: blocked by the constraint and missing data | | F7 | Move the signup form above the fold or reorder the template | Not scored | Not scored | Not scored | n/a | Excluded: violates R4 and R5 (A5, A6) | **Expected effect and post-ship check (each fix)** | Fix | Expected effect on the primary goal | How to check afterward | |---|---|---| | F1 | More first-screen visitors reach the signup form | Hero CTA click-through and signup conversion for paid-search sessions versus the baseline, same window length, C13–C15 passing | | F2 | More visitors continue past the first screen | Bounce, scroll depth past the hero, and conversion for the ad's sessions | | F4 | More visitors scroll or click before leaving | Scroll depth past the hero and CTA clicks | | F3 | Shorter time to a usable first screen | Real speed test before and after, then bounce on the affected device profile | | F5 | Higher CTA click-through or signup rate | CTA clicks and signup conversion, segmented by exposure | | F6 | Higher form completion | Form starts versus submits (requires a form-start event) | **Top priority: F1 (provisional).** It best satisfies R6, reaching the signup path for visitors who do not scroll past the hero, within R4 and R5. It is ranked first for drafting and validation only; it must not be treated as a confirmed high-impact fix until `<VALIDATE: above-fold-cta>` and `<VALIDATE: bounce-reason>` resolve. **Backlog map (deprioritized fixes; owner roles, individuals to be recorded by the requester)** | Fix | Score | Owner for revisiting | Revisit when | |---|---|---|---| | F4 | 9 | Designer | F1 and F2 have been measured, or `<VALIDATE: bounce-reason>` points to the video | | F3 | 8 | Developer | The Section 9 speed-test (C9) result is available | | F5 | 8 | Designer, with the page owner | The page inventory (C11) shows what trust signals exist | | F6 | 6 | Developer, with the page owner | Form fields are described and the template owner confirms what may change | | F7 | n/a | Page owner (decision only) | Only if the constraint itself is lifted | **Deliberately NOT recommended here:** - moving the form or restructuring the template (excluded by R4 and R5); - removing the hero video outright without evidence that it hurts (F3 and F4 are conditional); - pop-ups or exit-intent overlays, which add interruption against the ad's promise and are unsupported by any supplied data; - countdown timers or scarcity claims, which no supplied fact supports; - invented testimonials, ratings, or statistics; - redirecting paid traffic to a different page or template, which is outside this mandate (Section 1 exclusions) and is noted so that it is a deliberate exclusion, not an oversight; - bundling many changes into a single test, which would make any effect unattributable. Any fix not scored in this section is ruled out of this backlog until it is scored here. ## 7. Validation and Data Requirements **Fix-critical items** **`<VALIDATE: above-fold-cta>`** - **What must be established:** whether any signup CTA or route to the form exists above the fold, and where it points (A7). - **How to establish it:** run real user testing / confirm with real data. - **Dependent tasks/actions:** CTA score (Section 4); F1 impact rating; AC3 (Section 12). - **Status until resolved:** current best estimate is that, as described, no above-the-fold CTA exists (observed-only). F1 must not be prioritized as high-impact until resolved. **`<VALIDATE: headline-match>`** - **What must be established:** the actual first-screen headline and subhead wording, and whether it continues the ad's promise (A19). - **How to establish it:** run real user testing / confirm with real data. - **Dependent tasks/actions:** headline score and message-match score (Section 4); F2 impact rating; AC2 and AC9 (Section 12). - **Status until resolved:** current best estimate is a message-match score of 2, observed-only, based on the conversion path rather than wording. F2 must not be prioritized as high-impact until resolved. **`<VALIDATE: bounce-reason>`** - **What must be established:** why visitors leave, and which of the suspected contributors (no visible path to signup, the hero video, load time, mismatch with the ad) actually contribute (A15). - **How to establish it:** run real user testing / confirm with real data. - **Dependent tasks/actions:** Section 5 expectation inference; F1, F2, F4 impact ratings; AC3, AC4, AC9 (Section 12); confirmed-finding rows C7 and C16. - **Status until resolved:** current best estimate is that the contributors are suspected only. No suspected cause is treated as