YouTube revenue is wildly seasonal and most creators discover this the hard way — usually in January, when their RPM falls 35–50% from the December peak and they think the algorithm broke. It didn't. Advertiser demand crashed. This guide is the full 12-month calendar of CPM swings, why they happen, and how to plan around them.
The annual RPM curve
For a typical Tier-1 US-heavy channel, expect this baseline pattern (relative to a 1.00× annual average):
- January: 0.55–0.70× — the worst month of the year. Advertisers reset budgets, retail spending collapses post-holidays.
- February: 0.70–0.80× — recovery starts, slowly.
- March: 0.85–0.95× — Q1 close pushes advertiser spend, normal levels return.
- April–May: 0.95–1.05× — steady state. Tax season ends, B2B demand stable.
- June: 1.00–1.10× — Q2 close, summer ad pushes.
- July–August: 0.85–0.95× — summer dip. Brand teams on vacation, fewer agency campaigns.
- September: 1.05–1.15× — back-to-school, Q3 close.
- October: 1.10–1.25× — Q4 ramp begins. Holiday campaigns start booking inventory.
- November: 1.30–1.50× — Black Friday / Cyber Monday peak. Retail bidding war.
- December: 1.40–1.70× — highest RPMs of the year. Holiday shopping plus year-end "use it or lose it" budget burn.
Why Q4 spikes
Three forces compound. First, retail brands spend 35–45% of their annual digital budget in October–December against holiday sales. Second, agencies burn unspent budget in December rather than return it to clients. Third, political ad spend in election years adds a fourth bidder into every auction in October–November.
The auction model means more bidders = higher clearing prices for every impression. Your video didn't change. The competition for showing an ad against it doubled.
Why January crashes
Retail spending evaporates the week after Christmas. Agencies haven't yet deployed new-year budgets — those typically clear approval and start bidding in mid-to-late February. The result is a hard CPM floor where most impressions clear at the bumper-ad reserve price rather than competitive skippable rates.
This isn't a YouTube-specific phenomenon. The same pattern shows up across Meta, programmatic display, podcast networks — anywhere advertisers buy inventory.
Planning around the calendar
- Front-load uploads in October–November. Videos posted in Q4 get the seasonal bid lift on day one. The same upload in January earns 40–50% less for the same view count.
- Evergreen content peaks in Q4 too. A video posted in March that still gets views in December earns December RPMs on those views — seasonal lift applies to all live inventory, not just new uploads.
- Cash-flow plan for the January cliff. AdSense pays one month in arrears, so December earnings hit in late January — masking the January cliff for one month. The February deposit is the painful one. Hold reserves.
- Don't compare YoY without seasonal adjustment. A January-vs-September comparison is meaningless. Always compare same-month YoY.
Niche modifiers
Some categories smooth the curve, others amplify it.
- Personal finance / tax: January–April spike (tax filing season), softer Q4. Inverts the normal curve.
- Fitness / weight loss: Massive January spike (resolution advertisers). The one niche that prefers January.
- Gaming: Sharp November–December peak (console / AAA launches), then severe January crash.
- Education / EdTech: August–September back-to-school peak, plus January (new-year skill goals).
- Travel: January–March (winter escape booking) and May–July (summer planning) — two annual peaks.
Run your own numbers through the YouTube Revenue Calculator with December RPM at the upper end of your niche range and January at the lower — the spread is usually 2–2.5×, and forecasting against the midpoint is what causes the "January surprise" every year.