Black Friday through Cyber Monday compresses a month of volume into a weekend, which means it also compresses every operational weakness into a weekend: the sync gap that never mattered, the buffer that was never set, the promotion that leaked to a channel it should not have. The seasonal BUYING happened months ago, the Q4 backward calendar; the final fortnight is verification, and this is the checklist.
T-minus 14 days: the stock truth audit
- Cycle-count the A-items: the SKUs carrying the weekend get physically verified NOW, while discrepancies are fixable, record accuracy is the substrate everything else assumes.
- Reconcile inbound: every PO still in flight gets a confirmed arrival date against the receiving plan; anything landing during peak needs a receiving slot that does not steal packing labor.
- Kill the known liars: SKUs with chronic count drift or supplier rejection patterns get corrected or pulled from promotion lists, the weekend is not the time to advertise your least reliable stock.
T-minus 7 days: the systems drill
- Stopwatch the sync: all five scenarios, per channel, with the forced-failure test included, silent breakage discovered on Black Friday costs a weekend, discovered now it costs an hour.
- Raise the buffers on thin, fast SKUs: peak velocity plus multichannel demand is the race condition’s natural habitat; a one-to-three-unit buffer on the contested tail is the cheapest insurance of the season.
- Arm the alerts: low-stock thresholds tightened to peak velocity (a threshold sized for 5-a-day is decoration at 50-a-day), oversell-risk flags on, sync-health notifications routed somewhere a human sees them on a holiday weekend.
- Verify promotion containment: every BFCM discount is a scoped, end-dated override, channel-limited, MAP floors intact, and none of it edits base prices. The classic BFCM pricing incident is a store sale syncing to every marketplace unscoped.
T-minus 2 days: the freeze
- Catalog freeze: no new SKUs, no re-categorization, no variant restructuring into the weekend; changes queue for December.
- Promise sanity: handling times per channel set to what the bench can do at 5x volume, the metrics survive the weekend only if the promises were honest going in.
- The runbook: one page, taped up, who watches which alerts when, what the response to a sync failure is, where the order queue is triaged from, and the decision already made about when a SKU gets pulled rather than risked.
During the weekend: monitor, do not manage
The prepared version of BFCM is boring: cover forecasts glanced at daily, alerts responded to, packing to the deadline sort, and nothing else touched. Every intervention you feel tempted to improvise mid-peak, a price change, a buffer edit, a new listing, is a variance introduced at maximum velocity. The sellers who have a bad weekend mostly did it to themselves on Saturday afternoon.
Common questions
What if a SKU sells out mid-weekend?
If the out-of-stock behavior was configured and the zero is honest, nothing, it is a success state. The failure version is selling three past zero because the counts drifted; that is what the T-14 audit prevented.
Should I hold reserve stock off-channel?
A deliberate reserve released Sunday night is a legitimate play IF it is modeled as a separate pool with a release rule, not a pile someone remembers to add back.
Do I reorder during BFCM?
Only what the reorder points fire on with post-peak lead times in mind, most of what sells out is done for the season, and January stock arriving for February demand is dead stock with a receipt.
What is the single highest-leverage prep item?
The sync drill. Everything else degrades gracefully; a silent sync failure at peak velocity is the one that compounds by the hour.
Peak-proof is a checklist, not a hope
Buffers, in-app alerts, verified sync, and one honest pool under the biggest weekend of the year, from $49/month with unlimited orders. See pricing.