Data Story

Stack & Stars

Do storefront tech choices rhyme with public sentiment? Compare reviews and ratings across inferred commerce stacks.

cannabiscanadawebbuiltwithtech-stackreviewsgoogle
Dataset scope
3914
stores
790
city keys
6,997
max reviews
4.60
median rating
Stack labels are heuristic (BuiltWith + capture fingerprints). Treat patterns as observational signals, not causal claims.
Loading stack and stars data
Hypothesis

Commerce-native stacks are associated with higher review volume and slightly higher ratings within the same cities.

Question: Are Google review counts and ratings associated with the inferred commerce stack?

Method: Infer stack labels from BuiltWith + capture tech + network fingerprints, then compare review distributions by stack with within-city comparisons.

Prediction: Median review counts differ by stack in high-N cities, not just in national aggregates.

Test: Compare medians and deciles by stack; repeat within top cities to reduce aggregation bias.

Narrative Arc
Act I

A shelf of storefronts appears: each platform becomes a section.

Act II

Jars fill with starlight: review volume becomes glow; ratings lift the shelf line upward.

Act III

We zoom into cities to test whether differences persist when geography is held constant.

Datasets
  • cannabis_stores_canada.sqlite
  • 11_stack_and_stars_tech_vs_reviews.json
Limitations
  • Reviews are a public attention proxy, not revenue.
  • Stacks are inferred; BuiltWith/capture data can miss or mislabel vendors.
  • Geography confounds most aggregates — prefer within-city views.
Next

Browse more cannabis stories in the collection index.

Back to indexarrow_forward