Ask Headset: User Guide

Emily Walden

Jun 15, 2026

406

Ask Headset is an AI-powered assistant that helps you explore Headset Insights cannabis market data through natural language questions. Instead of building reports manually, you can simply ask questions like "What were the top edible brands in California last month?" and receive instant insights.

This guide will help you get the most out of Ask Headset by explaining how it works, what data is available, and how to write effective questions.

What Can Ask Headset Do?

Ask Headset can answer questions about:

Sales Performance: Revenue, units sold, and pricing trends for brands, products, and categories
Market Analysis: Compare performance across states, track market growth, and identify trends
Category Insights: Understand how different product categories (Flower, Edibles, Vapor Pens, etc.) perform
Brand Comparisons: See how brands stack up against competitors
Brand Loyalty & Cross-Shop: Measure what share of a brand's customers stay loyal, cross-shop with other brands, or switch away; identify the top other brands its customers also buy
Demographic Analysis: Learn which customer segments (age groups, genders) prefer certain categories or brands
Pricing & Discounts: Analyze average item prices and discount rates across markets
Basket Analysis: Average basket size, items per basket, basket size distributions, basket penetration by category, and cross-category attachment rates

Important: How Ask Headset Handles Markets

Ask Headset has access to all Headset Insights markets

Regardless of which markets are included in your Headset Insights subscription, Ask Headset can query data from all 20 available markets:

United States (16 markets): Arizona, California, Colorado, Connecticut, Florida*, Illinois, Maryland, Massachusetts, Michigan, Missouri, Nevada, New Jersey, New York, Ohio, Oregon, Washington

Canada (4 markets): Alberta, British Columbia, Ontario, Saskatchewan

Florida:* category-level sales data only

Ask Headset defaults to multi-market analysis

When you ask a question without specifying a market, Ask Headset will analyze data across all applicable markets - not just the ones in your Headset Insights subscription.

For example, if you ask "What are the top vapor pen brands?", Ask Headset will return results across all US markets combined, which may include markets outside your subscription.

Always specifiy your market of interest

To get results relevant to your business, include the state or province in your question:

❌ "What are the top edible brands?" (returns combined results from all markets)
✅ "What are the top edible brands in California?" (returns California-specific results)

Default Assumptions

When you ask a question, Ask Headset makes the following assumptions unless you specify otherwise:

If You Don't Specify...	Ask Headset Assumes...
Time period	Trailing 12 complete months
Country	United States (USD)
Number of results for "top" lists	15 items

Currency Handling

Ask Headset keeps US (USD) and Canadian (CAD) data separate. When you say "nationally" without specifying a country, Ask Headset defaults to US markets only and will note this in the response.

To analyze Canadian markets, be explicit: "What are the top brands in Canada?" or "How is the Ontario market performing?"

Data Availability & Limitations

Time Coverage

Data refresh: Daily (data is current through yesterday)
Historical depth: Up to 7 years, depending on the market

Brand Demographic Data

Ask Headset can tell you which customer demographics (age groups, genders) purchase specific brands. However, this data has some limitations:

Time window: Brand demographic data covers the trailing 180 days only (not the full historical period)
Minimum sample size: Only brands with at least 1,000 unique customers in a state/category appear in demographic results
Percentages only: For demographic analysis, Ask Headset reports percentage shares and index values - not raw dollar sales figures
Data confidence: Results include a confidence indicator (high, medium, low) based on sample size. Results marked "low" confidence should be considered directional

If Ask Headset returns no demographic data for a brand you're interested in, it likely means that brand has fewer than 1,000 customers in the Headset retailer sample for that market during the past 180 days.

Brand Loyalty Data

Ask Headset can analyze customer retention, repeat purchase behavior, and cross-shopping patterns for individual brands, or report category-wide averages. A few things to know:

Two time windows: Loyalty is measured by comparing two trailing windows. The Prior window (365–183 days ago) defines the customer cohort; the Current window (last 183 days) is where repeat and cross-shop behavior is measured. Both windows are anchored to yesterday and cannot be changed.
Three retention buckets: Every cohort customer falls into exactly one bucket — loyal-only (bought only this brand in the Current window), retained cross-shopper (bought this brand and others), or brand-switcher (returned to the category but did not repurchase this brand). The three together sum to ~100%.
Cross-shop detail: For any brand, Ask Headset can also list the top other brands its customers bought in the Current window — framed as "X% of the brand's customers also bought Y," not as competitors stealing share.
Demographic slicing built in: Loyalty can be sliced by age group and gender (e.g., "Is Wyld more loyal among Gen Z or Boomers?") without leaving this dataset.
Category baselines: You can also ask for the average loyalty profile across all qualifying brands in a category/state (e.g., "What's the average retention rate for CA vapor pen brands?").
Minimum sample size: Only brands with at least 1,000 unique customers in the Prior window appear. Below that threshold, Ask Headset will say so rather than guess.
Silent Generation excluded from age-group breakdowns due to small sample sizes.
Customer counts are sample, not market sizing: The customer numbers come from Headset's retailer sample. Use them for proportions and rankings, not as a count of every cannabis shopper in the state.

