What Actually Happened in December
Every December, we run accounts through what should be their strongest period. And every January, we audit the results against what was actually profitable—not what the dashboards claimed.
This month was no exception. Here's what contradicted expectations, what looked good but won't hold, and what "worked" only because of seasonal distortion.
1. What Contradicted Expectations
High ROAS campaigns underperformed on contribution
The campaigns with the highest reported ROAS in December were often the worst performers on actual contribution margin.
Why? Returns.
Fashion and gifting categories saw 35-50% return rates on peak purchases. That 6x ROAS campaign? After returns, it was closer to 3x. For some brands, that's below breakeven.
The lesson: December ROAS is a fantasy number. January returns reveal the truth. Track net revenue, not gross.
Brand campaigns claimed too much credit
During peak, brand search volume spikes. Customers who've already decided to buy search your brand name. Your brand campaigns intercept them and claim the conversion.
The result: brand ROAS looks incredible while non-brand efficiency appears to decline. In reality, the non-brand campaigns created the demand that brand campaigns harvested.
The lesson: Don't let December brand performance justify January brand spend. The demand environment is fundamentally different.
2. What Looked Good But Won't Hold
PMAX efficiency during peak
Performance Max loves peak season. More signals, more conversions, more data for the algorithm to work with. PMAX campaigns that struggled in Q3 suddenly looked competent in Q4.
This won't last.
In January, conversion volume drops. Signal quality degrades. The algorithm starts making worse decisions with less data. Campaigns that looked "fixed" in December will regress.
The lesson: Don't set January targets based on December PMAX performance. Expect efficiency to drop 20-40% as demand normalises.
Aggressive ROAS targets
During peak, you can hit aggressive targets because demand exceeds supply. More buyers competing for your products means higher conversion rates and better efficiency.
In January, demand contracts. The same targets become unachievable. Campaigns that "worked" in December will throttle volume dramatically or miss targets entirely.
The lesson: Your December ROAS target is not your January ROAS target. Adjust for demand reality, not dashboard history.
3. What "Worked" Because of Seasonal Distortion
New customer acquisition
Gifting season brings one-time buyers. These "new customers" have the lowest LTV of any cohort you'll acquire all year. Many will never return.
If your December CAC looked acceptable, check the cohort retention data before celebrating. A £25 CAC for a customer who never buys again is worse than a £50 CAC for someone who repeats.
The lesson: December new customer metrics are misleading. Measure 90-day retention before drawing conclusions about acquisition efficiency.
Overall account efficiency
Blended ROAS during peak masks enormous variance. Some campaigns were genuinely profitable. Others burned money but got hidden in the aggregate.
January exposes this. Without the tide of demand lifting all boats, the weak campaigns become visible. The accounts that looked healthy in December often need restructuring by February.
The lesson: Peak performance creates false confidence. A thorough audit in January reveals what's actually working.
The Bridge From Peak to Reality
December is not a baseline. It's an anomaly.
The brands that do well in Q1 are the ones who:
- Audit before acting — Don't carry December assumptions into January strategy
- Track net metrics — Use post-return revenue and contribution margin, not gross ROAS
- Adjust expectations — Set realistic targets for normalised demand
- Question PMAX — The algorithm's December success doesn't predict January performance
If your December numbers look good and you're planning to scale, pause. Book an audit first. The reality check will save you money.
This post is part of our monthly observations series. We publish what we're seeing across ecommerce accounts each month—contradictions, surprises, and patterns that inform how we manage spend.