If you're reading this, one of two things is true. Either your dispensary email open rate has been sliding for weeks and you can't figure out why, or it's already crashed and you're watching a once-healthy program go dark. The good news is the problem is almost always fixable. The bad news is the fix is counter-intuitive — it involves sending less email, not more, to a much smaller audience than you're used to, for three to four weeks before you can return to normal operations.
We wrote this playbook because we've recovered a dozen dispensary email domains over the last 18 months using the same protocol. The anchor case in this article — a single-location Queens dispensary we took over when their open rate had collapsed to 0.55% on a list of 64,171 subscribers — is the most dramatic example in our portfolio, but it's not an outlier. Every program we've taken over with a deliverability problem has followed the same decay curve and responded to the same recovery protocol. Twenty-five days after we started the rebuild, that same list was producing an 18.17% open rate on segmented sends. That's a 33× lift. It wasn't magic. It was the protocol below.
Why Dispensary Email Domains Decay in the First Place
Domain decay is not an accident. It happens for specific, predictable reasons, and it happens to dispensary programs more often than any other retail vertical because of how dispensary marketing usually gets staffed. Somebody inside the shop — the manager, the owner, a part-time marketer — is told to "build the email list" and given a Klaviyo or Mailchimp account. They start sending to everybody who's ever opted in, once or twice a week, forever. Nobody's watching engagement. Nobody's removing dead subscribers. Nobody's paying attention to Gmail's inbox placement data.
Then one week, open rate drops from 24% to 19%. Then to 14%. Then 9%. Then 3%. Then nothing. Here are the five most common causes of the slide, in the order we see them.
1. Sending to Dead Subscribers
The single biggest cause of dispensary domain decay is continuing to broadcast to subscribers who haven't engaged in six, nine, or twelve months. Gmail in particular watches your engaged-to-unengaged ratio very closely. If you're sending to 20,000 subscribers and only 4,000 are opening, Gmail's spam filter reads that as "this sender is sending unwanted mail" and starts routing your future sends to the spam folder — for every subscriber, not just the unengaged ones. The dead weight on your list is quietly poisoning your deliverability for your active subscribers.
2. Paid Opt-In Campaigns That Fill Your List With Unqualified Subscribers
If you've ever run a "sign up for email and get 20% off your first purchase" campaign and added those signups to your main broadcast list, you've probably seeded decay. Discount-driven signups have the lowest engagement rate of any acquisition source. They opt in to claim the discount, redeem it once, and never open another email. Mixed into your main list, they tank the average engagement rate and trigger Gmail's filters.
3. Missing or Broken Authentication (SPF / DKIM / DMARC)
Since February 2024, Gmail and Yahoo have required bulk senders (5,000+ messages per day) to authenticate their mail with SPF, DKIM, and a valid DMARC policy. If you started sending before that deadline and never updated your DNS records, you're probably running on borrowed time — Gmail is deprioritizing unauthenticated mail gradually, not instantly, which is why the decay can take months to become obvious.
4. Subject-Line Spam Triggers
Common dispensary subject lines ("50% off today only," "🔥🔥 DEALS INSIDE 🔥🔥", "LAST CHANCE") are textbook spam-filter triggers. In isolation, one such subject line is fine. As a recurring pattern, they train Gmail to categorize your sends as promotional-low-quality, which deprioritizes them. Combine that with high unopen rates and you compound the decay.
5. Spam Complaints You Don't Know About
If your spam complaint rate exceeds 0.3% — roughly 30 complaints per 10,000 sends — Gmail will move your future mail to the spam folder automatically, regardless of your other metrics. Most dispensary programs have no idea what their complaint rate is because they never check. If you can't answer "what's my spam complaint rate on my last broadcast" in under 30 seconds, check it now. It's the single fastest thing in this article.
Step 1: Diagnose Before You Touch Anything
Before you change a single send, you need to know what you're dealing with. The recovery protocol is the same regardless of cause, but the severity band determines how aggressively you cut volume and how long you warm back up. Run through this 7-point diagnostic and write down the numbers.
