Evidence coverage
the /100 — and the exact signals behind each category20/100
★★★★★
Sparse evidence coverage
Offer clarity4/20
Free shipping
Action path0/20
no signals detected
Trust & proof0/20
no signals detected
Friction reduction10/15
FAQ / objection handlingFree shipping
Value reinforcement7/15
Us-vs-them comparison
Risk reversal & urgency0/10
no signals detected
Evidence Coverage Score: deterministic points across a fixed 6-category conversion rubric (Offer 20 · Action 20 · Trust 20 · Friction 15 · Value 15 · Risk/Urgency 10 = 100), each capped so no single area dominates. Every point traces to an on-page evidence string. It measures coverage, not quality — no opinions, no predictions.
Palette
every color by on-page usage · the roles doing their jobfull palette ordered by on-page usage %
primary#e0e4f3
accent#e0e4f3
bg#ffffff
surface#f9f9f9
text#191c1d
accent
Body copy in their text role, on a surface card.
the roles, doing their job
Typography
their type system, rendered liveThe quick brown fox60px
Jumps over the lazy dog48px
A subheading in their voice40px
Body copy set at their measured paragraph size. This is how their product reads to a first-time visitor — same family, same scale, same weight.
- List item one, at body scale
- List item two
- List item three
Shape
the corner-radius scale, drawn4px
8px
20px
28px
32px
50px
Offer & pricing
rebuilt from extracted facts — nothing inventedRECONSTRUCTION — rebuilt from extracted facts
$99.99
$25.00$999$27.75
One-timeSubscribe
source: step 2 · Product page — every value extracted, nothing invented
The funnel
2 steps walkedTactics detected
applicable conversion tactics — click any for its evidence string·Proof at every step
·Stacked free gifts
·Countdown / urgency timer
·Anchor / strikethrough pricing
·Subscription / auto-ship
·Per-unit price anchor
✓Us-vs-them comparison2×
home: comparison table pdp: comparison table
·On-page review wall
·Express wallets
✓Free shipping2×
home: free shipping pdp: free shipping
·Money-back guarantee
✓FAQ / objection handling
pdp: FAQ/accordion
·Proof above the fold
Proof & trust language
the social-proof strings, harvestedFree shipping — Free shipping (1)
Build brief — copy for your agent
the whole teardown as one grounded spec: tokens · proof language · detected tactics · funnel facts# Agent Brief — build in the style of fitbit.com > Machine-extracted from a live funnel teardown (2026-07-08). 100% facts — measured > tokens, detected patterns, harvested proof strings. No LLM interpretation. > Tokens: `tokens.json` / `tailwind.css`. **Category (label):** — **Funnel:** Homepage / Lander → Product page → Cart → Checkout ## Palette (area-weighted from live CSS) | Hex | Usage | Likely role | |-----|------:|-------------| | `#ffffff` | 49.1% | background | | `#f9f9f9` | 41.3% | background | | `#effffe` | 2.2% | background | | `#f9f7f4` | 2.1% | background | | `#e0e4f3` | 1.4% | background | | `#f9fafc` | 0.9% | background | | `#fdfaf7` | 0.4% | background | | `#e7ebf3` | 0.4% | background | Curated roles: `primary=#e0e4f3` · `accent=#e0e4f3` · `bg=#ffffff` · `surface=#f9f9f9` · `text=#191c1d` ## Type - Display: `Google Sans Flex` - Body: `Google Sans Flex` - Size scale (px): 10, 11, 12, 14, 16, 20, 32, 40, 48, 60 - Radius (px): 4, 8, 20, 28, 32, 50 ## Patterns detected (fact-based) — 3/13 - **Us-vs-them comparison** — A comparison table framing the brand against alternatives. _(home: comparison table; pdp: comparison table)_ - **Free shipping** — Free shipping / delivery surfaced in the funnel. _(home: free shipping; pdp: free shipping)_ - **FAQ / objection handling** — An accordion / FAQ block answering objections. _(pdp: FAQ/accordion)_ ## Proof language (harvested strings → templates) - `Free shipping` — e.g. "Free shipping" ## Funnel facts - Depth reached: 2 steps - Proof points: 1 across 2 steps - Free-gift line items (max on a step): 0 - Discount/anchor pricing: no · Countdown: no · Subscription: yes - Price continuity lander→checkout: not detected ## To replicate Match the palette + type tokens above, deploy the detected patterns at the same funnel positions, and reuse the proof-language templates with your own numbers.



