The Email Flows Cannabis Brands Skip (And Why It Hurts Later)

Most cannabis brands don’t skip email flows because they don’t believe in them. They skip them because the brands around them did.

That’s the part worth sitting with. Flow gaps are rarely a strategy decision. They’re a pattern copied from the nearest competitor who also skipped them, usually because that competitor copied someone else. The whole category ends up running on a shared blind spot that nobody actually chose.

Welcome series, post-purchase, win-back, browse abandonment, review request, replenishment, VIP. Most cannabis programs run one or two of these. Usually the welcome. Usually half-configured. The rest sit in the ESP as empty shells or unfinished drafts from whoever set up the account two years ago.

The loss doesn’t show up in any one campaign. It shows up as a flat revenue line that never compounds.

This piece is long because the problem is structural, and structural problems don’t respond to quick fixes. What follows is the actual playbook: which flows matter, what the numbers say they should produce, where cannabis-specific compliance changes the build, and how to sequence the work so you stop losing money without noticing.

If you’d rather skip all of this and have someone just do it, that’s fine too. The link is at the bottom. But read this first, because you should know what you’re actually buying.

Why flows matter more than campaigns

Start with the math, because the math is where this argument gets won.

Across Klaviyo’s full ecommerce dataset, the average revenue per recipient on an automated flow email is roughly $1.94. The average on a one-off campaign send is about $0.10 to $0.11. That’s close to a twenty-fold difference on the same subscriber, in the same channel, on the same week.

For well-built ecommerce programs, Klaviyo typically drives somewhere between 25% and 45% of total revenue, and the flow-to-campaign split usually lands around 50/50. Mature programs with a complete flow stack often push flow revenue to 30% to 40% of total email revenue. If your flow revenue is below 15% of email revenue, you have gaps. Not opinions. Gaps.

Now layer in a specific cannabis data point that most operators miss. Klaviyo’s 2023 industry data flagged “Tobacco & Cannabis” as the number one industry for email click-to-purchase conversion rate, at roughly 3.7 times the cross-industry average. The intent is there. The buyers are ready. What’s usually missing is the automated infrastructure to meet them at the moment they’re ready.

That’s the whole thesis. In cannabis, flows are not the nice-to-have. They are the channel.

The retention problem nobody wants to talk about

Cannabis retention is worse than you think it is. Worse than most operators admit in a room.

A Happy Cabbage study across five separate retailers found that only 16% of customers continued shopping at the same dispensary over a five-year window. Flowhub’s retention data shows that most dispensaries lose 45% to 55% of first-time shoppers — they buy once and vanish. Between 70% and 80% of a typical dispensary’s customer file hasn’t shopped in more than two months at any given time.

The standard response to this is a loyalty program and a points system. That’s not wrong. It’s just not enough.

The reason retention is so poor in cannabis is not that customers don’t like the product. They buy again. They just buy again somewhere else. The brand that wins the second purchase is almost always the brand that shows up first, with the right message, at the moment the customer is ready to buy again. That timing cannot be done manually. It has to be automated. Which brings us to the flows themselves.

The flows every cannabis brand should be running

Seven flows carry almost all the weight. Below each one is what the data says it should actually produce, so you have a reference point instead of a vibe.

1. Welcome series

The welcome is the highest-stakes flow you own, because it’s the only one where the subscriber has explicitly raised their hand and asked you to email them. If you don’t convert them here, you almost certainly never will.

Klaviyo’s flow benchmarks put the average welcome flow at roughly $2.65 revenue per recipient, with the top 10% of brands hitting $21.18. That spread — an eight-fold gap between average and top-decile — is not about cleverness. It’s almost entirely about whether the flow exists as more than one email, whether it segments on how the subscriber arrived, and whether it actually introduces the brand instead of just handing out a 15% off code and going silent.

A real welcome series for a cannabis brand is usually four to six emails over seven to ten days. It handles the discount (if you use one), but it also does the unglamorous work: explaining what you actually sell, why your sourcing matters, how to choose a product if they’re new to the category, and what they should expect in their inbox going forward. That last part is the single most overlooked piece of a welcome series, and it’s the piece that makes the difference between a subscriber who opens your third campaign and one who marks it as spam because they forgot who you were.

2. Abandoned cart

The highest-revenue flow in ecommerce, almost universally. Klaviyo’s own benchmark data puts the average abandoned cart flow at $3.65 revenue per recipient — roughly 38% higher than the welcome flow — and the top 10% of brands generate $28.89 per recipient. Cross-industry, abandoned cart flows deliver the highest average placed-order rate of any flow type at 3.33%.

In cannabis specifically, the opportunity is larger, not smaller. CannaVersions reports average dispensary abandoned cart rates of 50% to 75%, with 10% to 15% of those recoverable through a properly built flow. Think about what that math actually means: on a dispensary doing modest online order volume, a working abandoned cart flow is often a five-figure monthly revenue line that currently rounds to zero.

