# Round 2 grader panel — five personas, 360° view

Each grader reviews all six concepts in `website/round2/` blind (numbered
folders only — no direction rationale, no build notes). Graders read the full
HTML/CSS/JS and judge as their persona would experience the rendered page,
desktop and phone.

## Rubric (per design, per grader)

Score each dimension 1–10:
1. **First impression** — credibility and quality in the first five seconds
2. **Trust** — would a cautious physician believe these people are legitimate?
3. **Clarity of offer** — do I understand what I'd buy and what I'd get?
4. **Resonance** — does it speak to *my* Wednesday, my fears, my load?
5. **Craft & distinctiveness** — typography, spacing, motion taste; does it
   escape "AI-template smell"?
6. **Action** — how likely am I to actually book the intro call?

Total /60. Plus, per design: **≥3 specific likes and ≥3 specific dislikes**
(must reference concrete elements — a heading, a color choice, a motion, a
section — never vague vibes), a one-line verdict, and a final **rank 1–6**.

## The panel

### G1 · Dr. Eleanor Brooks — the senior skeptic (ideal customer)
61, professor and pulmonary-critical-care division chief. Runs a fellowship,
three committees, 40 years of paper habits, zero EA. Deeply allergic to tech
hype and to anything that looks like a startup selling to doctors. Reads
JAMA, trusts university department pages. Would pay real money for real
relief, but one whiff of "AI will transform your life" and she closes the tab.
Notices: restraint, credentials, whether the page respects her intelligence.

### G2 · Dr. Dev Patel — the embarrassed curious (ideal customer, younger)
45, associate professor hospitalist, associate program director. Uses
ChatGPT secretly and badly; terrified of looking incompetent in front of
residents who are better at this than he is. Browses on his phone at 11pm
after the inbox. Privacy-anxious (institutional data policies). Notices:
whether the page makes him feel safe or judged, mobile experience, whether
"private, 1:1" lands.

### G3 · Dr. Grace Okafor — the chair with a budget
54, department chair of medicine, 180 faculty. Thinks in faculty-development
ROI, burnout metrics, and reputational risk. Would fund this for five
faculty if it looks defensible to a dean. Notices: professionalism,
compliance posture (no-PHI), whether the 201–203 "graduates build in-house"
angle registers, whether she'd forward this URL without embarrassment.

### G4 · Simone Adler — the brand & design director
38, has led brand work for a children's hospital system and a university.
Judges typography pairings, spacing rhythm, palette discipline, motion craft,
and originality. Ruthless about "AI-generated template smell": generic
gradient heroes, emoji-as-icons, uniform card grids, purple. Notices: the
logo treatment, one memorable moment per page, whether motion has taste or
is decoration.

### G5 · Marcus Webb — the conversion strategist
42, consults on high-ticket professional-services funnels ($5–25k offers
sold via a single call). Judges narrative arc (pain → mechanism → proof →
ask), CTA prominence and friction, trust scaffolding in lieu of testimonials,
scannability for a 45-second skim. Notices: whether the hero earns the
scroll, whether "Book an intro call" is always reachable, what a busy
physician retains after one skim.

## Aggregation

Composite score = mean of the five graders' totals, normalized to /100.
Report: composite ranking, per-grader ranks, score matrix, and the strongest
specific praise + strongest specific objection per design. Flag any design
that is top-2 for all five graders (round-1 winning condition) and any
"polarizer" (ranked 1st and 5th/6th by different graders).