- Rolling 30-day open rate. Pull the last 30 days of broadcasts. Below 18% is a warning. Below 12% is a decay event in progress. Below 5% is a terminal diagnosis — you are effectively delisted.
- Gmail inbox placement rate. Use a tool like GlockApps, Mail-Tester, or Litmus to test placement across Gmail, Apple Mail, and Yahoo. Below 65% Gmail placement means your domain is in soft-delisting territory.
- Hard bounce rate on your last send. Above 2% is a major red flag. It suggests your list is full of stale addresses, and the bounces themselves are burning your reputation further.
- Spam complaint rate trailing 30 days. Above 0.3% is the industry threshold that triggers automatic spam-folder routing at Gmail.
- SPF / DKIM / DMARC status. Use mxtoolbox.com or dmarcian.com to check all three. You want 100% pass on all three for every send domain.
- Send-to-engagement ratio on your last broadcast. Messages delivered divided by unique opens. Below 12% means the majority of your list is dead weight.
- Subscriber cohort decay analysis. Look at subscribers who joined 30 days ago vs 90 days ago vs 180 days ago vs 365 days ago. If the 365-day cohort is opening at 3% and the 30-day cohort is opening at 48%, you know where the dead weight is.
Write these seven numbers down on paper. You're going to compare them against the same seven numbers in 25 days. The comparison is what tells you whether the protocol is working, and the early signal (days 7-10) is what tells you whether you need to call in an expert.
The 25-Day Recovery Protocol
Here is the exact protocol we ran on the Queens program, broken into four weekly phases. Every week is a standalone chunk of work. Do not skip ahead. The entire protocol depends on not over-sending in the first two weeks, and the temptation to "get back to normal" will be strong by day 10 — resist it.
The first week is about triage, not sending. You will send exactly two emails during this week, both of them to a very tight engaged cohort. Everything else is cleanup.
- Pause all scheduled broadcasts. Do not send anything to the full list. If you have automations running (welcome, abandoned cart, etc.), leave them — they target engaged users and they're not hurting you.
- Build an "engaged cohort" segment: subscribers who have opened OR clicked any email in the last 30 days. On the Queens program this was roughly 1,900 of 64,171 subscribers — about 3%.
- Suppress everything else temporarily. Do not delete yet; just move the unengaged cohort into a suppression pool.
- Send one mid-week and one weekend broadcast to the engaged cohort only. Strong subject lines, high-value content, no discount-stuffing, no emoji-laden subject lines.
- Verify SPF, DKIM, and DMARC authentication on your sending domain. Fix anything broken. If DMARC is not yet set up, put it in
p=nonemonitoring mode for the first 14 days before moving to quarantine. - Remove all hard bounces from the last 90 days. Most ESPs do this automatically, but verify. Hard bounces on your sending history are still hurting you.
Week two is about generating positive engagement signal for Gmail and Apple Mail to notice. You're still sending only to the engaged cohort, but you're increasing frequency slightly and paying close attention to what the recipients do next.
- Increase send cadence to 3 per week on the engaged cohort only. Not 4, not 5 — 3.
- Every send must contain a clear call to action that drives a click. Clicks are the strongest positive engagement signal, stronger than opens.
- Request a reply on one of the three weekly sends. "Hit reply and tell us your favorite strain this month — we'll feature the best three in next week's drop." Replies are the strongest possible engagement signal.
- Remove any subscriber who hard-bounced in week 01 from the engaged cohort.
- Monitor open rate and click rate daily, not weekly. If you see a drop mid-week, pause the next send and investigate before resuming.
- Set DMARC policy to
p=quarantineonce you've confirmed 14 days of clean authentication.
Week three is where you start reintroducing subscribers from the suppression pool — but only the ones most likely to engage. This is the most delicate week in the protocol. Move too fast and you'll undo the progress from weeks 1 and 2.
- Add subscribers who engaged 31-60 days ago to the active sending cohort. On the Queens program this roughly doubled the cohort from 1,900 to 3,800 subscribers.
- Maintain 3 sends per week. Do not increase frequency yet.
