Shopify generated £14.6B in BFCM 2025, peak £5.1M per minute. The 8-week checklist we use for every BFCM-readiness engagement, ordered by week and by priority.
BFCM doesn't care about your average month. Traffic spikes 5–20×. Every weakness is exposed at scale. The window to fix things is 8 weeks before, not 8 days.
This is the checklist we run for every BFCM-readiness engagement. Ordered by week, prioritised by impact-vs-effort.
Baseline everything before changing anything. The deliverable is a one-page report with these numbers:
Apps are the #1 cause of BFCM outages. Specifically: apps that work fine at normal traffic and fail under load.
Simulate peak traffic. Not on production — on a Shopify development store mirror. We use k6 to generate concurrent users following realistic flows: PLP browse → PDP → cart → checkout (test mode).
The numbers we look for: response time degradation curve, app failure points, checkout completion rate at 5×, 10×, 20× normal load.
Image and media optimisation is the highest-leverage performance work. Specifically:
BFCM landing pages should be built and live by week 4. Why so early? Two reasons:
Build campaign pages as proper OS 2.0 templates with sections — not as marketing-team-built page-builder slowdowns. Performance budget applies to campaign pages too.
Plus only. Launchpad automates the on-the-day operations:
If you're not on Plus, the same effects need a person + Slack reminder + manual button-pushing. Doable but error-prone at peak.
The four numbers to watch in real-time:
Hot-fix capacity reserved. Senior dev on-call for the 4-day window. We've never had to use it on a properly-prepared store. We've always had to use it on stores that skipped weeks 7 and 6.
