The average CPC gap observed worldwide between the Black Friday peak (week of November 28, 2026) and the mid-August trough reaches 42 to 58% across most e-commerce verticals in public Google Ads benchmarks 2025 (WordStream Industry Benchmarks, Search Engine Land Q4 reports). In the US, BF/CM is the biggest e-commerce peak of the year (+120 to +200% on week 47). In the UK, the peak combines Black Friday (heavily imported since 2014) + Boxing Day (December 26, traditional). In continental Europe, BF/CM took hold between 2019 and 2025 but coexists with regulated sales periods (FR, ES, IT). On B2B SaaS, the gap is reversed but of similar magnitude: the month of August sees CPC drop 25 to 40% from a demand deficit, more pronounced in Europe than in the US. Google Ads seasonality is neither a detail nor stable data from one sector to another — it's the number-one variance factor on full-year ROAS, ahead of copy quality and just behind tracking quality.
Yet most accounts at the start of the seasonal cycle apply a constant monthly budget and a fixed Target CPA target over 12 months — as if demand didn't vary. The result: overspending during troughs (high CPC from stable competition against low demand), underspending during peaks (volume left on the table), Smart Bidding entering chaotic learning at every poorly prepared transition. This guide settles it: worldwide map of 2026 peaks and troughs (US, Europe, other markets), month-by-month calendar by vertical, Smart Bidding mechanics (seasonality adjustments vs budget pacing), BF/CM strategy, and summer arbitrage by hemisphere. For general budget mastery, see our Google Ads budget pacing guide. To compare your CPC against 2026 medians by vertical, our CPC calculator delivers the result instantly.
Worldwide seasonality 2026: US + Europe + other markets
Google Ads seasonality refers to all the cyclical variations (annual, weekly, monthly) of the demand / acquisition cost pair observed on an account. It is measured at three levels: variation in search volume (Google Trends signal), CPC variation (competition signal), conversion rate variation (purchase intent signal). The three don't always move in the same direction — that's what makes the arbitrage non-trivial.
In the US, the e-commerce seasonal calendar is dominated by six structuring peaks: (1) Memorial Day (last Monday of May) opens the summer season; (2) 4th of July (July 4) — big peak apparel/outdoor/grilling; (3) Back-to-school (July-August, earlier than in Europe); (4) Labor Day (first Monday of September); (5) Halloween + Thanksgiving (late October - 4th Thursday of November); (6) Black Friday + Cyber Monday + Cyber Week (week 47-48 = absolute peak). No legally regulated sales periods — each retailer sets its own dates. US Q4 retail (October-December) captures 30 to 40% of annual revenue in most e-commerce verticals. Source: Search Engine Land Paid Search Reports.
In the UK, the calendar is hybrid: Black Friday has gained strongly (imported 2014, dominant since 2018), but Boxing Day (December 26) remains the biggest historic retail day (high-street + online). UK summer sales (mid-July to early August) and January sales (post-Christmas) are traditional. Bank Holidays structure the microcycles (Easter, May Day, Spring Bank Holiday, Late Summer Bank Holiday).
In continental Europe, the sales regulation differs by country: France (2nd Wednesday of January + last Wednesday of June, 4 weeks), Spain (rebajas semi-liberalized since 2012), Italy (regional saldi), Germany (no imposed date — Winterschlussverkauf end of January, Sommerschlussverkauf end of July by tradition). Black Friday / Cyber Monday gained strongly in continental EU between 2019 and 2025 but coexist with traditional sales. The school calendar structures tourism peaks (3 main waves: spring, summer, All Saints in FR; Pasen, zomer, herfst in NL; etc.). The August effect is more pronounced in continental Europe than in the US or UK: most B2B companies are at reduced capacity from July 25 to August 25, creating a 35 to 55% trough on B2B SaaS leads and pro services.
Other key markets. Brazil: Black Friday (3rd or 4th Friday of November — less dominant than US but in strong growth), Carnaval (February-March, leisure/travel peak), Mãe (May, Mother's Day). Australia / NZ: southern hemisphere — reversed seasons, summer sales in December-January, back-to-school in January, Black Friday massively adopted since 2020. Japan: Golden Week (April 29 - May 5, big peak), Obon (mid-August), Oshogatsu (January 1-3). India: festivals (Diwali in October-November = dominant Q4 + festive season Karwa Chauth, Dussehra), Republic Day (January 26).
