Inverted Pyramid: Newsroom Reader-Retention Simulator

Story Facts — Drag Priority

Newsroom Conditions

Lede / hook strength 40

Prose complexity (reader fatigue) 55

Readers simulated 150

Scroll duration (s) 5.0

—

Results

Avg read depth

Completion rate

Newsworthy information delivered

How it works

Real audience-analytics data (Chartbeat, Parse.ly) shows online readers abandon an article at a rate that rises the longer and denser it gets — a "hazard" of leaving, not a fixed reading time. This sim models each simulated reader's exit point y ∈ [0,1] (fraction of the story scrolled) with a survival function built from that hazard:

hazard  h(y) = k · (1 + f·y)
survival S(y) = exp( -k·(y + f·y²/2) )
k = k0 · (1 − hook/150)     f = complexity/50

Each reader draws u ~ Uniform(0,1) and exits at the depth y solving S(y) = u — an inverse-transform sample from the hazard model, so a stronger lede (lower k) or simpler prose (lower f) keeps more readers scrolling deeper, exactly as newsroom analytics predict.

The article itself is six facts about a single story, each with a fixed newsworthiness value. Reordering them changes nothing about why readers leave — only which facts they see before they do. The "inverted pyramid" — the oldest rule in journalism — says: front-load the highest-value facts, because attrition is guaranteed and unrecoverable. The sim's "information delivered" score is exactly the expected newsworthy value reaching the average reader, given the retention curve above:

Reorder facts — rebuild the article's structure; the 3D column re-shapes into (or out of) a true pyramid as slab widths track each fact's value.
Hook / complexity sliders — change how fast the simulated audience falls away.
Readers / duration — sample size and playback speed of the scroll animation.

Still reading Dropped off Finished story

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.