- Send to the newly-expanded cohort using your strongest subject lines from weeks 1 and 2. Don't experiment here — replicate what worked.
- Watch click-to-open rate closely. If CTOR drops below 8% after the expansion, pause and shrink the cohort back by one week.
- Run a deliverability test (GlockApps or Mail-Tester) mid-week. You want to see Gmail inbox placement above 75%, Apple Mail above 85%.
- Begin sunsetting any subscribers who haven't engaged in 180+ days. Suppress them permanently from all future sends. Do not delete them in case of future re-engagement campaigns, but remove them from broadcast lists forever.
Week four is the return to normal operations — but "normal" is different from where you started. Your list is smaller, your sends are tighter, and your infrastructure is cleaner. This is the new baseline and you need to maintain it.
- Expand the active cohort one more time: add subscribers who engaged 61-90 days ago. The Queens program ended week 4 at roughly 5,800 active subscribers out of the original 64,171.
- Resume your normal content calendar (weekly newsletter, product drops, weekend broadcasts) — but only to the active cohort. The 90+ day inactives stay suppressed.
- Run a full deliverability test. Gmail inbox placement should be at 85%+. Apple Mail should be at 90%+. If not, repeat week 03.
- Set DMARC policy to
p=rejectif monitoring has been clean for 14+ days. - Publish a weekly deliverability dashboard for yourself or your team. Open rate, click rate, unsubscribe rate, spam complaint rate, hard bounce rate. Watch them weekly forever.
The Do and Don't of a Recovery in Progress
Do
- Send fewer, better, more segmented emails to your most engaged subscribers only
- Request replies wherever natural — replies are the strongest positive engagement signal
- Monitor open and click rates daily, not weekly, during the protocol
- Fix SPF, DKIM, and DMARC authentication before anything else
- Sunset subscribers aggressively — dead weight is the enemy
- Run authentication and deliverability tests mid-week, every week
Don't
- Don't "blast the list" to "try to get engagement back"
- Don't add new subscribers via paid discount campaigns during recovery
- Don't run "we miss you" broadcasts to inactive subscribers — these almost always fail and amplify decay
- Don't change email service providers mid-recovery — migration compounds the reputation problem
- Don't panic on day 10 when open rate is still climbing slowly — give it the full 25 days
- Don't skip week 03's cohort expansion because "everything's working" — you must keep expanding or the audience permanently shrinks
The biggest mistake we see: dispensaries that successfully recover their program during weeks 1-2 and then immediately return to old habits in week 3 — blasting the full list, running discount signups, sending 4+ times per week. They undo a month of careful work in a single bad week. The protocol only holds if you treat it as a permanent operating change, not a temporary intervention.
How to Maintain a Healthy Domain After Recovery
The protocol above gets your domain back into inbox placement. The six practices below keep it there. This is the operating manual for any dispensary email program that wants to never repeat this recovery.
Monthly List Hygiene
Once a month, pull a report of subscribers who haven't engaged in 120 days. Move them to a suppression cohort. If they haven't engaged in 180 days, suppress them permanently. The math on this is unforgiving: sending to dead addresses costs you more than never having had them. Our 2026 Cannabis Email Marketing Benchmarks Report shows top-quartile programs run active member rates above 78% — that's only possible with aggressive monthly sunsetting.
Segment Before You Send
A healthy dispensary email program sends 3-5 segmented campaigns per week, not one broadcast to the full list. Segments should be based on actual customer behavior: frequency tier, last-purchase recency, product category affinity, loyalty tier. The full-list broadcast, when it happens at all, is a once-a-quarter tentpole moment — never a weekly habit.
Watch Click-to-Open Rate, Not Just Open Rate
Open rate is corrupted by Apple MPP pre-fetches. Click-to-open rate is not. CTOR is the cleanest signal you have for whether your content is resonating with the people who actually saw it. If CTOR drops below 9% for two consecutive weeks, something in your creative or offer is wrong. Fix it before the open rate slides too.
Keep Welcome Sequences Running
Your welcome sequence is the highest-engagement series in your program and it trains Gmail to expect good things from your sending domain. Do not pause welcome automations during recovery (or ever). A strong welcome series is a free reputation boost on every new subscriber.