Two or three emails is the sweet spot. One email recovers most of what’s recoverable. A second captures hesitation. A third catches the people who genuinely got distracted. Past that, you’re training subscribers to ignore you.

3. Browse abandonment

Lower intent than cart, higher volume. Average revenue per recipient is around $1.07, with top-decile brands hitting $7.21. This is the flow that catches the person who looked at three pre-rolls on a Tuesday night, closed the tab, and would have bought from whoever emailed them first on Friday.

Most cannabis brands don’t run this at all, because it requires decent browsing data and a bit of setup that doesn’t feel urgent until you realize how much traffic your site gets that never converts on the first visit.

4. Post-purchase

This one is almost never built properly, and it’s the flow that determines whether your customer ever comes back.

Klaviyo’s average post-purchase RPR is $0.41, but top-decile brands hit $5.14 — a twelve-fold gap. The reason the spread is so wide is that most post-purchase flows are just order confirmations and shipping updates, and most top-performing flows are actual sequences that confirm the order, set expectations, educate on the product they just bought, ask for a review at the right moment, and tee up the next purchase.

For cannabis, the post-purchase flow is also where you capture zero-party data that’s almost impossible to get elsewhere. What did they buy? What effect were they looking for? Did it work? That information, collected at the moment the product is fresh in their mind, powers every other flow downstream.

5. Replenishment

Cannabis products have a consumption cycle. Flower, edibles, cartridges, tinctures — every category has a rough average days-to-depletion based on purchase size and typical use. A replenishment flow fires at the right moment before that runs out.

This flow doesn’t show up on most benchmark lists because it’s category-specific, which is precisely why it’s an advantage. Brands that build it well see the second purchase happen weeks sooner than it otherwise would, and that compression of the repurchase window is worth more over a year than almost any single campaign optimization.

6. Win-back

Given that 70% to 80% of a dispensary’s customer file is usually more than 60 days inactive, a win-back flow is effectively the largest addressable audience you have. And it pulls double duty. On one side, it recovers lapsed revenue directly — CannaVersions reports 2% to 3% conversion on inactive customers through proper win-back sequences, which is not enormous per-subscriber but is very large in aggregate.

On the other side, and this is the part most operators miss, a win-back flow is a deliverability instrument. Sending to subscribers who consistently don’t engage drags down your sender reputation. Gmail and Yahoo now enforce a spam complaint ceiling of 0.30%, with Gmail recommending under 0.10% as the working standard, and roughly 30% of bulk senders are still partially non-compliant with the 2024 requirements. Non-compliant senders see spam-folder placement jump from a 5–10% baseline to 22–34%. A win-back flow that either re-engages dormant subscribers or cleanly sunsets them is what keeps your reputation intact while you’re sending campaigns to everyone else.

7. Review and referral

The smallest flow on this list, and the one that compounds the most. A review flow, timed correctly after product delivery, generates the social proof that every other channel — paid, organic, retail — uses to close the next sale. A referral flow turns your best customers into acquisition, which matters more in cannabis than almost any other vertical because paid acquisition channels are still severely restricted.

Neither of these flows will be a top-three revenue line. Both of them make every other flow work better.

What makes cannabis different

Everything above applies to ecommerce generally. What changes in cannabis is the risk layer around it, and that changes the build.

ESP risk is real. Most mainstream ESPs carry policy language that allows suspension of cannabis accounts with limited notice. That risk is not theoretical. When your account goes down, your flows go down with it. A proper cannabis program runs on a cannabis-tolerant ESP, with a portable list, full authentication, and a clear migration plan. If you can’t move your program to a new platform in under four hours with flows, segments, and templates intact, you don’t own your program. You’re renting it.

Deliverability is the actual moat. Inbox placement for regulated categories is consistently below the cross-industry baseline, which means the brands that engineer for deliverability — not creative, not offer, not subject lines — win. Authentication (SPF, DKIM, DMARC) is table stakes. A dedicated sending domain is table stakes. A click-tracking subdomain separate from the root is table stakes. Engagement-based segmentation is table stakes. None of that is glamorous. All of it is load-bearing.

Content constraints change the copy, not the flow. Compliant cannabis email copy can’t make health claims, can’t use certain trigger terms, and in many states can’t promote directly to non-medical audiences. That affects what you write inside the flow. It does not change whether the flow should exist. The number of cannabis brands that skip post-purchase flows because they’re worried about compliance is large, and the worry is misplaced. A post-purchase email that confirms the order, asks how the product performed, and requests a review has no compliance issue. It has a laziness issue.

What actually happens when you don’t build this

Flat revenue is the obvious symptom. Here’s the less obvious one.

Every send leans on the same fatigued segment. Engagement concentrates, then narrows, then flattens. Apple Mail Privacy Protection auto-loads tracking pixels for roughly 50% of subscribers, which means your reported open rates are inflated and you’re making decisions on unreliable data — and without flows generating behavioral signals (clicks, purchases, replies), you don’t have the other data you’d need to see the degradation until it’s severe.