On the vertical axis, three major worldwide seasonal profiles stand out. Profile A — generalist e-commerce: January peak (post-holiday clearance / EU sales) + July peak (summer sales / EU sales) + November peak (BF/CM) + December peak (holiday gifting). Four waves, relatively stable demand between. Profile B — B2B and SaaS: December trough (year-end closing) + August trough (pro vacations more pronounced EU than US) + March-June peak (H1 budgets spent) + September-November peak (H2 budgets spent). Bimodal pattern with two strong seasons. Profile C — local services and contracting: March-October peak (weather-dependent working season), November-February trough (weather + closing). Long unimodal pattern, sensitive to actual weather. Our MER (Marketing Efficiency Ratio) calculator measures overall marketing efficiency, not channel-by-channel.
To understand how Smart Bidding integrates these variations into its predictions, see our Smart Bidding Maximize Conversions vs Target CPA guide.
Month-by-month calendar: peaks and troughs by vertical
Here is the operational month-by-month calendar across the three major profiles, with US, UK, continental EU columns. Figures from public Google Ads benchmarks 2025-2026 (WordStream, Search Engine Land), orders of magnitude of demand variation vs smoothed 12-month average.
Three important readings. For US e-com, BF/CM (week 47-48) is the absolute peak of the year — Thanksgiving + Black Friday + Cyber Monday + Cyber Week represent 8 to 14% of annual revenue across 7 days. US brands typically allocate 25 to 35% of the Q4 budget on this single week. In the UK, BF gained strongly since 2014 but Boxing Day remains the absolute single-day peak (December 26, online + high street). In continental EU, BF/CM coexists with regulated sales — allocate 35 to 45% of Q4 budget on November BF, 30 to 35% on December, the rest split between October and January winter sales.
On worldwide B2B/SaaS, the August effect is massive but variable by region: continental EU (FR, IT, ES, BE) -40 to -55%, northern EU (DE, NL, SE, DK) -25 to -40%, UK -20 to -35%, US -15 to -30% (vacation less systematic in August in the US). On most observed B2B accounts, August remains a clear trough with demand falling but CPC not dropping proportionally (competitors who stay on share the demand deficit). Cutting 40 to 70% of the August budget is almost always the right arbitrage by region, except in very specific cases (a company whose target stays active in August, e.g. schools before back-to-school).
For local services and contracting, seasonality is dictated by weather — similar pattern in the northern hemisphere (US Sun Belt vs Snow Belt, UK, continental Europe). The March-October peak is almost systematic on outdoor work (roofing, siding, gardens, pools), but varies by 4 to 6 weeks year over year depending on actual weather. Rather than locking in a rigid calendar, monitor Google Trends on the 2-4 main keywords and adjust the budget weekly following the demand curve. Southern hemisphere (AU, NZ, southern BR, AR): reversed pattern — December-March peak on outdoor work.
Smart Bidding seasonality adjustments: when to use them
The Google Ads seasonality adjustment (Tools > Bid Strategies > Advanced controls > Seasonality adjustment) is an explicit signal you give Smart Bidding to inform it of an expected variation in conversion rate over a short period (1 to 14 days), compared to its learned behavior of the last 30 days. It is typically used for: a non-recurring promo event (7-day flash sale, limited brand operation), a one-off structural event (product launch), or a predictable variation not yet learned by the algorithm (first Black Friday for a recent account).
The exact mechanism
You tell the algorithm two things: the time window (start date, end date, max 14 days) and the percentage of expected conversion rate adjustment (-90% to +900%). Example: Black Friday promo from November 28 to December 1, +150% conv rate adjustment (you expect 2.5x more conversions per click than usual). The algorithm integrates this signal into its predictions and bids accordingly — typically more aggressively to capture volume.
The official documentation details the constraints: maximum 14 days, applicable only to Maximize Conversions, Maximize Conversion Value, Target CPA, Target ROAS. Doesn't apply to Manual CPC or Maximize Clicks.