Use a Subdomain for Bulk Sends
Set up a separate subdomain (e.g., mail.yourdomain.com) for bulk marketing sends and leave your primary domain for transactional and one-to-one mail. This isolates your marketing reputation from your transactional reputation, so a decay event on one doesn't contaminate the other.
Run a Quarterly Deliverability Audit
Four times a year, run a full deliverability audit: inbox placement test, authentication check, spam complaint rate review, subscriber cohort analysis, list hygiene review. Catch problems at the warning stage before they become decay events. This takes two hours per quarter and prevents every recovery you'd otherwise need.
When to Call for Help
The protocol above works for programs in the warning and early-decay stages. If your diagnostic numbers look like this, you can run it yourself with your existing ESP and some discipline:
- Open rate between 10-18%
- Gmail inbox placement between 60-80%
- Spam complaint rate below 0.25%
- Authentication partially working (SPF or DKIM in place)
If your numbers look like the Queens program's did — open rate at 0.55%, Gmail effectively rejecting all mail, complaint rate over 0.3%, authentication broken — you are in the severe-decay range and a self-directed recovery is unlikely to work. At that point your domain needs hands-on intervention: possibly a subdomain setup, possibly a temporary migration to a seed-and-warm IP, possibly a complete authentication rebuild. We've run this protocol end-to-end on programs in every severity band and the severe cases need expert help. Trying to rebuild a severely damaged domain on your own usually makes it worse — the half-measures prolong the decay and burn more subscriber relationships in the process.
If your open rate is below 10% and you're not sure where to start: we'll run a free 30-minute deliverability audit on your sending domain and tell you exactly what severity band you're in and which version of the protocol applies. If you're in the self-directed range, we'll point you at the right checklist. If you're in the severe range, we'll tell you honestly what a managed recovery would look like and what it would cost. Book a 30-minute strategy call →
Frequently Asked Questions
How long does the full recovery take?
Between 21 and 35 days for most dispensary programs. The 25-day figure in this article is a typical midpoint. Programs with severe decay (open rates below 5%) can take up to 45 days. The protocol doesn't get faster if you rush it — it gets slower if you try to skip phases.
Can I switch ESPs to fix this?
No. Changing ESPs does not reset your domain reputation. Reputation is attached to your sending domain and IP history, not to the tool you're sending through. Switching ESPs during a recovery usually makes things worse because you lose your sending history and start from scratch on a new IP with warmed-up expectations.
Will my open rate come all the way back to where it was?
Not immediately, and maybe not ever — but that's because where you started was probably unsustainable. The Queens program's 25% open rate in month zero was being driven by a list that was already quietly decaying. Their new 18% open rate after recovery was a healthier, more honest number on a cleaner list. The revenue per send was actually higher post-recovery despite the lower absolute open rate, because the remaining subscribers were real customers.
Do I need a new sending domain?
Rarely. For 90% of dispensary recovery cases, you do not need a new domain — you need a cleanup protocol on your existing domain plus proper authentication. A new domain is a last-resort fix that comes with its own warmup costs (a fresh domain has zero reputation and has to build from scratch over 4-6 weeks). Only consider this if your existing domain has been on a major blocklist for 60+ days.
What if I'm using a free-tier ESP (Mailchimp free, Klaviyo free)?
The protocol still works, but you're going to hit frustrating limits around segmentation and suppression. Free tiers usually don't let you build dynamic "engaged cohort" segments. You can work around this by manually exporting CSVs, segmenting in a spreadsheet, and uploading the reduced list back in — it's a pain but it does the job. If you're serious about email as a revenue channel, upgrading to a paid ESP pays for itself in the first month.
Does this protocol work for SMS decay too?
The principles are the same (cut volume, clean list, segment hard) but the execution is different. SMS decay is mostly opt-out-driven rather than deliverability-driven. We'll write a dedicated SMS recovery playbook soon. In the meantime, if your SMS opt-out rate is above 1%, cut your send cadence by half immediately and segment your next 10 sends by loyalty tier.