The Klaviyo BFCM Flow Audit: 8 Flows to Fix Before November
Most brands spend October writing Black Friday campaigns. Almost nobody spends October auditing the Klaviyo BFCM flows that will actually catch the traffic those campaigns generate. That is backwards. Your flows run on intent, and BFCM is the highest intent week of your year.
Here is the timing that matters. Black Friday 2026 falls on November 27 and Cyber Monday on November 30. You are reading this in August, which means you have all of August and September to fix things, October to test them, and November to leave them alone. That last part matters: late November flow changes are the ones that break accounts.
This is the audit we run on client accounts before Q4. Eight flows, what specifically goes wrong in each one under BFCM conditions, and the fix. If you do not have the core stack built yet, start with our guide to the Shopify email flows every store needs and come back here once they exist.
Why Flows Decide BFCM, Not Campaigns
Campaigns run on your calendar. Flows run on your customer's intent. During BFCM, intent spikes across every trigger you have at once: more browsing, more carts, more checkouts started and abandoned, more stockouts, more first-time buyers.
That is why flow revenue per recipient consistently outperforms campaign revenue per recipient, and why the gap widens during peak. A misconfigured flow filter that costs you three percent of recoverable revenue in March costs you a great deal more on November 27, because the traffic behind it is many times larger. Same bug, much bigger bill.
So the audit is not an optimization exercise. It is risk reduction on the highest-traffic weekend of your year.
The 8 Flows, and What Breaks in Each
1. Welcome series
What breaks: Your welcome flow offers a standing 10 or 15 percent discount. In November, that offer is worse than your sitewide BFCM deal. New subscribers who join on November 20 get a welcome email offering them less than the banner on your homepage, which teaches them that your email is not where the good offers live.
The fix: Add a conditional split on date. From November 1 through December 2, route new subscribers into a BFCM-specific branch that leads with early access rather than a percentage that will be beaten in a week. Our Klaviyo welcome series guide covers the base structure; this is a seasonal variant of it, not a rebuild.
2. Abandoned checkout
What breaks: Timing. A standard four hour first delay is fine in ordinary trading. It is too slow when the offer expires Monday night and your customer is comparison shopping across six tabs.
The fix: Compress the delays for the BFCM window. Move the first message inside an hour and tighten the follow-ups. Rewrite the copy to reference the actual deadline rather than generic scarcity. "Your cart is waiting" is weak on Black Friday. "Your cart is waiting and the sale ends Monday at midnight" is not.
3. Added to cart and browse abandonment
What breaks: Most accounts have one of these and not both. Browse abandonment is usually the missing one, and it is where BFCM window shoppers live. People spend late November browsing and comparing without adding anything to a cart.
The fix: Build browse abandonment if it does not exist, and tighten the trigger filter so it fires on meaningful product views rather than on every page load. If you only have capacity to build one new flow before November, build this one. Our abandoned cart email strategy post covers the cart side in depth.
4. Post-purchase
What breaks: Two failure modes, both common. The first is cross-selling at full price to somebody who just bought at 40 percent off, which reads as insulting. The second is thanking the customer and then going completely silent for 30 days, which wastes the only window where a first-time BFCM buyer is actually paying attention to your brand.
The fix: Build a BFCM variant of the post-purchase flow. Lead with order confidence and product education rather than a full-price ask. Save the second-purchase push for December, which is a whole topic in itself and the subject of the last post in this series.
5. Back in stock
What breaks: It is usually not set up at all. BFCM is the highest stockout period of your year, and every stockout without a back in stock flow is demand you generated and then discarded.
The fix: Enable it, and treat it as a revenue flow rather than a customer service courtesy. Make sure the notification is fast and the copy carries the same urgency as the rest of your BFCM messaging.
6. Winback and lapsed purchaser
What breaks: It keeps firing its normal reactivation offer during BFCM, and that offer is weaker than your sitewide deal. You are sending a lapsed customer 15 percent off in the same inbox hour that your campaign sends them 40 percent off.
The fix: Either suppress the winback flow for the BFCM window or upgrade its offer so it is at least as good as the public one. Suppression is the safer choice if you are short on time.
7. VIP and high-CLV early access
What breaks: It usually does not exist. This is the highest-margin BFCM automation most accounts are missing.
