Section Design Library

The bands real pages are built from

The unit you actually build. Every page is a stack of these bands — each card is one real section, cropped from a crawled page at its measured position. Pick the band you need.

Section

Showing 24 of 4783 sections

pepsico.com — a real Email Capture section
pepsico.com · Email Capture
tryprobe.co — a real Email Capture section
tryprobe.co · Email Capture
doppio.sh — a real Stats section
doppio.sh · Stats
doppio.sh — a real Stats section
doppio.sh · Stats
promptsloth.com — a real Stats section
promptsloth.com · Stats
linkeddit.com — a real Stats section
linkeddit.com · Stats
mapleads.ai — a real Stats section
mapleads.ai · Stats
bersyn.com — a real Stats section
bersyn.com · Stats
taprefer.com — a real Stats section
taprefer.com · Stats
mapgo.io — a real Stats section
mapgo.io · Stats
taprefer.com — a real Stats section
taprefer.com · Stats
agentkey.app — a real Stats section
agentkey.app · Stats
sowiks.com — a real Stats section
sowiks.com · Stats
supasidebar.com — a real Stats section
supasidebar.com · Stats
bulkpictools.com — a real Stats section
bulkpictools.com · Stats
mirrorfly.com — a real Stats section
mirrorfly.com · Stats
weely.ai — a real Stats section
weely.ai · Stats
getstaffscheduling.ca — a real Stats section
getstaffscheduling.ca · Stats
syncgtm.com — a real Stats section
syncgtm.com · Stats
tailscale.com — a real Stats section
tailscale.com · Stats
syncgtm.com — a real Stats section
syncgtm.com · Stats
kraflio.com — a real Stats section
kraflio.com · Stats
captainverify.com — a real Stats section
captainverify.com · Stats
nuvio.so — a real Stats section
nuvio.so · Stats
1797 funnels x-rayed2722 pages measured10 section types

X-ray your own funnel · free, no signup required

Method: every crawled page is segmented into full-width bands and each band classified into a section archetype by deterministic DOM signals: no LLM, no opinion. Prevalence counts each archetype once per page, across pages crawled all the way to the footer (partial crawls are held back, not guessed). "Typically sits" = the average of the section's position within the page body (0 = top, 1 = bottom). Scoped to the pages we've crawled, not "the web"; auto-updating as the corpus grows.