Black Friday Google Ads: A Week-by-Week Plan for Stores

First check that November is your peak. In GetProfit data on 96 stores (July 2025 – June 2026), it was among the three best months for 53 and not for 43. If it is yours, plan stock three to five months ahead. One to two months out, fix the feed, bring back missing bestsellers and grow budgets in small steps. Two weeks out, our rule is to raise budgets by 20–30% and lower target ROAS by 10–15%. At the peak, leave structure and targets alone. Afterwards, roll back in steps.

In 2026, Black Friday falls on 27 November and Cyber Monday on 30 November, so early October leaves about eight weeks. This guide covers those weeks and the month after them, for stores whose ads run mostly through Performance Max, with or without separate Shopping campaigns. Black Friday is one case of a wider pattern, and our guide to seasonal demand in ecommerce shows how to measure demand across your store’s whole year.

Is Black Friday really your peak?

Not for every store. In our data, November was among the three best months for 53 of 96 stores and not for the other 43. For each store, we took ad revenue over one seasonal cycle and removed the store’s own growth trend. Then we checked which months were among its three best and three worst.

GetProfit data, 96 stores, July 2025 – June 2026. Revenue here is what the ads report, not the store’s total sales.

November and December are the most common peak months, and they come one after the other. So the weeks after Cyber Monday can matter as much as the run-up to Black Friday. For the 43 stores without November in their three best months, a plan copied from the retail calendar puts the biggest push into a weaker month.

So check your own seasonality before you plan. Line up last year’s months by ad revenue, then split them by category: a store that looks flat can hide a category with a sharp November peak. If November is outside your three best months, our guide on whether Black Friday is worth it helps you decide how much, if anything, to put into the sale.

These numbers have limits. They cover one observed cycle, so a strong month may be a one-off event rather than a pattern. For the same reason, we make no year-over-year comparisons on this data.

The Black Friday plan, week by week

We plan every seasonal peak in four phases: an early warning three to five months ahead, active preparation one to two months ahead, the peak itself and a review afterwards. The dates in this plan are for Black Friday on 27 November 2026.

1. By mid-October: confirm the peak for the store and for each category. If your best month is December, or a category peaks in October, shift these dates to match.
2. October: list last season’s peak bestsellers and check each one. Is it in stock, approved in Merchant Center and inside a campaign? A product that sold well last November and is now missing is the cheapest fix of the season. Our guide to getting last year’s bestsellers back shows how to find them and return them to your ads before the sale.
3. October: clean the feed for the products you will push. Start with the products that carry the peak: fix disapprovals, empty fields and wrong prices there before the rest of the catalogue.
4. September to early November: grow the campaigns in small steps. For a November peak, our rule is to start in late August or September; if you haven’t, start now. Raise budgets by up to 20% at a time and judge each step after two weeks. Campaigns that grow before the peak meet it with fresh conversion data.
5. Early November: load the offers. Add sale prices with start and end dates to the feed, send Merchant Center promotions early enough for Google’s review, and prepare promotion assets for the ads.
6. From 13 November, two weeks out: take the pre-season step. Raise daily budgets by 20–30% and lower target ROAS by 10–15% in one planned move.
7. 27–30 November: check every day, change little. Watch spend, stock and disapprovals. Use a short-event tool only if you expect conversion rate to jump far beyond normal. For what to look at each day and which changes should wait, see monitoring campaigns during Black Friday week.
8. From 1 December: roll back in steps, unless December is your peak too.
9. Mid-December: judge the changes, then review the season. A change made on Black Friday reaches two full weeks of data around 11 December.

For steps 2 and 3 in detail, using last season’s data rather than a generic list, see the Black Friday checklist for Shopping and Performance Max.

Why running out of budget on Black Friday costs more than in a slow month

For most stores, money spent in the peak months brings back more than money spent in the slow ones. In GetProfit data on 96 stores (July 2025 – June 2026), the median store’s ROAS in its three worst months was 0.70 of its ROAS in its three best. Even at the 75th percentile, which only a quarter of stores exceeded, the ratio was 0.86, still below 1. A budget that runs out on Black Friday cuts spend that pays back better than spend in the slow months.

That is why the pre-season step, two weeks before the peak, raises the daily budget before demand arrives rather than after the campaign shows “Limited by budget”. According to Google’s API documentation (Campaign budgets overview), a campaign may spend less than its average daily budget on some days and more on others. Over a month, though, it pays no more than the average daily budget times 30.4. A budget sized for October leaves little room on the busiest days of the year.