The fix: Build a flow that gives your top decile of predicted customer lifetime value a genuine 24 to 48 hour head start on the sale. Not a fake head start. A real one, with real inventory behind it. These customers convert at full attention before the inbox turns into a warzone on Friday morning, and they do not need the deepest discount to do it.
8. Sunset and suppression
What breaks: It does not run, so you carry thousands of dead addresses into the highest-volume sending period of your year and damage the sender reputation you need on November 27.
The fix: Run it in September. Not November. This is important enough that it gets its own post later in this series, on why your BFCM sender reputation is decided in September rather than in November.
The Klaviyo Settings That Quietly Sabotage BFCM
The eight flows above are the visible layer. Underneath them sit account settings that cause the failures nobody diagnoses until it is too late.
Smart sending is not universal inside flows. This surprises people. Smart sending has to be configured per message inside a flow, not once at the account level. Open each promotional message in each flow and make a deliberate decision. During BFCM you are intentionally contacting people more often than usual, so a smart sending window set for ordinary trading will silently skip profiles you meant to reach.
Flow scheduling is your friend. You can schedule BFCM flow variants to switch on for the sale window and hand back to your always-on flows afterward. Set this up in October so nothing depends on somebody remembering to toggle a flow at 6am on Black Friday.
Coupon code pools run dry. If your flows use dynamic unique codes, size the pool for peak volume rather than for a normal month. A flow that cannot issue a code sends a broken email.
Dynamic blocks render sold-out products. Check your fallback logic. A hero block pointing at an out-of-stock bestseller on Friday morning is a wasted send.
Flow filters written for normal season behave oddly across a multi-purchase weekend. Filters like "has placed order zero times since starting this flow" do exactly what they say, which is not always what you want when a customer buys twice in four days.
Your August to November Timeline
| Window | Work |
|---|---|
| Aug 10 to Aug 31 | Audit all eight flows. Fix what is broken. Build what is missing. |
| Sept 1 to Sept 30 | Segment build and deliverability work. Run the sunset flow. |
| Oct 1 to Oct 31 | Test everything with real sends and seed lists. Schedule BFCM flow variants. |
| Nov 1 onward | Freeze. Monitor only. No structural changes. |
If you get through August with the eight flows fixed, you are ahead of most brands your size.
Get the Rest of the BFCM Plan
The flows are step one of four. Segments, deliverability, and the November send calendar are the rest, and the free BFCM playbook walks through all of them with the checklists we use internally.
If you have just read the eight flows above and counted three or four that are broken in your own account, and BFCM is a big enough number that guessing is not acceptable, our BFCM readiness audit is the paid version. We go through your Klaviyo account flow by flow, find what will break under peak volume, and hand you a prioritized fix list. It takes a week and it happens before the season, not during it.
Frequently Asked Questions
Which Klaviyo flows matter most for Black Friday?
Abandoned checkout, browse abandonment, and welcome are the three that carry the most incremental BFCM revenue, because they catch the intent your campaigns generate. VIP early access is the highest-margin flow because it converts your best customers without your deepest discount. Back in stock matters more during BFCM than at any other time of year because stockouts peak.
Should I turn off my normal flows during BFCM?
Not all of them. Suppress or upgrade any flow whose standing offer is weaker than your sitewide BFCM deal, which usually means winback and sometimes welcome. Leave the intent-driven flows running, because abandoned checkout and browse abandonment are doing their best work of the year during that window.
When should I stop making changes to Klaviyo flows before Black Friday?
November 1. Use October for testing with real sends and seed lists, then freeze. Structural flow changes made in the week before Black Friday are the single most common cause of self-inflicted BFCM problems, because there is no time left to catch what the change broke.
How many emails should a BFCM abandoned checkout flow have?
Three to four messages, compressed into a much shorter window than your normal flow. During BFCM the deadline does the persuasion work, so speed matters more than sequence length. One well-timed reminder inside the first hour usually beats a fifth message on day four.
Do I need a separate welcome series for BFCM?
You need a seasonal branch, not a separate flow. Add a conditional split on date so subscribers who join during the BFCM window get early access messaging instead of a standing percentage discount that your sitewide sale is about to beat.