Methodology
No number ships without a source row.
Every figure the Record publishes traces to a scan, a date, a sample size, and this page. Here is how the panel is built, what a real browser load counts, and where the estimates end.
- 1
The panel
Licensed-looking dispensary storefronts with their own website, found through public business listings by city and state. Marketplace menus, social profiles, and link shorteners are excluded. Chains count once, with a location count.
- 2
The load
Each homepage is loaded once in a real browser as a first-time visitor, with no cookies and no prior answer. Tags and outside hosts are observed on the wire, not guessed from the HTML.
- 3
Fires before consent
A tag counts as firing before consent when the browser saw its request before any consent answer was given. A visible age gate is not a consent answer. A posted privacy policy is not a consent answer.
- 4
The policy
The privacy policy linked from the homepage is read for its date, length, and whether it mentions pixels, vendors, opt-out, and sharing. A policy older than two years is flagged.
- 5
The traffic
Monthly visits are modeled by third-party estimators, labeled with the month they cover. They are estimates, never measured analytics, and every figure built on them says so.
- 6
The score
Consent, capture, trust, and action are scored 0 to 100 and combined. The banner that asks first takes every failing subject to an A, which is the point of the exercise.
- 7
What is published
Aggregates by state and platform only. Store-level results stay private unless the store opts in to a public Site Inspection. Every published stat carries the sample size, the scan date, and this method.
What the current set shows.
321 dispensary sites, scanned 2026-09-25. A small set is a small set: treat these as directional until the panel grows.
| Stat | Value | Sample | Date | Method |
|---|---|---|---|---|
| Median overall score across scanned dispensary sites | 31 of 100 | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Dispensary sites that show a consent banner on first view | 14 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites that fire at least one tag before a visitor can answer | 74 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites loading an ad pixel before consent | 5 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites replaying visits before anyone agrees | 14 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites with no privacy policy linked from the homepage | 40 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites whose privacy policy is older than two years | 9 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Sites that ask the age question before the menu | 77 % | 321 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Modeled monthly visitors across the scanned set who leave without a name (at 98% anonymous) | 26,901,216 visitors a month | 310 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
| Customers a year the scanned set would name at a 30% yes rate | 98,820,792 customers a year | 310 | 2026-09-25 | Each homepage loaded once in a real browser as a first-time visitor; tags counted if they fired before any consent answer; policy read from the homepage link. Traffic is modeled (DataForSEO Labs), not measured. |
Menu and site platforms seen: WordPress 54%, Dutchie 18%, Weedmaps 10%, Leafly 8%, Squarespace 5%, Wix 4%, Jane 4%, Webflow 4%, Shopify 2%, BLAZE 2%, Meadow 1%.
Late bulletin
See our future together!
Want to see what we can do together? Put in your real website address and we will analyze your traffic, review your marketing tools, and predict the final outcome of GreenLeadSource activated and collecting new customer contacts.
- A real browser load of your homepage, screenshot included
- Every tag that fires before a visitor can answer
- Your privacy policy, lined up against what the site does
- Your monthly traffic and the customers it would produce