When to use it and when to avoid it
Use it: short event (under 14 days), identified promo with quantifiable conv rate variation (you know from experience that your Black Friday boosts conv rate by 130%), new event for the account that the algorithm hasn't seen in its 12-month history.
Avoid it: long variation (full month like January sales or August trough) — use budget pacing instead. Variation already known to the algorithm (an account with 24 months of history on the same calendar — the algorithm has already learned). Combination with manual budget adjustment on the same event — double signal that confuses the algorithm.
The simple discrimination rule. If the event lasts less than 14 days and represents a quantifiable conv rate change, use Smart Bidding seasonality adjustment. If the event lasts more than 14 days (full month, season) and represents a structural demand variation (search volume changes, not just conv rate), use manual budget pacing and let Smart Bidding adapt via its continuous learning. Combining both on the same event brings nothing — the algorithm receives a contradictory double signal and either ignores one, or overreacts. On accounts observed in public benchmarks, accounts that combine both display a ROAS variance during the event 18 to 32% higher than those that use only one.
Manual budget pacing: rules before/during/after peak
Manual budget pacing remains the main tool for managing long seasonal variations (month, season). It complements Smart Bidding, not competes with it: budget defines the available envelope, Smart Bidding optimizes allocation within. Three operational rules to manage an identified peak.
Rule 1 — Pre-peak phase: ramp up progressively from D-14 to D-7
The peak is not an instantaneous event. The pre-purchase research phase starts 7 to 14 days before the peak in most verticals (BF/CM, sales, end of year, back-to-school). During this phase, users compare, add to cart without buying, save products — conv rate temporarily drops but deferred ROAS rises if tracking properly attributes multi-session conversions.
Recommended budget pattern: D-14 = standard budget. D-10 = +20% from standard. D-7 = +50% from standard. D-3 = +80% from standard. D0 (peak start) = +100 to +150%. This progressive ramp gives Smart Bidding time to learn the new dynamic without forcing a chaotic learning phase. A brutal rise from D-1 to D0 forces the algorithm to relearn at peak demand, losing 18 to 28% of performance over the first 3 days of the peak.
Rule 2 — Peak phase: hold the budget, change nothing
During the peak itself (D0 to D+ peak duration), the worst mistake is to modify the Smart Bidding target (CPA target or ROAS target) or change the bidding strategy. Any change of more than 15% restarts 3 to 7 days of learning. If you change in the middle of BF week, the algorithm spends the last 3 days of the peak in learning phase — wasted.
The only lever to activate during the peak: the daily budget itself. If you see your campaigns consuming 100% of the budget every day and the limited by budget alert lights up, raising the budget by 20-30% gives the algorithm more room. Don't touch the target. Don't add keywords. Don't change copy. Absolute stability of everything that isn't budget.
Rule 3 — Post-peak phase: progressive descent and cleanup
The post-peak (D+1 to D+7 after peak end) is as critical as the pre-peak. Two simultaneous phenomena: demand drops sharply (search volume divided by 2 to 4 within 48h after peak end), but competition stays high 3-5 days more (slow-to-adjust competitors keep their bids). If you leave the budget inflated, you pay high CPCs on collapsing demand — catastrophic post-peak ROAS.
Pattern: D+1 = peak budget / 1.2. D+3 = standard budget. D+5 = standard budget - 10% (slight sub-peak to recover margin). D+7+ = return to standard budget. Also use the D+3 to D+7 window to do post-peak cleanup: exclude parasitic search terms collected during the peak, pause keywords that underperformed, adjust manual bids if on hybrid strategy.
For an overview of CPA vs ROAS arbitrages during seasonal variations, see our Target ROAS vs Target CPA comparison.
Black Friday and Cyber Monday: the 2026 strategy
Black Friday 2026 falls on November 27 (day after US Thanksgiving), Cyber Monday on November 30. In the US, the extended BF/CM window (Cyber Week) generally covers Thanksgiving (November 26) → Cyber Monday (November 30) → Cyber Tuesday/Wednesday early December. According to public Google Ads benchmarks 2025-2026 (WordStream Q4 reports, Search Engine Land Cyber Week analysis), US BF/CM concentrates 8 to 14% of annual e-commerce revenue across 7 days, the highest demand concentration of the year. In the UK, BF (heavily imported since 2014) coexists with Boxing Day (December 26, online + high street absolute single-day peak). In continental Europe, BF/CM concentrates 12 to 22% of Q4 revenue across 7 days, the second peak after January winter sales.
