How Do You Measure Dispensary Customer Acquisition Without Reliable Ad Platform Tracking?
If you market a dispensary, the platforms may show you clicks, views, and conversions without answering the question the owner actually cares about: did this campaign bring in a new customer who bought something?
You won't attribute every sale perfectly. You can still build a decision-quality estimate by defining a new customer, controlling campaign identifiers, and connecting several imperfect signals to first-purchase data.
Define a new customer before calculating cost
You should choose a consistent new-customer rule before calculating acquisition cost, because otherwise one campaign may count loyalty signups while another counts first purchases, and the lower reported cost will reflect the definition rather than better marketing.
A practical definition is a person making a first completed purchase at any location owned by the business during a stated lookback period.
Decide how to treat anonymous cash buyers, merged profiles, delivery customers, returning customers after a long absence, and people who first shopped at another location. Document the rule before comparing campaigns.
Do not call every new loyalty signup a customer. A signup becomes acquisition only when it connects to the business outcome the campaign was meant to produce.
Include the full acquisition cost
For a campaign or channel, add:
- media and placement fees;
- creative and production;
- agency or contractor cost;
- landing-page and tracking cost;
- discounts used only to acquire the customer;
- event or sponsorship cost; and
- staff time that would not otherwise have been spent.
Divide that amount by the number of qualified new customers reasonably attributed to the effort.
`customer acquisition cost = acquisition spend / attributed new customers`
Keep fixed brand costs separate when they cannot be assigned fairly. Consistency matters more than forcing every overhead dollar into one campaign.
Build campaign identifiers you control
Use dedicated landing pages, tagged links, QR destinations, campaign codes, permitted offer codes, event registration fields, and optional point-of-sale source questions. Name campaigns consistently across the website, customer platform, point of sale, and reporting sheet.
External advertising links should use clear campaign parameters without placing sensitive customer information in the URL. Never use a person's name, phone number, email address, purchase, or health-related detail as a tracking value.
Platform pixels can be useful where they are allowed, but they should not be the only record. Browser limits, consent choices, device changes, cash purchases, and policy restrictions can break the path between exposure and sale.
Connect the first purchase carefully
Where lawful and supported by the point-of-sale system, record the first completed transaction date, location, pretax revenue, discount, gross profit, and declared or observed campaign source.
Use the minimum personal data needed and restrict access to it. Cannabis purchase information can be sensitive even when a particular privacy law does not assign it a special category.
Do not upload raw customer lists to an advertising platform merely because the interface allows it. Review platform terms, customer consent, applicable privacy requirements, retention, hashing, and whether the business can honor deletion and opt-out requests.
Use an attribution ladder
Rank evidence by strength instead of pretending all signals are equal.
Strong direct evidence can include a single-use campaign code tied to a first purchase. Moderate evidence may include a campaign landing-page visit followed by a consented signup and later purchase.
Weaker evidence includes a customer recalling the source, a rise in branded search, or a store-sales increase during a campaign. Those signals still matter, but they should not be reported as exact one-to-one attribution.
Create a time window appropriate to the action. A directions campaign may influence a visit within days, while awareness from a local sponsorship may take longer and appear through several channels.
Compare acquisition cost with contribution, not revenue
A $100 first purchase is not $100 available to recover marketing cost. Subtract product cost, transaction-related expenses, discounts, and other variable delivery costs to estimate contribution margin.
Then track whether acquired customers return. A campaign with a higher initial acquisition cost can be better when it produces customers who make profitable repeat purchases.
Useful cohort measures include:
- first-purchase contribution;
- percentage making a second purchase;
- days to second purchase;
- contribution over 30, 60, or 90 days; and
- opt-out, refund, or complaint rates.
Do not project lifetime value from a few weeks of optimistic data. Use observed cohorts and label forecasts as forecasts.
Establish a baseline before claiming lift
Compare the campaign period with a relevant baseline by location, day of week, and season. Account for holidays, pay cycles, store hours, inventory availability, price changes, weather, events, and competing promotions.
When possible, stagger a campaign across comparable locations or audiences. A holdout is rarely perfect in local retail, but it can help separate normal movement from campaign lift.
Record what else changed. Attribution becomes less credible when the business launches a new menu, changes hours, runs three promotions, and credits the entire sales increase to one ad.
Make the report honest enough to use
Report confirmed, probable, and unattributed new customers separately. Show spend, first-purchase contribution, repeat behavior, and the assumptions used.
A channel should not be scaled because it produced cheap clicks. Scale when the evidence suggests it produces lawful, qualified customers whose contribution and retention justify the acquisition cost.
Good measurement does not eliminate uncertainty. It makes the uncertainty visible enough that the next marketing dollar can be placed more intelligently.