The Retention Gap: Why Two in Three Cannabis Customers Never Return

Methodology

This analysis draws on transaction-level point-of-sale data contributed directly by partner retailers through Headset's retail intelligence platform, covering 9 U.S. states and 1,060 retail locations. Direct sourcing from partner POS systems produces a high degree of accuracy, though some human error in data entry is possible.

To measure retention, we used a cohorted approach anchoring each customer to their first visit, a 180-day purchase window, and a 180-day return window giving every customer an equal six month opportunity to return. Customers who visited once and did not return within that window are classified as one-timers. All value figures reflect spending within the 180-day observation period, not projected lifetime estimates. Retention measures loyalty to a specific store, not to cannabis retail generally.

Over Half of Customers Never Return

Across every market in this analysis, the majority of a retailer's customers will never return. Within six months of their first visit, 67.1% of cannabis shoppers do not come back. That single fact defines the central commercial challenge facing cannabis operators today.

The customer value gap between these two groups is substantial. Within the same six-month observation window, a one-timer generates $56 in total sales. The average repeat customer generates $316. That 5.6x difference is not about who they are or how much they spend per trip. Basket size differs by only 7% ($56 vs. $60) and average item price is nearly identical. The entire gap is frequency. The one-timer is not a worse shopper. They are a shopper that chose not to return.

The more damaging cost is hidden: acquisition. Every new customer requires marketing spend to walk through the door. This includes Weedmaps placement, promotions, and local visibility. For a repeat customer, that cost is paid once and absorbed across multiple visits. For a one-timer, the full acquisition cost lands against a single $56 transaction. At a 67.1% one-timer rate, most operators are running a constant replacement cycle, continuously paying to acquire customers they will never see again. Retention does not just build revenue. It makes the revenue you already have profitable.

One important thing this data does not show: a meaningful demographic difference between the two groups. One-timers and repeat customers are nearly identical across age and gender. Millennials are 42.5% of one-timers and 41.2% of repeat customers. Gen Z over-indexes only slightly as repeat customers (0.98x). Females over-index modestly as one-timers (1.03x). Neither gap comes close to explaining a 67.1% one-timer rate. Operators cannot solve this by acquiring different customers. The customer walking out the door for the last time looks almost identical to the one who comes back. What differs is what happened during and after the visit.

Where Retention Breaks Down by Market

The national one-timer rate of 67.1% is a useful headline, but it obscures meaningful variation across states. One-timer rates span 18 percentage points in this dataset, from 58.4% in Michigan to 76.4% in New York and Massachusetts. That range reflects a mix of market maturity, competitive density, regulatory history, and operator execution. Understanding which category your market falls into determines which retention levers are most likely to work.

High-loss, high-upside markets (NY, MA). New York and Massachusetts share the worst one-timer rate in the dataset at 76.4%, and both have only recently gained access to loyalty programs and promotional tools that other markets have had for years. Customers in both states are responsive to promotional signals precisely because they are unaccustomed to them. The basket gap in NY (21.7%) means retained customers also spend meaningfully more per visit, amplifying the revenue at stake per returning customer. The case for investing in loyalty infrastructure now, while that responsiveness is at its peak, is stronger here than anywhere else in the dataset.

Mature, retention-neutral markets (CA, MO, MI). Basket size is flat or negative between one-timers and repeat customers in all three states. Returning customers visit more often but do not spend more per trip. California's basket gap has turned negative (-3.5%), meaning one-timers actually spend slightly more per visit than repeat customers, reinforcing that retention value here is entirely frequency-based. Michigan at $8.91 average item price has reached commodity pricing, where promotional signals carry almost no additional weight, reflected in its near-flat discount ratio (0.95x). Discount-led retention strategies are unlikely to move the needle in these markets.

Middle markets (IL, MD, NJ, OH). One-timer rates between 60-69% with basket gaps in the 6-11% range. Ohio ($31.99 avg item price) and Illinois ($27.87) are expensive markets where heavier discounting correlates with stronger retention, consistent with price correction rather than deal-chasing. New Jersey and Maryland show moderate discount sensitivity with promotions working alongside store experience rather than independently.

This Is Not a Market Problem. It Is an Operator Problem.

The market variation is real, but it does not explain the full picture. The retailer variance data makes that clear.

Across 1,060 retailers in this dataset, retention rates range from single digits to nearly 85%. That spread reflects a genuine performance gap between operators who have built the conditions for customers to return and operators who have not. The industry average is not a ceiling. It is the average of wildly different execution quality.