The 4-phase strategy
Phase 1 — Preparation (D-30 to D-15): Merchant Center feed audit (BF disapprovals are extremely costly), pre-load Performance Max asset groups with BF creative (visuals, headlines, descriptions), tracking validation and Enhanced Conversions, total available BF/CM budget calculation and breakdown. During this phase, activate nothing on the campaign side — just prepare.
Phase 2 — Pre-peak (D-14 to D-3): progressive budget ramp-up (+30% D-14, +60% D-7, +90% D-3). Activation of Smart Bidding seasonality adjustments on BF campaigns if account is recent or conv rate variation is very different from the last BF. Possible launch of a 7-day teaser brand operation to capture the research phase.
Phase 3 — Peak (D0 to D+3, i.e. November 27-30): budget +120 to +180% from standard. Don't touch anything other than the budget. Monitor the limited by budget alert every 4-6h. Monitor feed approval rate (a technical crash during BF costs 3 to 5 times more than an off-peak crash).
Phase 4 — Tail and cleanup (D+4 to D+10): progressive descent to standard over 5 days, exclusion of parasitic search terms, cleanup of underperforming keywords during the peak, ROAS audit by campaign to identify losses (often: a poorly calibrated PMax asset group that burned 30% of the peak budget without converting).
The 3 mistakes that wreck a BF
Mistake 1 — Modifying the Smart Bidding target during the peak. The stressed marketer's instinct is to lower the CPA target when seeing CPA rise during BF. That's exactly the opposite of what you should do: during a peak, conv rate rises (so CPA would drop at equal bid), Smart Bidding can therefore bid higher without degrading the target CPA. Lowering the target cuts delivery mid-peak. Hold, even raise by 10-15% if you tolerate a bit more CPA to capture more volume.
Mistake 2 — Activating Performance Max without brand exclusion. PMax e-commerce without a brand exclusion list buys back your brand during BF, inflates its apparent ROAS (brand conversions are expensive in organic, so easy to buy back), and the global account CPA explodes. Activate PMax brand exclusions before any peak.
Mistake 3 — Cutting the BF+1 budget. November 28, 2026 (Saturday) remains an important peak-tail day (15-25% of BF volume). Cutting the budget on the 28th morning leaves this tail to the competition. Hold the budget +100% until Monday, November 30 midnight.
Summer season (July-August): trough or opportunity?
The received wisdom that "summer is slow and you should cut the budget" is one of the costliest seasonal mistakes we see in audits. The reality is more nuanced and depends heavily on the vertical.
July: e-commerce and tourism peak, B2B trough
July is dominated in the US by 4th of July weekend (July 3-5, apparel/outdoor/grilling peak +30 to +50%) and Amazon Prime Day (mid-July, triggers cross-retail competition). In the UK, summer sales mid-July (+35%). In continental Europe, legal summer sales (FR last Wednesday of June to last Wednesday of July, similar in other countries) push e-commerce demand to +50 to +75% across most generalist verticals. Worldwide, massive peak on tourism (+80 to +120% search volume hotels, rentals, activities), restaurants in tourist zones (+30 to +50%), outdoor sport (+15 to +25%). For these verticals, July is a budget-protect month.
Conversely, B2B and SaaS sees demand drop from July 15-20, with a trough that sets in and reaches its low point mid-August. For these verticals, progressively cutting the budget from D-15 to D-1 August makes sense.
August: massive B2B trough, tourism and non-B2B services peak
August is the most polarized month of the year worldwide. On public Google Ads benchmarks 2025-2026, the patterns observed:
- US Back-to-school: massive peak +50 to +80% on kids apparel, school supplies, student electronics. The season starts mid-July in the US (before Europe), extends to Labor Day (September 1, 2026).
- B2B and SaaS worldwide: continental EU -40 to -55%, northern EU (DE, NL, SE) -25 to -40%, UK -20 to -35%, US -15 to -30% (less systematic vacation). Cutting 40 to 70% of the budget is almost always profitable by region.
