Growth
What Black Friday Taught Us About CRO This Year
The Black Friday and Cyber Monday numbers are in, and the headline reads like good news: record online spend, up meaningfully year over year. The footnote is the real story. With inflation running where it has all year, much of that growth is price, not volume, and the deep, early discounting that dominated the season tells you retailers knew exactly how price-sensitive this shopper was. Inventory gluts met nervous consumers, and the discount became the strategy.
We spent the weekend watching dashboards for our commerce clients, and now that the dust has settled, here is our honest retro on what moved conversion this season and what turned out to be theater.
What actually moved the number
Across the accounts we can see, the wins clustered in unglamorous places:
- Speed under load. The strongest correlation with conversion all weekend was simply staying fast at peak. One client's mobile LCP degraded by about a second during their email-driven traffic spikes, and conversion sagged in lockstep until a misbehaving personalization script was cut at midday. The performance work is CRO. It always was.
- Price clarity beat price drama. Pages that stated the deal plainly, real strikethrough, real percentage, shipping cost visible before checkout, outperformed the countdown-timer carnival. This year's shopper arrived skeptical, comparison-tabbed, and fluent in fake urgency. Respecting that fluency converted better than exploiting it.
- Payment flexibility. Wallets and buy-now-pay-later did visible work, especially on mobile and especially at higher cart values, which is exactly what you would predict in a squeezed-budget year. Every tap you remove from checkout matters more when the buyer is ambivalent.
- The email list, again. Owned channels delivered the cheapest revenue of the weekend by a comic margin. Post-ATT paid acquisition costs made this the year the list finally got the credit, and the budget, it deserved.
What was theater
Equally instructive is what did not matter. Homepage takeover animations, elaborate gamified promos, and most of the personalization layer produced nothing distinguishable from noise in our post-weekend reads. One honest test: a client's "personalized recommendations" module, A/B tested against a simple bestsellers grid over the weekend, lost. Bestsellers are social proof, and social proof travels better than algorithmic guesswork when catalog data is thin.
I want to be careful here: the reads are directional. Holiday weekends are the worst possible test environment, traffic mix shifts hourly, promotions confound everything, and anyone claiming clean causal wins from BFCM data is selling something. Which is itself the meta-lesson.
The December game is different
BFCM optimizes for decisive, deal-primed buyers. The December shopper is a different creature: later, more anxious, and buying for someone else. The CRO priorities shift accordingly, and this is the week to make the shift:
- Shipping cutoff dates, visible everywhere, updated honestly. Nothing converts a December browser like credible delivery confidence, and nothing torches trust like a missed promise.
- Gift certainty: return windows extended through January, stated up front, gift receipts and wrapping surfaced in the cart rather than discovered at the end.
- A digital gift card path that takes under a minute, because from around December 21 it becomes your best-converting product and most teams treat it as an afterthought.
Write it down now
Last recommendation, the one nobody follows: hold the retro this week, while it hurts. Every screenshot, every incident, every "we should have," into a document dated today. The single most valuable CRO asset for holiday 2023 is an honest record of holiday 2022, written before the memory smooths itself over.
Next year's baseline, as we keep learning in this business, is built from whatever you bother to write down now.
Building something this could apply to?
We take on a small number of flagship projects each quarter.
Start a project