The top 20 retailers by retention retain 68.7% of customers within the six-month window. The bottom 20 retain just 5%. Top retailer customers spend 37% more per basket ($67 vs. $49), shop across twice as many categories (2.59 vs. 1.26), and generate more than 5x the customer value over the same period ($340 vs. $63). The bottom retention retailers cluster heavily in NY and MA, the two states with the worst one-timer rates in the dataset. That concentration reflects both market difficulty and execution gaps compounding each other. But stores in those same difficult markets are also among the top performers. Market conditions set the context. Operator execution determines the outcome.

Retention rates are also heavily shaped by geography before any operational factors come into play. Stores with the highest retention tend to be anchored in stable residential communities: suburban neighborhoods, small towns, and rural markets with limited competition where customers live nearby, have few alternatives, and in many cases have ongoing needs. Stores with the lowest retention are often structurally exposed to transient customer bases, sitting in tourist destinations, seasonal markets, state border corridors, and high-traffic urban areas where a significant share of foot traffic has no realistic intention of returning. A meaningful portion of what looks like a retention problem at the store level is a customer mix problem driven by location. A tourist-area store in New York and a residential suburban store in Michigan are not competing for the same customer, and comparing their retention rates without that context produces misleading conclusions. Until you account for the type of customer a store is realistically able to serve, retention comparisons across the network will always be skewed by geography.

What Retention Is Actually Worth

Most retention thinking in cannabis focuses on the same conversion: getting a one-timer back for a second visit. That is a valuable problem to solve. But the customer value data shows it is only the beginning.

Repeat customers are not a monolith. They are a spectrum, and the majority sit at the shallow end. Within the six-month observation window, over half of all repeat customers (55.7%) fall into the Occasional tier, visiting just 2-3 times with roughly two months between trips. At the other end, fewer than 5% reach the Highly Engaged tier, visiting weekly and generating nearly 10x the customer value of an Occasional shopper over the same period.

Basket size is nearly flat from Occasional through Loyal, ranging from $61 down to $56. It compresses further at the Highly Engaged tier ($45), where customers are making frequent, smaller restocking trips rather than considered purchases. All of the value progression from $142 to $1,353 is a frequency effect. Customers who visit more often are not spending more per trip. They are simply choosing your store more consistently.

The gap between tiers is significant. An Occasional customer generates $142 within the six-month window. A Regular customer generates $488, more than 3x the value with no change in basket size. The difference is not what they spend per trip. It is how consistently they choose your store. Customers who visit more frequently are already making that choice. The question for a retailer is whether their store is the one those customers keep coming back to.

What Drives Retention

Three findings from this dataset consistently predict whether a customer comes back, and each one has direct implications for how a store operates.

Category Signals Who Is Likely to Return

The category a customer purchases on their first visit is one of the strongest predictors of whether they return within six months. Concentrate and Vapor Pen buyers return at more than 2x the rate of Topical buyers, and meaningfully above Pre-Roll customers.

This pattern reflects the nature of each product. A customer who purchases a Vapor Pen device has an ongoing need for cartridges, so the hardware creates a natural reason to return. Concentrate buyers tend to be engaged, higher-involvement customers. Pre-Roll and Topical buyers are more likely visiting for a specific, one-time occasion.

This does not mean Pre-Roll or Edible customers cannot become regulars. It means that Concentrate, Vapor Pen, and Flower customers arrive already showing the signals of a returning shopper. For these customers, quality of experience matters most: knowledgeable staff, awareness of what is new, and a reason to choose your store again. Getting those interactions right on the first visit is the highest-leverage retention opportunity a retailer has.

Category Breadth Reflects Engagement

Customers who shop across multiple categories visit more often and generate more customer value. One-timers average 1.4 distinct categories. Repeat customers average 2.3. Customers in the Highly Engaged tier average 3.2 across the six month analysis period.

This is not a prescription to push customers toward more products. It is a signal about who your most engaged customers already are. A customer who shops across Flower, Vapor Pens, and Concentrates is telling you something about how they relate to cannabis retail. They are a broad explorer, and broad explorers are your most valuable and most loyal audience. Making sure they have an excellent experience across every category they shop, and that your staff is knowledgeable about all of them, is how you earn their continued preference.

The First-Visit Discount Effect

One of the more important tools retailers can leverage to engage customers is discounting. When examining first time offers, customers who received any discount on their first visit retained at 36.8% within six months. Those who received no first-visit discount retained at 26.9%. That 10 percentage point lift is observable at the individual customer level and holds across nearly every state in the dataset.