- Generalist e-commerce fashion/home equipment: continental EU -15 to -25% volume, US/UK lukewarm demand (-5 to -15%). Maintain standard budget or lower by 10-20%.
- Tourism, vacation rentals, restaurants in tourist zones: peak +50 to +120% worldwide in northern hemisphere. Raise budget by 30 to 80%.
- Outdoor sport (cycling, hiking, camping, water sports): peak +20 to +40% in northern hemisphere. Maintain standard budget or +20%.
- Childcare / school supplies: peak +20 to +30% end of August (back-to-school). Raise budget +30%.
- Local services / contracting: peak slowed by Europe vacations but stays positive. US smoother. Maintain standard budget.
Common mistake: applying the "August trough" reflex to all verticals without auditing the actual pattern over 24 months of account history. On poorly audited accounts, that's typically 18 to 32% of conversions left to competition in August on verticals where August is actually a peak.
Before cutting the August budget on a sectoral reflex, look in Google Ads at the conversion pattern of the past 24 months on the relevant campaign. Tools > Reports > Performance > segment by month. If August of the last 2 years shows conversion volume greater than or equal to the smoothed 12-month average, don't cut. If volume is 25 to 50% below, cut proportionally. If volume is more than 60% below, cut drastically but keep a minimum presence (10 to 20% of the budget) to avoid losing quality score position on strategic keywords.
Common seasonal management mistakes
Five recurring mistakes we see on most public benchmarks, that collectively cost 15 to 30% of annual ROAS.
Mistake 1 — Constant monthly budget over 12 months. The account that doesn't vary its Google Ads budget by season systematically overspends during troughs and underspends during peaks. On an e-commerce account at $100,000/year (equivalent ~€92,000 / £80,000), the performance gap between flat budget vs seasonalized budget reaches 18 to 32% of observed annual ROAS.
Mistake 2 — Smart Bidding target frozen all year. The optimal CPA or ROAS target varies by season: more diluted margin in sales, higher average basket end of year for gifts, higher conversion rate at peak, lower at trough. Keeping the same target over 12 months forces Smart Bidding to compensate via volume — often by cutting delivery during troughs.
Mistake 3 — Modifying the target mid-peak. See sections 4 and 5. Any target modification of more than 15% during the peak restarts 3-7 days of learning and degrades peak performance.
Mistake 4 — Combining seasonality adjustment and manual budget on the same event. See section 3. Contradictory double signal, the algorithm gets confused.
Mistake 5 — Forgetting to return to baselines after the peak. The account that pushes +120% budget during BF and forgets to come back down to standard on D+10 pays 2 to 3 weeks of overspending in the peak tail. Configure a calendar alert or an automated rule for return to baselines.
For accounts that want to automate this seasonal management without depending on a manual checklist, our SteerAds auto-optimization engine detects seasonal patterns over 24 months of account history, automatically applies pre/during/post peak budget pacing, adjusts Smart Bidding targets by season, and triggers seasonality adjustments when appropriate — without restarting an unmastered learning phase.
For multi-brand or multi-country accounts that need to orchestrate seasonality across multiple Google Ads accounts simultaneously, also see our multi-account MCC strategy. For online training providers (vertical with its own specific seasonality: September back-to-school peak, January new-resolutions peak), see our Google Ads online training guide.
France regional case: regulated sales + amplified August B2B effect
France is an interesting regional case in Europe for two reasons. First, sales are strictly regulated by law: 4 weeks in winter starting on the 2nd Wednesday of January, 4 weeks in summer starting on the last Wednesday of June. This creates almost-deterministic peaks on the e-commerce side that we also find but less rigid in ES (semi-liberalized rebajas) and IT (regional saldi). Second, the August B2B effect is more pronounced than in the US or UK: most B2B companies are at reduced capacity from July 25 to August 25, creating a 35 to 55% trough on B2B SaaS leads and pro services. For B2B accounts operating both US and FR, plan for an August budget asymmetry: maintain 70-85% of US budget, cut 40-60% of FR budget. Similar case for IT and ES, less pronounced for DE and UK.
