Sixty-five percent of advertisers plan to use Google's seasonality adjustments for Black Friday. New data from Optmyzr suggests most of them will regret it.
The study analyzed 5,886 Google Ads accounts across three consecutive BFCM periods (2022 through 2024). Advertisers who applied seasonality adjustments saw CPCs rise 34% during the holiday window, double the 17% increase experienced by those who left Smart Bidding alone. ROAS fell 15.7% for the adjustment group. The hands-off cohort? Their ROAS rose 5.7%.
That's a 21-point swing in efficiency, and it repeated every year in the dataset.
The Precision Problem
Seasonality adjustments work by telling Google's algorithm to expect a specific conversion rate lift. If you input +50%, bids rise roughly 50% immediately. The system takes your forecast literally. It does not hedge.
The math breaks down fast. If actual conversion rates rise only 40% instead of 50%, you're overbidding by about 7% relative to the true value of each click. During BFCM, when volume is enormous and CPCs are already elevated, that miscalibration compounds into real money within hours.
Optmyzr's analysis frames this as a "precision-risk problem." Most marketers cannot predict conversion rate lifts to the decimal point. The adjustment feature assumes you can. When you're wrong, and you will be, the algorithm amplifies your error rather than correcting it.
The Peak Is Flattening
Here's the part that makes the math even harder: the Black Friday conversion spike itself is shrinking. Among advertisers who didn't use adjustments, conversion rates rose 17.5% over baseline in 2022, 11.9% in 2023, and just 7.5% in 2024.
Shoppers have learned to spread purchases across a longer promotional season. Global-e's BFCM 2024 report found that conversion rate increases began three weeks before the holiday weekend, with sales momentum building throughout November. The traditional double-spike pattern (Black Friday, then Cyber Monday) is softening into a month-long plateau.
Anyone setting a seasonality adjustment based on what Black Friday "usually" does is likely overestimating the lift. The historical reference point keeps moving, and it's moving down.
Smart Bidding Already Knows
The core insight from the Optmyzr study is almost embarrassingly simple: Google already knows it's Black Friday.
BFCM is one of the most predictable spikes in global commerce. Google's bidding models are trained on vast amounts of historical data. They know exactly when Black Friday is, and they know consumer behavior shifts globally during this week. For advertisers who didn't apply adjustments, Smart Bidding consistently detected the conversion rate spike, increased bids appropriately to capture volume, and kept ROAS stable or improved it.
The algorithm behaved rationally without human intervention. The humans who intervened introduced volatility.
Google's own documentation hints at this. It recommends seasonality adjustments "only if you expect major changes to conversion rates" and notes that "Smart Bidding already manages seasonal events." The feature is designed for unpredictable moments: a flash sale your algorithm has never seen, a one-day promotion with no historical pattern. Black Friday is the opposite of that.
The Volume Trade-Off Is Shrinking
Adjustments did buy volume. In 2022, advertisers using them grew conversion value about 25 points more than everyone else. But by 2024, that advantage had shrunk to about 6 points, and it cost them a ROAS swing of more than 21 points.
That's a lot of efficiency to give up for a small amount of extra revenue. If your CFO is modeling CAC payback or gross margin contribution, the math doesn't work. You're paying a premium for incremental volume that's getting smaller every year while the efficiency penalty stays large.
Triple Whale's BFCM 2025 analysis showed Google Ads CPA climbed 34% year over year during the holiday window, while ROAS fell 21%. Rising acquisition costs are already putting pressure on returns. Adding a manual override that inflates bids further makes the problem worse.
Guardrails, Not Overrides
The alternative isn't passivity. It's setting constraints that let the algorithm work while protecting against runaway spend.
Budget pacing matters more than bid overrides. Optmyzr's BFCM guide recommends monitoring spend projection to estimate when you'll run out at the current burn rate, then adjusting before it happens. Last year, orders accelerated midday on Black Friday (around ) and again in the evening on Cyber Monday (around ). If your budget drains before those windows, you've lost the best hours.
Impression share lost to budget is the metric to watch. Grow My Ads recommends aiming for less than 2% budget-related impression share loss on top-performing Shopping campaigns. If you see double-digit percentage losses in last year's data, that's the fix for this year.
Margin-based rules can tighten or loosen Target ROAS/CPA using COGS inputs, pushing profitable items and limiting low-margin risk. This is a guardrail, not an override. You're telling the algorithm what success looks like, not telling it what conversion rates will be.
The Two-Week Pilot
If you're skeptical of the data, run a holdout. Split your campaigns into two groups: one with seasonality adjustments, one without. Measure CPC, ROAS, and conversion value for each cohort across the BFCM window. Compare against a pre-holiday baseline (the two weeks before Thanksgiving works well).
You'll have your own answer by the first week of December. If the adjustment group outperforms, you've learned something about your specific account. If it doesn't, you've avoided paying the premium next year.
The Optmyzr study isn't a randomized controlled trial. It compares two groups of advertisers, which isn't the same as a true A/B test. But the pattern held across three years and nearly 6,000 accounts. That's a strong signal.
The Board-Ready Version
For the executive summary: seasonality adjustments assume you can predict conversion rate lifts more accurately than Google's models. The data suggests you can't. The feature introduces volatility during the highest-volume, highest-stakes week of the year. The volume advantage it once provided is shrinking. The efficiency penalty is not.
The recommendation is simple. Leave Smart Bidding alone. Set budget guardrails. Monitor impression share. Run a holdout if you need proof. The algorithm already knows it's Black Friday.