The size of the first-visit discount has almost no effect on whether a customer returns. Retention is nearly flat from 1-10% through 50%+, ranging from 35.6% to 38.2%. But discount size has a significant effect on what the customer is worth when they do return. The 1-10% cohort generates $193 in 180-day value versus $109 for the 50%+ cohort. A deep welcome discount attracts a customer at roughly the same return probability but 43% lower value.

The Sweet Spot Is Light Engagement, Not Heavy Promotion

In a separate analysis evaluating average customer discount across their six month lifespan, we found that Retailers who build no promotional engagement into their customer experience are strongly associated with poor retention outcomes. Nationally, the no-discount cohort has an 85% one-timer rate, the worst of any group in this analysis. What is clear is that stores with active promotional programs retain customers at dramatically higher rates than those without.

Light discounting (1-10%) produces the strongest outcomes across every metric. Beyond that threshold, retention and customer value decline with each additional tier. Two factors likely explain why heavy discounting underperforms. First, deep promotions tend to attract price-motivated shoppers who follow the best deal rather than building a relationship with a specific store. Second, heavy discounting is often a symptom of store distress, where poor retention and heavy discounting are both symptoms of the same underlying problem.

Discount Behavior Varies Significantly by Market

The national pattern masks meaningfully different dynamics across states. Three cohorts emerge.

Light discount markets (CA, MO, NJ) follow the national pattern cleanly. The 1-10% sweet spot holds with steady declining returns at higher tiers. California is the flattest market in the dataset on first-visit discounting, with retention barely moving regardless of discount size (34.8% to 35.9%). Promotional signals carry minimal weight in this mature, price-competitive environment.

High-price markets (OH, IL) show their highest customer value at the 30%+ tier. Ohio's average item price of $31.99 is the highest in the dataset, and its first-visit discount responsiveness reflects this: retention jumps from 31% at no discount to 50% at 50%+. In markets where cannabis is structurally expensive, deeper discounts function as price correction rather than deal-chasing signals.

Regulatory transition markets (NY, MA) have only recently been permitted to offer loyalty programs, coupons, and bundled promotions. Both show strong retention at light discounting (NY 67.8%, MA 61.3%) because customers are unaccustomed to promotional signals and respond strongly to modest ones. NY shows diminishing returns above 10% on the first visit (32.8% at 1-10%, dropping to 25.5% at 50%+), consistent with deep discounts attracting price-motivated customers in a market still building its loyalty infrastructure.

Price-compressed markets (MI, MD) show flat or mixed discount sensitivity. Michigan at $8.91 average item price has reached commodity pricing, where promotional signals carry almost no additional weight. Michigan also has the highest no-discount retention baseline in the dataset at 37.8%, suggesting store experience is the dominant driver of return behavior there.

The Practical Takeaway

Build promotional engagement into your operation regardless of market. The no-discount cohort underperforms in every state. Offer a modest welcome discount on visit one, keep it in the 1-10% range, and do not depend on ongoing deep discounting to maintain the relationship. The data shows the welcome signal matters. The depth of the discount does not. And never build your store's primary value proposition around price. Customers who return only because you are the cheapest option are the most fragile relationship in your base.

What Retailers Can Do

The data points to four concrete actions. Each is grounded directly in a finding from this analysis.

Show up after visit one. The no-discount cohort has an 85% one-timer rate. A welcome offer, a loyalty program enrollment, or any modest signal that the store noticed the visit is the single highest-leverage retention action available. Keep it light. The first-visit discount data shows that a 1-10% welcome offer generates $193 in 180-day customer value compared to $109 for a 50%+ discount. The signal matters. The depth does not.

Know which customers are most likely to return, and make sure their experience reflects that. Concentrate, Vapor Pen, and Flower customers are statistically your most likely returning shoppers, returning at 37.4%, 36.8%, and 33.7% respectively within six months. These customers already show the behavioral signals of someone who will choose a store again. Quality of experience, staff knowledge, and product availability are what determine whether they choose yours.

Earn the return visit at every tier. Over half of repeat customers (55.7%) sit in the Occasional tier, visiting every two months. These customers are already choosing to return. Making sure they have a consistent, high-quality experience, that staff knows them, that product is available, and that the store feels worth coming back to, is how you become their preferred destination rather than just one option among several.

Benchmark against the right market. A store in New York operates in a fundamentally different environment than one in Michigan. NY and MA retailers are only now building the promotional infrastructure other markets have had for years, and their one-timer rates of 76.4% reflect that transition as much as execution quality. Michigan at 58.4% is operating in a commodity-priced market where store experience is the primary retention driver. Retention performance should be evaluated against market-specific comparables. Multi-state operators should resist applying a single retention target across markets with structurally different dynamics.

‍