confirmed, and F1, F2, and F4 must not be prioritized as high-impact until resolved. **`<VALIDATE: template-scope>`** - **What must be established:** whether copy and above-the-fold edits on this page can be made without altering the four other product-line pages (A8). - **How to establish it:** run real user testing / confirm with real data. - **Dependent tasks/actions:** every shipping action in Section 12 (AC2–AC8); constraint checks in Section 8 (T4, T5). - **Status until resolved:** current best estimate is that edits might propagate to all five pages, and the analysis treats them that way. No fix may be prioritized as high-impact or shipped until resolved. **Routine items** **`<VALIDATE: page-speed>`** - **What must be established:** real load metrics for the page on mobile and desktop profiles, and the hero video's share of page weight (A13). - **How to establish it:** check current analytics or run a real page-speed test before finalizing. - **Dependent tasks/actions:** page-speed score (Section 4); F3 prioritization. **`<VALIDATE: conversion-baseline>`** - **What must be established:** the definition, window, denominator, and sample of the 2.1% conversion rate, 58% bounce rate, and heatmap, and whether the signup event fires correctly (A12). - **How to establish it:** check current analytics or run a real page-speed test before finalizing. - **Dependent tasks/actions:** confidence labels in Section 4; prioritization of anything relying on these numbers; C8, C13, C14, C15. **`<VALIDATE: mobile-share>`** - **What must be established:** the device split of paid-search sessions and per-device conversion and bounce (A14). - **How to establish it:** check current analytics or run a real page-speed test before finalizing. - **Dependent tasks/actions:** mobile-experience score (Section 4); mobile-specific prioritization. **Existing-data-first.** Existing analytics, the heatmap, and any prior analysis of this page are checked before any new user testing is commissioned. Real user testing is reserved for fix-critical questions that existing data cannot answer. Any performance figure quoted anywhere in this document is either sourced (A9–A11 came from the requester) or flagged with a `<VALIDATE: id>` tag, and no figure is ever supplied for plausibility. ## 8. Investigation Sequence and Requirement Traceability **Task list (dependency-ordered)** 1. **T1: Page review and data pull.** Load the live page (the URL was not supplied; the requester provides it) and capture the current version. Inventory the headline, CTAs, video behavior, trust signals, form fields, and required disclosures. Pull analytics (conversion, bounce, device split, scroll and heatmap, landing-session volume) and check for a prior analysis of this page. IDs: A1–A4, A7, A9–A11, A14, A19; R2, R10. Dependencies: none. 2. **T2: Run the Section 9 checks** (analytics-pull and speed-test rows) before trusting any number. IDs: A9–A13; R8, R9. Dependencies: T1. 3. **T3: Draft the scorecard per Section 4.** Score the rows marked "Not scored" once T1 supplies evidence, and downgrade any score resting on an unresolved `<VALIDATE>` item to observed-only. IDs: A2–A4, A7, A19; R2, R3, R6. Dependencies: T1, T2. 4. **T4: Deterministic constraint check for R4 (template not restructured).** Compare each proposed change against the template structure and confirm section order and containers are unchanged. Render the four other product-line pages before and after in staging and confirm they are identical. IDs: R4, A5, A8. Dependencies: T3 (candidate changes), T1. 5. **T5: Deterministic constraint and accessibility check for R5 (copy and above-the-fold only).** Enumerate every proposed change and tag it copy, above-the-fold, or neither; PASS only if none is tagged neither. Run the accessibility checks in Section 10 on the changed first screen. IDs: R5, A6. Dependencies: T3. 6. **T6: Resolve the fix-critical validations** (`above-fold-cta`, `headline-match`, `template-scope`, `bounce-reason`), using existing data first and real user testing only where needed. IDs: A7, A8, A15, A19; R7, R10, R11. Dependencies: T1, T2. 7. **T7: Fix prioritization.** Re-score Section 6 and remove the (stand-in) marks once T6 resolves. IDs: R6, R12. Dependencies: T4, T5, T6. 