More budget buys more only when the budget is what holds a campaign back. In GetProfit data on 125 stores (June 2025 – June 2026), the median share of search impressions lost to rank was 50.2%, and the median share lost to budget was 13.6%. In 110 of the 119 stores where we could measure both, rank was the bigger limit.

Impression share describes the pool a campaign is already eligible for, which the catalogue and the budget have narrowed. It is not the size of the market, and for Performance Max it is incomplete.

A bigger budget also does little when the target is what holds a campaign back. That is why the pre-season rule moves the budget and the target ROAS together.

The target has mattered even more since 17 August 2026. On that date Google began a change to bidding for campaigns limited by budget, rolling out over a few weeks. Its announcement (Bidding and Budgeting Updates to Scale Your Growth) says the change is meant to help these campaigns deliver more predictable performance in line with their targets. Our article on why Google Ads targets no longer overperform explains what that means for a store with a loose target.

Clicks may cost more at the peak as well. Google Ads CPC in Q4 checks whether they do in November and December, and whether that should change your budget or target.

All these ROAS figures are based on revenue, not profit, because our data has no purchase prices. A Black Friday discount also changes the profit side: it raises the ROAS a store needs to break even.

How big a step, and how often?

Small, planned steps keep Smart Bidding learning on clean data. Here are the rules we work by, next to Google’s own limits.

To decide how many weeks ahead to start and how to split the increase into steps, see raising your budget before Black Friday.

Small steps matter because of learning. Google Ads Help (Duration of the learning period for campaigns and what affects it) lists what puts a bid strategy into “Learning”: a new or reactivated strategy, a changed setting, or campaigns, ad groups or keywords added to or removed from it. In Shopping campaigns, an ad group target change can trigger it too. Google adds that calibration can take a few conversion cycles, typically one to two, though it can be faster with more conversion data. A target change in the last days before Black Friday leaves no time for that learning period to pass.

Our guide to target ROAS for Black Friday covers which way to move the target, by how much and when to put it back. For the same step sizes outside the season, see scaling Google Ads without losing ROAS.

Seasonality adjustment, promotion mode or a campaign total budget?

Google offers three tools for a short sale, and they move different levers. They are built for short events, not for a whole season. Google’s API documentation (Create seasonality adjustments) calls seasonality adjustments an advanced tool for telling Smart Bidding to expect a change in conversion rate during an upcoming event. It says they are ideal for short events of 1–7 days.

Google announced promotion mode as a beta for Search and Performance Max campaigns on 15 June 2026. The third tool, a campaign total budget, sets one sum for a fixed period instead of a daily amount.

A few details decide how each one behaves in practice:

• Seasonality adjustment. Set it only for the days the sale runs, and size it from your own conversion rate on the same days last year. Our rule caps an adjustment at 14 days, the point after which Google says it may not work as well. Our article on seasonality adjustments in Google Ads shows when an adjustment helps and when it does harm.
• Promotion mode. Google’s announcement says the beta lets you schedule temporary changes to the ROAS tolerance and add daily budget during peak periods. You can also combine its ROAS tolerance adjustment with campaign total budgets. To join the beta, Google asks you to contact your account team. According to the trade site PPC Land, Ginny Marvin, Google’s Ads Product Liaison, wrote on LinkedIn that setup starts with the promotion’s start and end dates. The span is 3 to 14 days, so the mode switches off by itself. PPC Land also reports that it can run alongside a seasonality adjustment, though Google’s documentation does not say how the two interact.
• Campaign total budget. Google’s API documentation (Campaign budgets overview) lists it for Search, Standard Shopping and Performance Max campaigns, among others, with set start and end dates. You can’t change the budget type after you create a campaign, so it fits a new sale campaign rather than an existing one.

Our article on Google Ads promotion mode compares which of the three fits a 3–14 day sale and whether to combine them.

What the feed needs before the sale

In Shopping campaigns and Performance Max, the offer reaches shoppers through the feed, so the sale has to be in the feed before it starts.

Sale price with dates. Add a sale price with its effective dates, so the price switches on and off by itself. In Google’s Merchant API reference (ProductAttributes), the sale price effective date is the date range during which the item is on sale. Give the start and end with a time and your time zone, so the sale starts and ends at your local midnight.

