Your BFCM Sender Reputation Is Decided in September
Mailbox providers do not evaluate a single send. They evaluate a pattern. That is the entire mechanism behind BFCM email deliverability, and it is why the campaign that lands in spam on November 27 was actually decided eight weeks earlier.
By the time you can see the problem in your Black Friday numbers, the input that caused it is old news. You cannot fix it in November. You can only fix it now.
Black Friday 2026 is November 27 and Cyber Monday is November 30. That gives you September for cleaning and the start of your ramp, October for building volume, and November for executing against a reputation that is already set. Here is how to use each one.
If you are reading this in November and something has already gone wrong, skip to the last section.
The Volume Spike Problem
Run the numbers on your own sending and the problem becomes obvious.
A typical DTC brand sends four to six campaigns in September, to moderate segments. In the BFCM window that same brand sends fifteen to twenty campaigns, to progressively larger audiences, compressed into about ten days. That is a four to five times increase in volume arriving almost overnight from a sender whose recent history says it is a modest, occasional emailer.
Mailbox providers respond to unexplained spikes the way you would expect. They throttle, they route to the bulk folder, and in the worst cases they block outright. None of that is a punishment for selling things in November. It is a pattern-matching system doing exactly what it was built to do.
The fix is to make the spike explainable by building toward it gradually. The widely cited guidance is to increase daily send volume by no more than five to ten percent per day, with the lower end for very high volume senders and the upper end for smaller ones. Applied across September and October, that is enough runway to arrive at BFCM volume without the jump looking anomalous.
Clean Before You Scale, In That Order
You cannot warm up on a dirty list. Ramping volume on a list full of dead addresses just teaches mailbox providers that you send more and more mail to people who never engage.
So the sequence is cleaning first, ramping second. Not simultaneously, and never in the other order.
What comes out:
- Profiles that have never engaged and have been subscribed more than 90 days
- Hard bounces
- Anyone who has marked you as spam
- Role addresses and obvious typo domains
- The long tail of subscribers who exist only to make your list count look impressive
In Klaviyo, this is what your sunset flow is for, and our guide to the Klaviyo sunset flow and list cleaning covers how to build one properly. If you built the segments from the previous post in this series, your suppression logic is already defined and this step is mostly execution.
Now the objection, because every founder has it: shrinking the list right before the biggest revenue weekend of the year feels insane.
It is not, and the arithmetic is straightforward. Sending to 40,000 addresses that reach the inbox beats sending to 60,000 where a third land in spam and drag the rest down with them. Deliverability is not a per-message property. Poor engagement from the dead portion of your list suppresses inbox placement for the living portion. You are not losing 20,000 customers. You are removing 20,000 addresses that were actively making your email worse for everyone else.
The commonly repeated advice is to start cleaning about three months before BFCM. Counting back from November 27, that is the first week of September. The timing is not arbitrary.
Authentication: Run These Checks Now
Authentication is the part that is genuinely binary. It either passes or it does not, and it takes an afternoon.
Check DMARC, DKIM, and SPF. All three, on the domain you actually send from. Run our free DMARC check tool and read the output rather than assuming a green light somewhere in Klaviyo means everything is correct. For the Klaviyo-specific setup, including the dedicated sending domain, our Klaviyo DMARC setup guide walks through it.
Confirm you meet the bulk sender requirements. Google and Yahoo have required authenticated sending domains, one-click list unsubscribe, and a spam complaint rate below 0.3 percent since 2024. These are not best practices, they are conditions of delivery to a large share of your list.
Do not jump your DMARC policy to reject in October. Move from none to quarantine, monitor the reports, and only tighten further once you can see what is passing and failing. Tightening policy right before peak, without monitoring, is a way to block your own mail.
Everything above is table stakes, and the wider deliverability picture is covered in our ecommerce email deliverability guide.
