We Live in a World of Funnels — Semiotic

Home · Getting Started · Charts · Features · Playground

This page requires JavaScript for interactive chart demos.

AI / Machine-readable docs: llms.txt · llms-full.txt · CLAUDE.md · schema.json · API Reference · Examples

View on GitHub · npx semiotic-ai for CLI access

Machine-readable page content

Canonical: https://semiotic.nteract.io/examples/world-of-funnels

We Live in a World of Funnels

Precision, accuracy, and the paths we erase

You think you are reading an essay. You are moving through a funnel. The trick is deciding when the clean shape is a useful instrument and when it starts hiding the actual system.

Selected conversion15%1,380 of 9,400

Hardest visible drop-43%Product -> Cart

Hidden flow exposed52%traffic outside the straight line

Best A/B lift+93%bigger buttons vs. original

Same case file, two charting attitudes. The first is excellent at comparison. The second admits that behavior has loops.

PreciseFive ordered stages make the checkout leak painfully clear.

AccurateSearch, support, returns, and loops are part of the product, not noise.

Candy Shop: the funnel as evidence

A simple funnel turns a process into comparable widths. It is direct, boardroom-friendly, and useful for a first pass. It is also a decision to treat all off-path behavior as loss.

Precise readingThe stage strip, leak tape, and funnel agree on the biggest narrowing.

01Arrive9,400
02Product6,700
03Cart3,820
04Checkout2,650
05Bought1,380

Product->Cart-2,88043% leak

Arrive->Product-2,70029% leak

Checkout->Bought-1,27048% leak

A/B funnels reward the intervention that moves the line

Product funnels are often used this way: compare cohorts, color the winners, ship the design that carries more people deeper. The vertical view makes drop-off legible as retained value plus lost opportunity.

Precise readingEqually effective here: cohort testing is a straight comparison, so the ordered funnel is the right instrument.

Precision says the story is shrinking

Toggle the argument, not the case file. The precise view emphasizes ordered loss. The accurate view reveals repair, rereading, searching, and exits that create design questions a single funnel cannot ask.

Precise readingThe chart compresses the case into one comparable path.

Motifs are paths with names

Teams already talk about happy paths and bad paths. A temporal process sankey makes those motifs explicit: repair loops, comparison loops, and exits all occupy time and capacity even when they are not the goal.

Precise readingThe temporal view still works, but particles pause so path volume and timing carry the comparison.

The Semiotic point

Precise readingThe conclusion treats the funnel as a sharp measuring instrument.

The danger is not that funnel charts are wrong. The danger is letting the chart's clean shape become the system's design brief. Semiotic's value here is the ability to move between precise ordinal views and more accurate flow views without leaving the same React data-visualization surface.