A dispensary’s point-of-sale system keeps the most complete record in the building. It knows the SKU, price, discount, time of day, the budtender who rang it up, and, if the customer is enrolled in loyalty, who they are and how often they return. For inventory, reorders, promotions, and compliance reporting, that record is exactly what an operator needs.
It also ends when the repeat-purchase decision begins.
Whether a customer comes back is not settled at the counter. It is settled hours later, at home, when the product either does what the customer bought it for or does not. The receipt captures the transaction in detail but nothing about the outcome. Most customer retention strategies rely on the first record and guess at the second.
What the POS Actually Knows
Cannabis POS and loyalty platforms are good at recency, frequency, and spend. They can tell you who has not been in for 30 days, which categories a customer buys, and what basket size looks like by daypart. Layer in a loyalty program and you can attach points, tiers and targeted offers to all of it.
Every one of those signals describes behavior. None of them describes experience. A repeat purchase is an outcome, and the POS records the outcome without its cause. That distinction sounds academic until you look at two customers with identical receipts.
Two Identical Receipts
Consider two customers who buy the same product on the same Friday afternoon, at the same price, with the same loyalty discount. In the POS, they are the same row.
The first bought it to unwind after work, and it did exactly that. She will be back next week and will probably ask for it by name. The second bought it for the same reason and spent the evening restless and unsettled. He will not complain, ask for a refund, or leave a review. He will quietly try another store next time, or come back to yours and pick something different with no record of what went wrong.
Nothing in the transaction data distinguishes them. A recency report will flag the second customer only after he has already gone, and by then the reason is unrecoverable.
Why the Label Does Not Close the Gap
The obvious objection is that product attributes should predict experience. If a store knows what it sold, it knows roughly what the customer felt.
The research suggests otherwise. A 2022 analysis of commercial cannabis across six U.S. states found that the common labels attached to products, including indica, sativa, and hybrid, do not consistently align with the products’ measured chemical profiles.1 Potency is not a reliable proxy either. In a University of Colorado study, participants assigned to higher- or lower-THC flower and concentrates showed very different levels of THC in their blood, yet the researchers concluded that differences in short-term subjective effects “did not track specifically with strength of the cannabis consumed.”2
Add format and the variability widens. An edible, a vape cartridge, a tincture, and a pre-roll can produce different onsets, durations, and experiences for the same person. The label describes the product. It does not describe what happened to the customer, and it was never designed to.
A Structural Gap, Not a Missing Feature
The temptation is to treat this as a software problem: add a survey, add a rating field, buy another dashboard. But the gap is structural. The experience happens off premises, and after the sale, in a window no retail system observes. It varies by person, product, and format, so you can’t infer it from the SKU. And the only bridge most stores have, the budtender who remembers what a regular said last time, is the least durable asset in the business. When that employee leaves, the store loses the knowledge of why its best customers come back.
That is the real cost. Operators do not lack data. Their data answers questions they don’t need answered and says nothing about the one that drives retention.
What Operators Can Do Now
None of the following requires new software, and most of it can start this week.
Ask the after-sale question on purpose. When a known customer returns, the most valuable thing a budtender can ask is how the last purchase worked for them. Make it a standard part of the second visit, not an improvisation.
Record the answer in a structured way. A free-text note that says “liked it” is nearly useless at scale. Three fields are enough: what the customer wanted from the product (sleep, relief, focus, socializing, unwinding), whether it delivered (yes, partly, no), and the format they used. Most POS and CRM systems already have a profile notes field for this.
Recommend in outcome language. If the counter conversation starts with how the customer wants to feel rather than with THC percentage or strain type, the follow-up question has something to measure against. “Did it help you unwind?” is answerable. “Did you like the 28 percent?” is not.
Read basket changes as a signal. A customer who switches categories or brands every visit may be exploring. They may also be searching for something that works but can’t find it. Without the experience data those two customers look the same, and only one of them is at risk.
Make the knowledge outlive the employee. Whatever staff learn about why a customer returns belongs in the system, not in one person’s memory. Turnover is a fact of cannabis retail. Losing your retention intelligence every time someone quits does not have to be.
Where state rules permit follow-up outreach, build it around the experience, not the discount. A message that asks how a product worked earns information. A coupon earns a transaction and teaches you nothing.
The Second Question
The receipt will always be the best record of what sold. It will never tell you why someone came back. Operators who build that second record, one structured conversation at a time, will understand their customers in ways transaction data alone cannot, and they will see churn coming while there is still time to act.
The post What Your Dispensary’s POS Data Doesn’t Tell You About Your Customers appeared first on Cannabis Industry Journal.