The September and October Warmup Schedule
This is the part worth screenshotting.
| Window | Send to | What to do | Watch |
|---|---|---|---|
| Sept 1 to 14 | Engaged 30 day only | Clean the list. Fix authentication. Establish a baseline. | Bounce rate, complaint rate |
| Sept 15 to 30 | Add engaged 60 day | Begin widening. Increase cadence by roughly 25 percent per week. | Open rate by mailbox provider |
| Oct 1 to 15 | Add engaged 90 day | Approach BFCM cadence on your strongest segments. | Inbox placement, spam folder tests |
| Oct 16 to 31 | Full BFCM audience | Rehearse peak cadence. Break things now, not later. | Everything |
| Nov 1 onward | As planned | Execute. Monitor only. | Complaint rate above all |
Two thresholds to hold yourself to as you ramp. Complaint rate stays below 0.3 percent, and ideally well below it. Hard bounce rate stays under two percent. If either moves in the wrong direction as you widen, stop widening and go back a step. The ramp is not a schedule you owe anybody. It is a response to what the data tells you.
Watch open rates broken out by mailbox provider rather than in aggregate. Aggregate open rate hides the exact problem you are looking for. If Gmail placement degrades while everything else holds steady, you will only see it if you are looking provider by provider.
If You Have Been Dormant
There is a version of this reader who has sent almost nothing since spring and is planning to email 80,000 people on Black Friday.
Be honest about the situation: that cannot be fully fixed in the time available. A dormant sender attempting peak volume is the single highest-risk profile in email marketing, and the likely outcome is bulk-foldering that damages the domain well into next year.
The correct move is to accept less reach in exchange for keeping the asset. Start immediately, send only to people who engaged in the last 30 days, ramp more slowly than the table above, and plan a BFCM campaign that reaches a fraction of your list and actually arrives. A smaller send that lands beats a larger send that does not, and it leaves you with a domain you can rebuild from in Q1.
What to Do If It Is Already November
Triage only.
Do not add volume. Do not switch sending domains mid-crisis, because a new domain has no reputation at all and you will be starting from zero on the worst possible week. Do not buy or import a list.
Narrow to your most engaged segment, cut send frequency, and protect what reputation you have. Then put September in the calendar for next year, because that is the only place this problem is actually solvable.
The Rest of the Plan
The free BFCM playbook has the full sequence: flows, segments, this deliverability work, and the November send calendar.
Deliverability is the part of BFCM prep where the gap between doing it and doing it properly is hardest to see from the inside, and where the cost of getting it wrong is highest. If you want somebody to run the six week program above on your account rather than reading about it, that is what our BFCM readiness audit covers: authentication, list health, a warmup schedule built for your volume, and a prioritized fix list before the season starts.
Frequently Asked Questions
When should I start warming up my email list for Black Friday?
Early September, roughly twelve weeks before Black Friday. Clean the list first, then ramp send volume gradually across September and October so that peak BFCM volume is the end of a visible trend rather than a sudden spike. Warmup started in November is too late to change the pattern mailbox providers are already evaluating.
How do I check my sender reputation before BFCM?
Start with authentication, since it is binary and quick to verify: check DMARC, DKIM, and SPF on your actual sending domain. Then look at your own metrics broken out by mailbox provider, watching complaint rate, hard bounce rate, and open rate trend. A complaint rate approaching 0.3 percent is a warning sign that needs action before you add volume.
Will sending more emails on Black Friday hurt my deliverability?
Sending more email is not the problem. Sending suddenly more email, to less engaged people, from a sender with no recent history of doing so, is the problem. If you ramp gradually from September and hold your audience to engaged segments, elevated BFCM frequency is normal and mailbox providers treat it as such.
Should I clean my list before Black Friday even though it shrinks my audience?
Yes. Poor engagement from dead addresses suppresses inbox placement for the rest of your list, so the addresses you remove were reducing the reach of the ones you keep. Sending to 40,000 people who receive your email beats sending to 60,000 where a third never see it.
What spam complaint rate is too high for BFCM?
Google and Yahoo require bulk senders to stay below 0.3 percent, and that is a ceiling rather than a target. Treat anything above 0.1 percent as a signal to slow your ramp and tighten your audience, particularly if it is climbing as you widen segments through October.