If Ask Headset returns no loyalty data for a brand, it almost always means the brand had fewer than 1,000 customers in the Prior window for that state/category.

Basket Analysis Data

Ask Headset can analyze basket-level shopping behavior: basket size, what's in the basket, and which categories get bought together. A few things to know:

Current month access: Basket data is current through yesterday. If you ask about the current month, Ask Headset will note that the month isn't complete.
Two question types:
- Market-level basket metrics: average basket size, items per basket, basket size distribution by $10 bucket, how discounting affects basket size
- Category-level basket metrics: basket penetration, basket composition, cross-category attachment rates
Time window: Basket data covers up to 15 months of history (shorter than the 7-year window for sales data).
Florida excluded: Basket data is not available for Florida. For Florida questions, Ask Headset will fall back to category-level sales data.

What Ask Headset Cannot Do

Access data from markets not listed above
Provide retailer-specific or store-level data (this requires Headset Bridge)
Make future predictions or forecasts
Combine USD and CAD sales into a single total (these are kept separate)
Attribute customer movement causally (e.g., "Brand X stole customers from Brand Y") — loyalty data describes what customers bought, not why they switched

Prompting Tips

Be Specific About Geography

Always include the state or province you're interested in:

Instead of...	Try...
"Top flower brands"	"Top flower brands in Colorado"
"How is Wyld performing?"	"How is Wyld performing in Washington and Oregon?"
"Edible market trends"	"Edible market trends in Midwest markets over the last 6 months"

Be Specific About Time Periods

Ask Headset defaults to 12 months, but you can request different periods:

"...in Q4 2024"
"...over the last 6 months"
"...in January 2025"
"...comparing 2025 vs 2024"
“… over the trailing 60 days”

Be Specific About Categories

When analyzing brands, specify the product category for more relevant results:

Instead of...	Try...
"Stiiizy's demographic breakdown"	"Stiiizy's demographic breakdown for vapor pens in California"
"Which brands do Gen Z prefer?"	"Which edible brands do Gen Z prefer in Colorado?"

Use Clear Comparisons

When comparing, be explicit about what you're comparing:

"Compare Cookies vs. Connected flower sales in California"
"How do Millennial vs. Baby Boomer preferences differ for edibles in Illinois?"
"Compare vapor pen sales in New York vs. New Jersey"

Example Questions

Good Questions ✅

Sales & Market Analysis:

"What were the top 10 vapor pen brands in California last month?"
"How has the average price of flower trended in Colorado over the past 12 months? Compare that to the preceding 12 months."
"Compare Wyld and Wana edible sales in Washington for 2024"
"Which states had the highest discount rates last quarter?"
"What is the month-over-month sales growth for Michigan?"

Category Demographics:

"What age groups prefer edibles vs. flower in Illinois?"
"Do male or female customers spend more on concentrates in Colorado?"
"Which product categories over-index with Baby Boomers in East Coast markets?"

Brand Demographics:

"What is the demographic breakdown of Cookies flower customers in California?"
"Which vapor pen brands over-index with Generation Z in New York?"
"Compare the demographic profiles of Wyld vs. Camino edibles in California"

Brand Loyalty & Cross-Shop:

"What percentage of Stiiizy's California vapor pen customers were loyal-only, cross-shopped, or switched away over the last six months?"
"Who are Jaunty’s biggest cross-shopped brands in New York vapor pens?"
"How does Wyld's customer loyalty in Colorado edibles break down by age group?"
"What is the average customer retention rate for vapor pen brands in Michigan?"
"Compare loyalty rates for Claybourne Co. vs. CannaBiotix flower in California."

Basket Analysis:

"What is the average basket size in Washington over the last 6 months?"
"What percentage of baskets in California fall in the $20–$30 range? What are the most common discount rates for those baskets?"
"What categories are most often purchased alongside vapor pens in Colorado?"
"What's the attachment rate between pre-rolls and flower in Michigan?"
"How has basket penetration for edibles trended in Michigan over the last 12 months?"

