Sooper Works · IP Lab. Concept Leaderboard

Data window 20 Jun – 2 Jul 2026 · web_events · refreshed 2 Jul 14:00

Mode A — historical bands. Hold, Repeat and Return graded on existing traffic; Hook shown as a modelled, position-corrected index (provisional). Clean-rig certificates arrive with Mode B tests. Tiers are computed within band (never across).

Concepts graded
56+38 maturing
Library Hold base
43%
Story Return base (7d)
10.3%
Repeat base (same-visit)
2.9%
Site return, new visitors
22.8%

In flight — Mode B test slot

What a live test looks like. No Mode B test is running yet — this is the worked example.
Captain Mould
EXAMPLE · MODE B PREVIEW
In test · day 3 of 7 · certificate freezes 6 Jul · Return matures 13 Jul
4,120 mobile slot views so far · next checkpoint: day-7 read
Hook
13.8% ±1.1
Starters
569
Hold (provisional)
51% ±4
Repeat
6.1%
Return
matures day 10–14

Hold × Return — the franchise map

Dot size = unique starters. Dotted crosslines: length-adjusted hold par and story Return base. Top-right = completes and comes back. Click a dot for the report card.

Hook × Hold — promise vs product

Hook = modelled, position-corrected index (1.00× = par; provisional — read as above/at/below par). Bottom-right = great story, weak front door (re-package it). Top-left = strong front door, weak story (shell risk). Top-right = the golden quadrant.

Leaderboard

Click a column to sort · click a row for the report card. Hook stays empty until a concept passes through the controlled slot.
Concept Band Stories Length Starters Hook (modelled) Hold Hold vs length Repeat Return 7d Tier Status

Reading this board: Hold = unique starters who finish. Repeat = completed sessions re-read immediately ("again!"). Return = came back within 7 days to the exact story they first met — the depth-neutral franchise signal; inventory can't flatter it, so a 1-story concept and a 40-episode series compare fairly. Series additionally return via other episodes (the catalogue-depth dividend) — shown in brackets and on report cards, never graded. All ratios per unique visitor; ± figures are 95% Wilson intervals; tiers are z-scores within band (Hook is displayed, not tiered; Hold enters tiers length-adjusted). Hold vs length: short stories complete more easily, so raw Hold flatters them. Expected Hold is fitted against per-story length across the library (≈67% at 1 min → 42% at 15 min); the index is actual ÷ expected — 1.00× = par for a story of that length. Both charts plot the length-adjusted index. Hook (modelled): click-through vs the position-expected average, fitted jointly across 5 listing pages (same story at different positions identifies the position curve; new visitors, first session only). 1.00× = par for its slot. Read it as above-par / par / below-par — robustness checks put mid-table model uncertainty at ~±0.3, and series tiles pinned at fixed top positions are only weakly identified. Famous titles carry name-recognition a new concept won't have. It is a shortlist generator; the Mode B slot issues the real grade. Certificates (frozen, comparable grades incl. Hook) require the Mode B controlled slot — spec in statistical-model.md.