By the time the revenue line tells you something’s wrong, you’ve been on a slow deliverability decline for months. Rebuilding sender reputation from that position takes a 4-to-8-week warming process on new infrastructure if the damage is bad enough to force a migration. That’s 4-to-8 weeks of suppressed revenue while you recover from a problem that flows would have prevented.

This is the full cost of skipping flows. Not the revenue you could have earned with them running. The revenue you’re losing on campaigns that should be working and aren’t, because the program underneath them has quietly decayed.

The sequence that actually works

If you’re starting from zero or close to it, the order matters. Most brands get this wrong because they build the flow that sounds most fun (usually welcome) and then run out of energy before the rest get done.

Build in this order:

First, abandoned cart. Fastest payback, largest immediate revenue, and it works even if nothing else in the program is functioning. Two emails minimum, three if you have the volume. No discount on email one. A small incentive on email two only if your data supports it.

Second, welcome series. Four to six emails. Segment on acquisition source if you have more than one. The first email goes out within minutes of signup, not hours. The sequence should finish before day ten.

Third, post-purchase. Three to five emails. Order confirmation, expectation-setting, product education, review request, next-purchase tee-up. This is also where you start capturing zero-party data through simple preference questions.

Fourth, win-back. Define “lapsed” based on your actual repurchase window, not a default 90 days that Klaviyo suggests. For most cannabis brands, 45 to 60 days is a more accurate signal. Three emails. Sunset the non-engagers cleanly at the end.

Fifth, browse abandonment. Lower priority because it requires more setup and generates lower per-recipient revenue, but it fills a gap the first four flows don’t cover.

Sixth, replenishment. Category-specific, requires some data work to dial in timing, and is the flow that separates mature programs from everyone else.

Seventh, review and referral. Lightest lift, longest compounding impact.

A full build typically takes four to six weeks if you’re doing it properly. The first two flows should be live in week one. If your agency or internal team is telling you the abandoned cart flow will take a month, find a different agency.

What this is worth

A conservative estimate. A cannabis brand doing $100k/month in online revenue with no functioning flows, building the stack above, should see email rise to 25% to 35% of total revenue within 90 days of the full build being live and seasoned. That’s an incremental $25k to $35k/month, most of it margin, most of it compounding monthly as the flows accumulate data and improve.

A top-decile build, over 12 months, closes the gap between the $1.94 average flow RPR and the $16.96 top-decile flow RPR on a growing list. That’s where the category-leaders live. Not on better creative. On better infrastructure.

Your two options

You can build this yourself. Everything above is public information. The flows are documented. The benchmarks are published. The deliverability rules are on Gmail’s own developer site. A disciplined internal marketer with six to eight weeks of focused time and a cannabis-tolerant ESP can absolutely stand this up.

Can you do this in-house?

Technically, yes. If you have someone on your team who understands email infrastructure, knows the difference between SPF, DKIM, and DMARC without looking it up, has built segmentation logic before, can write compliant cannabis copy that still converts, understands how Klaviyo attributes revenue across overlapping flows, and has eight weeks of uninterrupted runway to do the work, you can absolutely run this internally. Most internal marketers are sharp enough to follow a playbook. The playbook is not the hard part.

Why you probably shouldn’t

The hard part is what you don’t know is broken until it’s already broken.

Cannabis deliverability has a half-life. A welcome flow built on a shared IP pool can look fine for six months and then quietly collapse when one of the other senders on that pool triggers a reputation event you had nothing to do with. A post-purchase flow sending to a list with 30% unengaged subscribers will drag your Gmail placement below the 85% threshold where recovery becomes a multi-week project, and you won’t see it in your open rates because Apple Mail Privacy is inflating 50% of your data. A DMARC policy set to p=none (which is the default almost everyone leaves in place) means your domain is technically compliant and functionally exposed — and the day a spoofed email goes out under your name is the day you find out.

Most in-house builds I see are not wrong. They’re incomplete in ways that surface nine months later, when the revenue line flattens and nobody can explain why. By then, the fix isn’t building the flows. The fix is unwinding eight months of accumulated reputation damage, which is slower, more expensive, and more fragile than getting it right the first time.

There’s also the ESP risk. Most mainstream platforms reserve the right to suspend cannabis accounts with limited notice, and “limited notice” in practice sometimes means hours. If your list, flows, and templates aren’t structured for portability — and most in-house builds aren’t, because portability isn’t something you think about until you need it — a platform action takes your entire email program offline during the peak revenue week you built it for.

None of this is hypothetical. All of it is recoverable. It’s just much cheaper to not need to recover from it.

If you’d rather talk through where your program sits before something breaks, email me: hello@highopens.com

If this feels familiar, you’re probably seeing it too.