8. **T8: Report blockers** on every unresolved item, by name, before any change ships. IDs: all open R and A items. Dependencies: T7. **Traceability table** | Requirement/Assumption | Element/Fix | Owner | Acceptance Check | |---|---|---|---| | R1 | Signup form and primary CTA (F1) | Metric owner | Analytics signups reconcile to backend trial-account records for the same window (C1) | | R2 | Scorecard (Section 4) | Analyst | Each score cites an element on the dated live-page capture (C2) | | R3 | Hero copy versus ad (F2) | Copywriter | Side-by-side of ad text and first-screen copy recorded (C3) | | R4 | Shared template structure | Developer | Structure diff is empty; other four pages identical before and after (C4) | | R5 | Scope of changes | Developer | Every change tagged copy or above-the-fold, none "neither" (C5) | | R6 | Form reachability (F1, F4) | Designer | Scroll-depth pull by device reproduces A11 (C6) | | R7 | Cause of bounce | Analyst | Each candidate reason supported or ruled out by real data or testing (C7) | | R8 | Metric definitions | Metric owner | Each figure has a stated formula, window, and source (C8) | | R9 | Speed and device data | Developer | Real speed-test report and device-split pull saved (C9, C10) | | R10 | Page inventory | Page owner | Inventory complete with capture references (C11) | | R11 | Template scope | Template owner | Written scope confirmation or change control from the other four lines' owners (C12) | | R12 | Approval register | Page owner | Every shipped fix has a completed Section 12 row | | A1 | Page identity | Page owner | Live URL confirmed as the free-trial page (C2) | | A2 | Hero video | Designer | Video present on the capture; behavior recorded (C11) | | A3 | Feature blocks | Page owner | Three blocks present on the capture (C2) | | A4 | Form position | Designer | Form position confirmed on the capture at desktop and mobile widths (C2, C6) | | A5 | Template constraint | Developer | C4 | | A6 | Scope constraint | Developer | C5 | | A7 | Above-fold CTA | Page owner | CTA inventory in C11; `<VALIDATE: above-fold-cta>` resolved | | A8 | Edit scope | Template owner | C12 | | A9 | Conversion rate | Metric owner | Definition and window documented (C8, C13) | | A10 | Bounce rate | Metric owner | Definition and tool documented (C8) | | A11 | Heatmap finding | Analyst | Scroll pull by device with adequate sample (C6, C13) | | A12 | Data quality | Metric owner | C8, C13, C14, C15 all PASS | | A13 | Video load impact | Developer | Speed-test report with the video's share of page weight (C9) | | A14 | Mobile share | Analyst | Device split and per-device figures (C10) | | A15 | Cause of bounce | Analyst | C7, confirmed-finding, remains NOT CONFIRMED until Section 12 approvals | | A16 | Primary goal | Metric owner | Goal event in analytics matches free-trial signups (C1, C14) | | A17 | Traffic mix | Analyst | Source breakdown pulled and recorded (C1) | | A18 | Ad headline | Ad owner | Full ad text pulled from the ad account (C3) | | A19 | Headline match | Copywriter | Wording compared side by side; `<VALIDATE: headline-match>` resolved (C3) | ## 9. Data-Quality and Readiness Checks A scorecard is only as good as the data behind it. Every performance number here (A9–A11) must be checked for quality before it is trusted, because a tracking gap or a mismatched definition looks exactly like a visitor-behavior finding and is not one. **Stand-in note.** Four fix-critical items (`above-fold-cta`, `headline-match`, `bounce-reason`, `template-scope`) are currently held by stand-in estimates (Section 7). Any fix priority built on a stand-in remains UNCONFIRMED until the item resolves. Operational default chosen: PageSpeed Insights (field and lab data) for speed tests, with alternatives permitted if documented alongside the results. | Assumption/Req | Check level | Method | Expected result | Pass condition | Prerequisites | |---|---|---|---|---|---| | C1 · R1, A16, A17 | analytics-pull | Pull paid-search landing sessions and completed signups for the window; reconcile signups against backend trial-account records | Analytics signups reconcile to real trial accounts | Both counts exist for the same window and any gap is explained | Analytics and backend access | | C2 · R2, A1, A3, A4 | observation | Capture the live page (dated desktop and mobile screenshots plus an HTML snapshot); check every Section 4 evidence cell against it | Every scored row cites an element present on the capture | No scored row cites an element absent from the capture | Live URL (not supplied) | | C3 · R3, A18, A19 | observation | Place the full ad text beside the first-screen copy and record wording overlap and promise continuity | A documented match or mismatch statement per visible element | Side-by-side exists and is labeled observed, not confirmed cause | Ad headline, description lines, and final URL from the ad account | | C4 · R4, A5 | observation | Compare each proposed change with the template structure; render the four other product-line pages before and after in staging | No structural change; other pages identical | Zero structural differences; zero differences on the other four pages | Staging access; change list | | C5 · R5, A6 | observation | Enumerate each proposed change and tag it copy, above-the-fold, or neither | Every change tagged copy or above-the-fold | No change tagged neither | Change list | | C6 · R6, A4, A11 | analytics-pull | Pull scroll-depth and heatmap data by device for the window; measure the share of sessions reaching the form | Scroll data reproduces the supplied finding | Depth distribution exists for desktop and mobile with sample per C13 | Heatmap tool access | | C7 · R7, A15 | confirmed-finding | Combine existing analytics (bounce by device, load time, entry query) with real user testing on the first screen | NOT CONFIRMED until Section 12 approvals granted | Each suspected reason is supported or ruled out by real data or testing, and the finding is approved through Section 12 | C1, C6, C9, C10 complete; existing-data-first review | | C8 · R8, A9, A10, A12 | analytics-pull | Document the definition, window, filters, and denominator of the conversion, bounce, and heatmap figures | A written definition per figure | Each figure has a stated formula, window, and source | Analytics access | | C9 · R9, A13 | speed-test | Run a real page-speed test on the real URL with mobile and desktop profiles; record load metrics and the hero video's share of page weight from the network waterfall | Recorded metrics with tool, date, and profile | Report saved and video weight identified | Live URL | | C10 · R9, A14 | analytics-pull | Pull the device split and per-device conversion and bounce for paid-search sessions | Device split recorded | Per-device figures exist with sample per C13 | Analytics access | | C11 · R10, A2, A7 | observation | Inventory above-the-fold elements, existing CTAs and their targets, trust signals, form fields, and required disclosures | A complete inventory | Every element type in Section 4 has a recorded "present" or "absent" entry with a capture reference | C2 capture | | C12 · R11, A8 | observation | Inspect the template configuration or CMS to see whether the hero and copy are per-page or template-wide; ask the template owner | Written scope confirmation | Scope confirmed page-level, or change control obtained from the four other lines' owners | Template access | | C13 · sample size (A9, A10, A11, A12) | analytics-pull | Count sessions and signups in the window; compute an interval around the conversion rate | An interval narrow enough to compare against | Interval narrower than the smallest uplift agreed with the metric owner; otherwise the figure is reported as directional only | Metric owner's uplift threshold | | C14 · tracking integrity (A12, A16) | analytics-pull | Complete a real signup in a real browser under normal consent conditions; confirm the event fires once and is attributed to paid search; check for duplicates and misses | Event fires exactly once per completed signup | Exactly one event recorded with correct attribution | Test account; analytics access | | C15 · concurrent changes (A12) | analytics-pull | Compare change logs (ad account, site releases, other tests on the shared template, seasonality) across the window | A list of overlapping changes | No overlapping change, or the window is trimmed to exclude it | Ad, release, and test logs | | C16 · R12, A15 | confirmed-finding | Measure any shipped fix's effect on signups against the baseline | NOT CONFIRMED until Section 12 approvals granted | Fix shipped through an approved row; before and after measured under passing C13–C15 | An approved Section 12 row for the fix | **What each level proves.** An analytics-pull PASS proves only that the number is real: it was measured, defined, and sampled adequately. Only a confirmed-finding proves that an element caused the number. The two are never conflated. ## 10. Accessibility, Privacy, and Compliance Boundaries - **Shipping a fix while a fix-critical item is unresolved — High; BLOCKED.** No fix ships while any of the four fix-critical items in Section 7 is unresolved; a stand-in estimate does not count as resolution. This is the explicit no-ship rule. - **Cross-product-line propagation — High; BLOCKED BY `<VALIDATE: template-scope>`.** Because the template is shared (A5), an above-the-fold or copy edit could reach the four other product lines. Mitigation: check C4 and C12 must pass before any change ships, and a rollback must be checked on all five pages. - **Accessibility regression from a design change — Medium.