Beta Campaign Wireframes — Jun 8, 2026

Two surfaces for the closed-beta feedback campaign: an operator-facing admin site and a tester-facing gated walkthrough. Both use the Confidence Engine skin (ink/cream/moss palette, --ce-* tokens). Screenshots depend on Playwright MCP — run mcp__playwright__browser_navigate + mcp__playwright__browser_take_screenshot against each file path below.

Screen index

Admin site (`docs/ux/beta-admin/`)

File	Description
`index.html`	Tester list (default view) — table of all testers with status, sentiment bar, promote/revoke actions; stats row; "Next Round" trigger bar
`screens/add-tester.html`	Add tester form — email, first name, persona pre-classification (4 radio cards), internal notes, send-welcome checkbox
`screens/weekly-digest.html`	Digest preview — live email preview rendered in-page, schedule controls, include/exclude toggles
`screens/settings.html`	Settings — welcome + reminder email template editors with variable chips; CF Access policy status (read-only); round settings (expiry, auto-reminder schedule)

Tester site (`docs/ux/beta-tester/`)

File	Description
`index.html`	Welcome screen — personalised greeting, time estimate chips, "what to expect" 3-item list, founder note, start CTA
`screens/walk-dashboard.html`	Walk: Dashboard — simulated dark-mode dashboard mockup in left pane; 3 questions (free text + Likert + Y/N) in right pane
`screens/walk-backtest.html`	Walk: Backtest — backtest form + results simulation; 3 questions on clarity and intent-to-use
`screens/walk-tax.html`	Walk: Tax Tools — tax summary tiles + closed-position list with ST/LT/wash badges; 3 questions on utility and workflow
`screens/walk-strategies.html`	Walk: Strategies — strategy cards with rule fields; 3 questions on resonance and comprehension
`screens/survey.html`	Final survey — NPS (0–10), 4-item Likert battery, persona self-id grid, 2 free-text questions, submit
`screens/thank-you.html`	Thank-you — confirmation, "what happens next" (round 2 invite, early access), founder note

Design decisions made without explicit instruction

Two-column layout for walkthrough screens — mockup preview on the left, questions on the right. Avoids page-scroll confusion; tester can see the mockup while answering. Could alternatively be tabs or full-screen steps if feature-dev prefers.
Sentiment score 0–100 — computed server-side from Likert average + NPS + free-text weighting. The admin table shows it as a bar + number; the exact formula is a backend decision.
Admin lives in a standalone mini-app (not console.raxx.app) — mocked as Raxx.Beta with its own top-nav. If marketing-strategist picks a sub-route of console instead, the nav chrome needs to swap but the content is identical.
CF Access sync is implied, not built — the settings page shows a "Last sync" timestamp and "Sync now" affordance. The actual CF API call is a feature-developer decision.
Walkthrough URL format — assumed beta.raxx.app/walk/<token>. Marketing-strategist should confirm.
Persona options — used the four from the brief (structure-needer / skill-needer / both / unknown). Survey uses slightly softer plain-English labels ("I have a strategy" / "I'm still learning") to avoid jargon on the tester side.
Reminder auto-send cadence — defaulted to Day 3 + Day 7 in settings. This is a configurable default, not a hard-coded spec.
All mock data — Alice / Bob / Charlie / Dana / Elias / Fatima as placeholder testers. Ticker symbols (AAPL, SPY, TSLA) are illustrative only.

Open questions — needs PM / marketing-strategist / BLR resolution before feature-dev

Exact survey rubric — the survey questions in survey.html are placeholders aligned to the brief. PM card with the final rubric needs to land before feature-dev wires this to a data model.
Walkthrough URL structure — beta.raxx.app/walk/<token> assumed. Marketing-strategist to confirm sub-domain vs path.
Admin hosting decision — beta-admin.raxx.app vs sub-route of console.raxx.app. Changes which nav chrome wraps the admin.
NDA / legal copy on welcome screen — BLR may require a brief acknowledgement before the tester proceeds. Add a checkbox step between welcome and walkthrough if so.
Sentiment score algorithm — the 0–100 number is shown in the admin table. The weighting formula (NPS vs Likert vs per-screen scores) is unspecified; PM or data-science to define.
Round 2 cohort trigger logic — "Promote to next round" toggle is manual per the mockup. Auto-promotion rules (e.g., sentiment > 60 + status = complete) are unspecified.
Settings page access control — who else besides Kristerpher can access beta-admin? Console RBAC or standalone CF Access policy?

Handoff notes for feature-developer

All HTML is self-contained with inline <style>. No external dependencies.
The CE palette is declared as CSS custom properties in :root — match these tokens to the production theme system.
The left-pane "mockup frame" in walkthrough screens is a simplified simulation of the real UI. When the production app mockups are available (from raxx-mockups), replace the inline SVG/tile simulation with actual screenshots or iframes.
Progress bar fill width is set manually per screen (0%, 14.3%, 28.6%…) — compute (step - 1) / 6 * 100 in the real implementation.
The thank-you.html screen shows the full progress bar filled to 100% with all dots green — style only, not a JS animation.
Contact email throughout: support@raxx.app.
Legal footer address throughout: MooseQuest LLC dba Raxx · 502 W 7th St, Ste 100, Erie, PA 16502-1333.

Screenshots

Screenshots to be regenerated with Playwright MCP. Navigate each .html file at file:/// + absolute path, capture at 1440×900. Pending screenshot generation — HTML mockups are complete and reviewable without screenshots.