Worldwide Google Ads seasonality is not a detail to manage on the side — it's the main lever of ROAS variation over the year, comparable in impact to a campaign structure overhaul or a major bidding strategy change. Mastered, it brings 15 to 30% additional annual ROAS at equal structure. Ignored, it silently drags down performance and forces your Smart Bidding to carry a weight it shouldn't have to carry alone — also see official Google Ads documentation for more details.
To go further, also see our guides Google Ads local restaurant, Google Ads legal services worldwide, Google Ads real estate lead gen.
Sources
Official sources consulted for this guide:
FAQ
Should I use Smart Bidding seasonality adjustments or adjust the budget manually?
Both tools serve different purposes. The Smart Bidding seasonality adjustment (Google Ads UI > Tools > Bid Strategies > Advanced controls) is meant to signal to the algorithm an expected variation in conversion rate over 1 to 14 days — useful for a short, well-defined event (promo launch, 7-day brand operation). Manual budget pacing serves to anticipate a structural variation in demand (entire month, season) that doesn't reduce to a change in conv rate. Practical 2026 rule: seasonality adjustment for one-off peaks under 14 days with identified promo; manual budget pacing for underlying variations (back-to-school, sales, summer season). Don't combine both on the same event — the algorithm gets confused.
How many days before a peak should I raise the budget?
On accounts observed in public Google Ads benchmarks 2025-2026, the optimum sits between 7 and 14 days before the identified peak, with a progressive ramp-up of 30 to 80% from the initial budget. Raising too late (D-3 or less) doesn't give Smart Bidding time to stabilize its signals and the algorithm enters learning phase during the peak — wasted. Raising too early (D-30) burns budget without return. For Black Friday specifically, the D-14 to D-7 window captures the pre-purchase research phase, D-3 to D0 captures triggered purchases, D+1 to D+5 captures the post-peak tail. Allocate 60% on the pre-peak phase, 30% on the peak, 10% on the tail.
How do I manage Smart Bidding during post-Christmas peaks (US Boxing Week / UK Boxing Day / EU sales) in 2026?
In the US, the Boxing Week window (December 26 - January 2) remains a peak for apparel and electronics. In the UK, Boxing Day (December 26) is one of the biggest retail days of the year. In continental Europe, winter sales start on dates set by law (FR 2nd Wednesday of January, ES January 7, IT January 5, DE no imposed date but Winterschlussverkauf at end of January). Across these 4-week windows, two approaches. Mature account with stable Smart Bidding history of 6 months: leave Target ROAS or Target CPA running, the algorithm has learned the previous years' seasonality (Smart Bidding integrates rolling 12-month data). Adjust the target by 5 to 10% to account for promo margin dilution. Recent account: switch to Maximize Conversions for the 4 weeks, time enough to capture volume without CPA constraint, then return to Target CPA mid-February.
Does the August trough really penalize all e-commerce verticals?
No, that's the classic mistake especially in the northern hemisphere. On public Google Ads benchmarks 2025-2026, August is a trough for B2B (-30 to -55% volume vs average, more pronounced in continental Europe than in the US), renovation/contracting services (-30%), e-com fashion/home equipment (-20%). But August is a PEAK for: back-to-school (US +50 to +80% over July-August on kids apparel and supplies), tourism (+80% volume Europe, US road trips +30%), restaurants in tourist zones (+30 to +50%), e-com outdoor sport (+15 to +25%), vacation rentals. The rule: there is no universal off-season, there are verticals that shift their peak. Audit the monthly pattern of the past 24 months before cutting the August budget — many brands leave conversions on the table by pre-supposing a trough that doesn't exist for them.
Should I use Accelerated delivery during a peak?
No — Google deprecated Accelerated delivery for nearly all strategies in 2020. Today, only Standard delivery exists on Smart Bidding, which distributes the budget intelligently throughout the day according to the predicted hourly conversion. To accelerate budget consumption during a peak, two real levers: 1) increase the daily budget (Smart Bidding has a daily spend cap at 2x the configured budget, so raise the cap), 2) increase the CPA target or lower the ROAS target to allow the algorithm to bid higher. The nostalgia for Accelerated delivery comes from the manual CPC era — it has no place in 2026 Smart Bidding.