Merchant Center promotions. Promotions add a special offer link, such as “15% off” or “Free shipping”, to your products on Google Search, Shopping and Chrome (Overview of Promotions sub-API). To run them, you need a product data source and a promotions data source in Merchant Center, a Google Ads account for Shopping campaigns and enrolment in the Promotions program. Google reviews and approves each promotion before showing it, so send your promotions before the pre-season step rather than on the day.

Promotion assets and countdowns. These sit in the ads rather than in the feed. Our guide to promotion assets and countdowns shows how to add them and which campaigns show them.

What to leave alone during the peak

For stores with a November peak, the days around Black Friday bring the most valuable traffic of the year. They are the wrong time for anything that sends a campaign back into learning.

• The campaign structure. Our rule: no splitting, merging or new asset groups in the peak weeks. Google also lists campaigns, ad groups or keywords added to or removed from a bid strategy as a reason for “Learning”. Settle the structure in October.
• The target ROAS, upwards. Our rule is never to raise it at the peak. A changed setting is another of Google’s reasons for “Learning”.
• The bid strategy. A new or reactivated strategy also puts the campaign into “Learning”.
• A category that goes quiet just before the sale. Keep it running: a slow week before the peak is part of the season, not a verdict on the category.
• Data exclusions for the sale. Google’s API documentation (Create data exclusions) describes them as a tool for dates with conversion tracking problems, meant for outages or major issues. It warns that using them often or for long periods can hurt Smart Bidding. If tracking breaks on Black Friday, that is when to use one.
• Judging fresh changes. In its first 7 days, a change shows mostly the learning, not the result.

After Cyber Monday: December, January and the roll-back

For many stores the peak runs past BFCM, the days from Black Friday to Cyber Monday. December was among the three best months for 50 of the 96 stores in our data (July 2025 – June 2026), almost as often as November. January was among the three best for 31 stores and among the three worst for 20. So on 1 December, check your own calendar before you undo anything.

Where demand does fall, undo the pre-season step in the same small steps you used to build it, and watch how deep the cut goes. In our data, 61 of 96 stores spent less than 70% of their peak-month spend in their worst months. Demand does fall in the slow months, since ROAS there was 0.70 of the peak in the median store. But most stores cut the budget as well, so part of the slow season can be self-inflicted. Our guide on what to do after Black Friday covers what to roll back first, how fast, and whether to keep spending through December and January.

How to review Black Friday afterwards

Start with one question: did the peak come in below expectations because demand for your products was weaker, or because your ads got fewer impressions? The two point to different fixes.

Give the data time to settle before you judge. Changes made in Black Friday week reach two full weeks of data in mid-December. Compare with last year only if you have that year’s data in the same form. For the full method, see reviewing your peak season.

Example: a tableware shop’s plan, and what a discount does to it

Example store, not client data.

A tableware shop with 3,000 products spends 40,000 a month and gets 180,000 in revenue, a ROAS of 4.5 (450%). Its main Performance Max campaign has a target ROAS of 450% and an average daily budget of about 1,316 (40,000 ÷ 30.4). Its gross margin is 35%, so its ads break even at a ROAS of 1 ÷ 0.35 ≈ 2.86, or 286%.

On 13 November the shop takes its pre-season step. The daily budget goes from 1,316 to 1,645 (+25%), and the target ROAS from 450% to 400% (−11%). On Black Friday itself Google may spend more than 1,645, but a full month at the new budget would cost no more than about 50,000 (1,645 × 30.4).

Last year the shop’s conversion rate from Black Friday to Cyber Monday was about 1.5 times its usual 3.75%. So it sets a seasonality adjustment of +50% for 27–30 November and none for the days after.

Then the shop plans a 20% discount on a third of its products, and the arithmetic changes. On a discounted product the price falls by 20% while the cost stays the same. The margin falls from 35% to (35% − 20%) ÷ 80% = 18.75%, and the break-even ROAS rises to 1 ÷ 0.1875 ≈ 5.33, or 533%.

A target of 400% sits above break-even at full price and at a 10% discount, but below it at 20%. The shop has three honest options: put the discounted products in a campaign with a higher target, discount less, or accept a loss on those orders on purpose. It has to choose in October, before it freezes the structure for the peak. Our data has no margins or repeat purchases, so only the shop can tell whether Black Friday buyers come back.

