Measuring the Lifetime Value of Cannabis SEO Customers

Why the first order lies about SEO

Cannabis has unusually loyal buyers. Someone who finds a dispensary through search and likes the experience often comes back monthly for years, which means the first order is a small fraction of what that customer is actually worth. If you credit SEO only with that first purchase, you will systematically undervalue it and probably underfund it. The whole reason to bother with lifetime value is to stop judging a channel that acquires durable customers by the thinnest slice of their spend.

A simple estimate you can defend

You do not need a data science team. A serviceable LTV is average order value, times purchases per year, times the years a customer typically stays, times your gross margin. Say a customer spends 60 dollars an order, buys twice a month, stays two years, and you run 50 percent margin. That is 60 times 24 times 2 times 0.5, roughly 1,440 dollars of margin, against a first order that showed maybe 30 dollars. Even with loose inputs, the gap between 30 and 1,440 is the entire argument — that gap is the point. Keep the assumptions visible so the estimate can be challenged and adjusted rather than trusted blindly.

The attribution gap you cannot ignore

Here is the honest hard part. The data that says "this customer came from organic search" lives in GA4. The data that says "this customer bought twelve more times" lives in your point-of-sale, usually under a loyalty account with no thread back to that first web session. The two systems rarely talk. So you can see acquisition and you can see repeat purchasing, but connecting a specific SEO-acquired visitor to their two-year purchase history is often impossible with the tools most dispensaries run.

Working around the gap

Two practical moves help. First, a shared identifier: if you capture an email at the online signup and the same email joins the loyalty program, you have a stitch point, imperfect and privacy-limited but real. Second, and more reliable, a cohort view: instead of tracing individuals, compare the repeat behavior of customers acquired through organic search against those from other channels. If the organic cohort reorders more often or stays longer, that is a defensible LTV story even without per-person tracking. Most operators live in the cohort world — and that is fine.

A worked example

A delivery brand judged SEO purely on first-order revenue and nearly cut the budget. Before doing so, they ran a cohort comparison in their POS: customers whose loyalty signup email matched an organic-search email reordered 40 percent more often over twelve months than paid-social customers. They could not trace every individual, and they said so, but the cohort trend was enough to reframe SEO from a mediocre first-order channel into their best source of repeat buyers. The budget stayed, presented as a range with its assumptions spelled out.

Frequently asked questions

How do I estimate customer lifetime value?

A workable estimate is average order value times purchases per year times the number of years a customer stays, times your gross margin. You do not need a perfect model. Even a rough figure makes the point that a returning cannabis customer is worth several times their first order, which is what changes how you value the SEO that acquired them.

Why is LTV hard to measure in cannabis?

Because the acquisition data and the purchase data usually live in two systems that do not talk to each other. GA4 knows a visitor arrived from organic search, but the repeat purchases happen at the point-of-sale, often under a loyalty account with no link back to the original session. Bridging that gap is the whole challenge.

Can I connect GA4 to my POS to see this?

Sometimes, and imperfectly. A shared identifier such as an email captured at both the online signup and the loyalty program can stitch the two together, but privacy rules and platform limits make full stitching rare. Most operators settle for a cohort approach: compare repeat behavior of organic-acquired customers against other channels rather than tracing every individual.

Should I report LTV as an exact number?

No. LTV projects future behavior from assumptions, so present it as an informed estimate or a range, and state the assumptions behind it. A defensible estimate that you can explain beats a precise-looking figure that falls apart the moment someone asks how you got it.