Half of Shopify brands don't peak in BFCM. They still pay for it.

Three things. Protect your margin instead of buying into an auction you can't win, keep testing so the traffic you already have converts better, and get your site ready for the first two weeks of December when ad costs drop and attention comes back. The mistake most non-peak brands make is going quiet in November, which means they arrive in December with the same untested site they had in October.

Because you're bidding against every brand that is. Paid inventory is a shared auction, so when thousands of advertisers raise budgets and put a 40% offer behind their creative, the clearing price goes up for everyone in the auction, including you. Your creative didn't get worse. The competition got more aggressive and better funded, and your evergreen offer has to compete against a discount.

If you're running BFCM offers, yes, mostly. Your traffic mix is distorted and your results won't generalize. If you're not running offers, the opposite is true. Your November traffic is more intentional than your usual mix, which makes it good test traffic. The thing to change is not whether you test but what you test: incremental optimization rather than structural experiments you can't read cleanly against a noisy backdrop.

After November 30 the brands that ran Black Friday offers pull them offline and cut spend. Ad costs fall and consumer attention frees up. For a brand with a tested, optimized site, that's the cheapest high-intent traffic of the quarter. The window is short, so it goes to whoever prepared during November rather than whoever reacts in December.

It depends on your volume. If November traffic falls far enough that you can't reach significance in a reasonable window, run fewer tests on higher-traffic pages rather than the same number spread thin. What doesn't work is stopping entirely and restarting in January, which costs you the December window.

That's a different problem from an expensive auction and it needs a different answer. Spending harder against soft demand makes the month worse. The session covers how to separate a demand problem from a conversion problem, and what each one calls for.

Shoplift tests at the theme level. Instead of changing one button or one headline and measuring that in isolation, you run a complete variant of the store experience against your current one and see how the whole thing performs. It's built natively into Shopify, so tests go live the same week rather than sitting behind an engineering backlog. For a brand deciding what its site should look like on December 1, that speed is the difference between shipping the change and missing the window.

Yes, and the emphasis shifts. For subscription brands, November is usually a retention problem before it's an acquisition problem. Churn and the subscriber experience carry more of your revenue during a month when new customer acquisition gets expensive. The session and the playbook both cover the retention side.

No. Register and we'll send you the recording either way. Live attendees can ask questions, which is the main reason to show up.