What the portal shows before and after the peak

Two questions in this guide need your own history: whether November is your peak, and whether the changes you made around it paid off.

The portal answers the first in its assortment and seasonality section. It takes 24 completed months and gives the store, and each notable category, one of five states, from “Confirmed by two years” to “Too little data”. If the two years look nothing alike, it builds no seasonal plan and says so.

The portal’s change log answers the second. It records changes to budgets, bids, conversion goals, product sets and whole campaigns. It gives a first, preliminary conclusion a week after a change and a confident one after two weeks, and stops judging after a month. For changes made on Black Friday, the confident conclusion arrives in mid-December.

Who changed what in your ads — and how it ended. The portal keeps a log of changes: budgets, bids, conversion goals, product sets, whole campaigns. For each one it works out what things were like before and after, and says it plainly: did it work or did it hurt. The portal changes nothing without your consent.

See what changed in your account →

Frequently asked questions

When should I start preparing Google Ads for Black Friday?

Three to five months ahead for stock, one to two months ahead for the campaigns and two weeks ahead for the pre-season step. For Black Friday on 27 November 2026, that means growing campaigns from September, or from now if you haven’t started, and taking the pre-season step from 13 November.

Should I create a separate Black Friday campaign?

Usually not. A new campaign starts with a new bid strategy, and Google lists that as a reason for “Learning”. Splitting also divides your conversions: the portal treats 30 conversions a month per campaign as the minimum and 50 or more as comfortable. If you do want a separate sale campaign with a fixed spend, use a campaign total budget; it works only on new campaigns.

Should I raise target ROAS during Black Friday?

No. Raising it is a setting change, which Google lists among the reasons for “Learning”, and it comes in the days that matter most. Our rule is the opposite: lower the target by 10–15% two weeks before the peak, then put it back in steps afterwards. If you discount deeply, check the lowered target against your break-even ROAS first.

Can every account use promotion mode?

Not yet. Google describes it as a beta for Search and Performance Max campaigns and asks advertisers who want it to contact their account team.

Sources

• Create seasonality adjustments — Google Ads API documentation: an advanced tool for expected conversion-rate changes during upcoming events; ideal for 1–7 days, may not work as well beyond 14. Checked 2 October 2026.
• BiddingSeasonalityAdjustment — Google Ads API reference: channel types Display, Search and Shopping. Checked 2 October 2026.
• Bidding and Budgeting Updates to Scale Your Growth — Google, published 15 June 2026: promotion mode in beta for Search and Performance Max, temporary ROAS tolerance and extra daily budget, combination with campaign total budgets, beta through the account team; the change for budget-limited campaigns rolling out from 17 August. Checked 2 October 2026.
• Promotion mode is here - Google’s Ginny Marvin explains what actually changed — PPC Land, trade press, reporting Ginny Marvin’s LinkedIn comments: 3 to 14 day window with a set end date; runs alongside seasonality adjustments in Search and Performance Max; interaction not documented by Google. Checked 2 October 2026.
• Campaign budgets overview — Google Ads API documentation: daily spend varies, no more than 30.4× the average daily budget in a month; campaign total budgets for Search, Standard Shopping, Performance Max and other types with set start and end dates. Checked 2 October 2026.
• Create campaign budgets — Google Ads API documentation: the budget type can’t be changed after a campaign is created. Checked 2 October 2026.
• Duration of the learning period for campaigns and what affects it — the reasons for the “Learning” status, including ad group target changes in Shopping; a few conversion cycles, typically 1–2, to calibrate. Checked 2 October 2026.
• Create data exclusions — data exclusions are for dates with conversion tracking problems; frequent or long use can hurt Smart Bidding. Checked 2 October 2026.
• ProductAttributes — Merchant API reference: the sale price effective date is the date range during which the item is on sale. Checked 2 October 2026.
• Overview of Promotions sub-API — the special offer link on Search, Shopping and Chrome; requirements; Google reviews promotions before showing them. Checked 2 October 2026.
• GetProfit data: 96 stores, July 2025 – June 2026 — best and worst months, ROAS of the worst months against the best, spend in the worst months.
• GetProfit data: 125 stores, June 2025 – June 2026 — search impressions lost to rank and to budget.
• GetProfit portal methodology — step sizes for budget and target ROAS, the pre-season rule, verdict windows for changes and the four phases of a season.