** The main exposures are text contrast over or beside the video, missing captions or a transcript for the hero video, keyboard operability and focus after a CTA jumps to the form, and autoplay motion without a pause control. The video's current accessibility state is not described and is unassessed. Mitigation: run an automated audit plus a manual keyboard and screen-reader pass on the changed first screen, checked against WCAG 2.2 AA (a documented default), and require no new failures compared with the live page. - **Visitor-level data boundary — Medium.** The analysis uses aggregate analytics, heatmaps, and, if commissioned, consented user tests. Session recordings of identifiable individuals, form-field contents, and personal data are not viewed; any recording or heatmap tool must mask input fields and respect the site's existing consent settings. - **Never remove without sign-off — High.** Any required legal disclosure, consent or cookie mechanism, terms and privacy links, and trial-term or billing disclosures near the signup path stay on the page. None is described, so C11 inventories them, and the hero region is checked for any disclosure before it is edited. - **Brand-guideline conflicts — Medium.** A change that conflicts with the template's brand system is caught by a brand or design-owner review before a Section 12 row can be marked ready. - **Sensitive data collection — Not yet assessed.** The signup form's fields are not described. If it collects payment details or other sensitive data, extra privacy and security review is triggered. That review also applies to any above-the-fold capture element that would collect the same data. **Risk that stays genuinely unassessable until Section 7's fix-critical items resolve:** the risk that a hero change reduces signups by replacing an element that is currently carrying conversions (for example, an existing hero CTA or the video itself), which cannot be sized until `<VALIDATE: above-fold-cta>` and `<VALIDATE: bounce-reason>` resolve. ## 11. Implementation, Verification, and Rollback Plan **Analysis and internal review (runnable now).** Tasks T1–T5 and drafting of candidate copy for F1 and F2 for internal review only. Nothing is published. **Blocked on validation.** Every change to the live page is blocked by `<VALIDATE: template-scope>` and, depending on the fix, by `<VALIDATE: above-fold-cta>`, `<VALIDATE: headline-match>`, or `<VALIDATE: bounce-reason>` (Section 12 shows the status per action). **Approved low-risk fix implementation (→ Section 12).** A copy-only change (AC2) or an above-the-fold CTA that links to the existing form (AC3), staged and not exposed to full traffic, after its Section 12 row is approved. **Staged rollout checks.** Where feasible, expose the changed hero to a partial share of paid-search traffic, using the site's existing testing mechanism or an ad-platform experiment, before any full page change (AC7). Feasibility is unconfirmed. On a shared template, the test must not expose the other four product lines. The result is not read until C13–C15 pass. **Approved full rollout.** Proceeds only when ALL of the following hold: - Section 9's checks pass at their stated levels, including C4, C5, and C13–C15; - all Section 12 actions in the chain are complete and approved; - every fix-critical Section 7 item is resolved, not estimate-stubbed. **Bounded post-change verification.** After the fix ships, one concrete, low-cost real check: does the targeted metric move? For F1, compare hero CTA click-through and signup conversion for paid-search sessions over a window of the same length as the baseline. Also verify that no excluded element from Section 1 was accidentally changed, by diffing the template structure, the form and feature-block structure, and the four other product-line pages against their pre-change versions. **Rollback.