Questions That Need Improvement ⚠️

Question

Problem

Better Version

"What are the top brands?"

No market or category specified

"What are the top edible brands in California?"

"How is the market doing?"

Too vague

"What is the year-over-year sales growth for Colorado by category?"

"Tell me about Stiiizy"

Too broad

"What were Stiiizy's vapor pen sales in California last quarter, and what is their demographic breakdown?"

"National sales trends"

Doesn't specify US or Canada

"What are the US national sales trends for vapor pens over the last 12 months?"

"Who buys edibles?"

No market specified

"What age groups and genders buy the most edibles in Michigan?"

"Who's stealing my customers?"

No brand, market, or category; causal framing

"What other brands do Stiiizy's California vapor pen customers also buy?"

Understanding Your Results

Demographic Index Values

When viewing brand demographic data, you may see "index" values. Here's how to interpret them:

Index = 100: The brand's appeal to this demographic matches the category average
Index > 100: The brand over-indexes with this demographic (stronger appeal than average)
Index < 100: The brand under-indexes with this demographic (weaker appeal than average)

For example, if a brand has a "Generation Z Index" of 130, it means Generation Z customers are 30% more likely to purchase that brand compared to the category average.

Brand Loyalty Terminology

Loyal-only: Customers who bought the brand in the Prior window (365–183 days ago) AND only this brand again in the Current window (last 183 days).
Retained cross-shopper: Customers who bought the brand in the Prior window AND bought this brand plus at least one other brand in the Current window.
Brand-switcher: Customers who bought the brand in the Prior window AND returned to the category in the Current window but did NOT repurchase this brand (they bought other brands in the category instead). Note: this does not mean a competitor "took" them — they simply didn't repeat this brand.
Cross-shop: When Ask Headset lists other brands a brand's customers also bought, it's reporting the share of the brand's Current-window customers whose baskets also contained the other brand. Use it to understand competitive overlap, not to claim causality.

Data Confidence Levels

For demographic analysis, results include a confidence level:

High: 5,000+ customers — reliable for detailed analysis
Medium: 2,500–4,999 customers — good for directional insights
Low: 1,000–2,499 customers — use with caution; consider these insights directional

Basket Terminology

Basket analysis uses a few terms worth defining:

Average basket: The average dollar value of a single transaction, after discounts and before taxes
Basket penetration: The % of all transactions that included a given category. If Vapor Pens have 35% basket penetration in California, 35% of CA transactions included at least one Vapor Pen item.
Attachment rate: The % of baskets containing Category A that also contained Category B. If the Pre-Roll → Flower attachment rate is 20%, 20% of Pre-Roll baskets also included at least one Flower product.
Basket composition: How a basket's dollar value breaks down between the focal category and other items. A high "% of basket to category" means shoppers buy that category as the main event; a low value means it's mostly an add-on.

Providing Feedback

Your feedback helps improve Ask Headset. After each response, you'll see thumbs up 👍 and thumbs down 👎 buttons.

👍 Thumbs up: The response was helpful and accurate
👎 Thumbs down: The response was unhelpful, inaccurate, or unexpected

When providing a thumbs down, please include written feedback describing what went wrong or what you expected. This helps our team identify and fix issues.

Quick Reference

Topic	Default	How to Change
Time period	Trailing 12 months	Specify: "last 6 months", "in Q4 2024", "YTD"
Country	United States (USD)	Specify: "in Canada", "in Ontario", "Canadian markets"
Market scope	All US markets combined	Specify state: "in California", "in Colorado"
Results count	15 for "top" lists	Specify: "top 5", "top 25"
Demographics time window	180 days (fixed)	Cannot be changed
Basket data history	15 months (fixed)	Cannot be changed; FL excluded
Brand loyalty windows	Prior 365–183 days ago; Current last 183 days	Cannot be changed
Brand loyalty minimum sample	1,000 customers in Prior window	Cannot be changed

Summary: Keys to Success

Always specify your market — Include the state or province in every question
Be specific about time periods — Don't rely on defaults if you need a specific range
Include product category — Especially for brand and demographic questions
Remember the data boundaries — Brand demographics are 180-day, percentage-based, and require 1,000+ customers; brand loyalty uses 365–183 vs last 183-day windows with the same 1,000-customer minimum
Use feedback buttons — Help us improve by rating responses and providing comments

If you have questions not covered in this guide, please contact Headset Support.