Data Story

The Headline Ledger

What do stores promise first? A letterpress index of H1/H2 phrases across captured sites.

cannabiscanadawebcontentheadingslanguage

Dataset scope

1918

captures

800

phrases

H1/H2

only

TF-IDF

later

Headings are often templated. This is a “promise index,” not a truth or sentiment classifier.

Loading headline ledger data

Red-team note

Word clouds lie when they ignore denominators. This view emphasizes capture counts and distinct phrases, and it fades low-N slices.

Browse more cannabis stories in the collection index.