** Confirm the current live version of the page before assuming a revert restores it exactly. Reverting the page does not undo data already collected under the changed version. Containment means republishing the prior version and flagging the rollback in the analytics record (annotating the date), not assuming a silent revert. Because the template is shared, a rollback is verified on all five pages. ## 12. Change and Publication Approval Register | Action ID | Purpose | Element/Scope | Expected Effect | Trigger Condition | Status | Approved by / Date | |---|---|---|---|---|---|---| | AC1 | Circulate the draft scorecard internally as a labeled draft diagnosis | Sections 1–11 as a document; no page change | Stakeholders can review and supply the missing inputs (URL, ad text, data access) | Requester wants review | READY FOR HUMAN REVIEW | ____ / ____ | | AC2 | Implement the hero headline and subhead copy fix (F2), staged | Hero copy only | Wording continues the ad's promise; effect on signups unmeasured | `headline-match` and `template-scope` resolved; C3, C4, C5 pass | BLOCKED BY VALIDATION (`<VALIDATE: headline-match>`, `<VALIDATE: template-scope>`) | ____ / ____ | | AC3 | Implement the primary hero CTA linking to the existing form (F1), staged | One above-the-fold element plus an anchor target | More first-screen visitors reach the form; effect on signups unmeasured | `above-fold-cta`, `bounce-reason`, and `template-scope` resolved; C4, C5 pass | BLOCKED BY VALIDATION (`<VALIDATE: above-fold-cta>`, `<VALIDATE: bounce-reason>`, `<VALIDATE: template-scope>`) | ____ / ____ | | AC4 | Implement the hero layout rebalance (F4), staged | Above-the-fold layout | More visitors scroll or click before leaving | F1 and F2 measured, or `bounce-reason` points to the video; C4, C5 pass | BLOCKED BY VALIDATION (`<VALIDATE: bounce-reason>`, `<VALIDATE: template-scope>`) | ____ / ____ | | AC5 | Implement the hero video load treatment (F3), staged | Hero video only | Shorter time to a usable first screen | C9 shows the video contributes materially; C4, C5 pass | BLOCKED BY VALIDATION (`<VALIDATE: template-scope>`) | ____ / ____ | | AC6 | Implement an above-the-fold trust element (F5), staged | One above-the-fold element from real assets | Higher CTA click-through or signup rate | C11 shows no equivalent trust signal exists; assets and claims cleared; C4, C5 pass | BLOCKED BY VALIDATION (`<VALIDATE: template-scope>`) | ____ / ____ | | AC7 | Run a partial-traffic test of the changed hero | Share of paid-search traffic to the staged variant | A measurable difference in CTA clicks and signups | One of AC2–AC6 approved and implemented; C13–C15 pass | BLOCKED ON PRIOR ACTION (AC2 and/or AC3); also `<VALIDATE: bounce-reason>` | ____ / ____ | | AC8 | Ship the full page change | Approved changes only, on the live page | Targeted metric moves; measured after ship | AC7 read out; every fix-critical item resolved; C4, C5 pass | BLOCKED ON PRIOR ACTION (AC7) | ____ / ____ | | AC9 | Share confirmed message-match and bounce findings outside the immediate team | Confirmed findings only (Section 5 rule) | External stakeholders receive verified statements | C7 and C3 confirmed-finding rows pass | BLOCKED BY VALIDATION (`<VALIDATE: bounce-reason>`, `<VALIDATE: headline-match>`) | ____ / ____ | | AC10 | Roll back a shipped change | Prior page version, all five template pages verified | Restores the prior page; collected data is not undone | Post-change check shows harm, or an excluded element changed | BLOCKED ON PRIOR ACTION (AC8 or AC7) | ____ / ____ | Approving one action never implies approval of another. Listing a rollback action does not mean it is to be executed now. A fix promoted from the deprioritized backlog (F6, for example) receives its own row only when promoted. ## 13. Completion Evidence and Handoff A final report must contain: - **Scores and fixes.** The confirmed scores and the fixes actually shipped, each mapped to a real page version or URL and its capture date. - **Section 9 checks.** Every check, C1–C16, itemized as PASS, FAIL, NOT RUN, or BLOCKED with one line of evidence each. - **Findings by level.** Observed, data-supported, and confirmed findings reported separately. An observed issue is never presented as a confirmed cause. - **Performance data.** Actual before and after performance data, only if a fix actually shipped and was measured; otherwise "NOT YET MEASURED". - **Open validations.** The status of every unresolved Section 7 item, by name: `<VALIDATE: above-fold-cta>`, `<VALIDATE: headline-match>`, `<VALIDATE: bounce-reason>`, `<VALIDATE: template-scope>`, `<VALIDATE: page-speed>`, `<VALIDATE: conversion-baseline>`, and `<VALIDATE: mobile-share>`. - **Residual limitations.** Any limitation still true at handoff, for example a device segment too small to analyze reliably (C13) or a live URL that was never supplied. - **Next steps.** The exact next steps to continue any outstanding validation or approval item, including who owns each. Fabricating analytics screenshots, test results, or approvals is prohibited. A well-argued critique alone is not evidence of the actual impact until a fix ships and is measured.