Black Friday Google Ads 2026: Cyber Five Returned 8.41x, and the Prep Starts in August

Black Friday pays the stores that prepare for it. Across the 113 store accounts we ran for a full twelve months, the five days from Thanksgiving to Cyber Monday 2025 returned 8.41x against a 7.18x yearly average. Those accounts spent 39% more per day than their yearly average and brought in 63% more revenue per day.

Here is the number most Black Friday guides leave out, and it is why preparation beats spending. October to December produced 26.1% of the median account’s annual Google Ads revenue; a perfectly flat year would give you 25%, and March and May each matched November. The holiday season is a short, expensive sprint inside a normal year, not the thing that makes or breaks the year for most stores. The stores that treat it like a sprint (an offer decided early, campaigns left alone, throttles loosened, a plan for December) do well. The ones that treat it like a rescue mission launch a brand-new Performance Max campaign the week before Thanksgiving and spend the most valuable days of the year paying for the algorithm’s education.

This year Black Friday is November 27 and Cyber Monday is November 30. I tell my team that if you start the Black Friday conversation with a client in August, you’re perfectly on time, and if you wait until October you may not have the runway.

The short answer: the stores that do well on Black Friday prepare early and change little in November. Lock your structure by mid-October, decide the offer and its break-even early, loosen ROAS targets a notch before the rush, and budget for click prices that peak in Cyber Monday week and the week after. The data is below, and the plan is at the end.

The median store has three strong months, not one: March and May matched November

On the fixed set of 113 accounts, where Q4 was 26.1% of the median account’s year, November’s revenue index was 117. March was 116 and May was 117. December came in at 107 and October at 94. For the median account, where every store counts equally regardless of size, November was only 105.6 against a 100 average. The blended curve gets its November bump from a handful of large giftable stores; the typical store has a good November and a few other months that are just as good.

Why does this matter? Because the Black Friday plan most stores run assumes the opposite: go big in November and coast the rest of the year. If your spring is as strong as your November, the bigger risk isn’t under-spending on Black Friday. It’s overspending into it at poor efficiency, then walking into January with an algorithm that spent six weeks learning from discount shoppers. There’s nothing we can do about seasonality; it’s part of the game. What we can do is look at it clearly before we decide what to spend.

Giftable stores still live in Q4. Parts and supply stores mostly don’t

Split by vertical, the textbook holds exactly where you’d expect it to. Outdoor recreation stores took a median 33.0% of their year in Q4 and food and beverage stores 32.1% (42.0% blended, because the biggest food stores are also the most seasonal). Auto parts stores took 23.8%, which is less than an ordinary quarter. Industrial and B2B supply sat at 25.2%.

If you sell gifts, food or seasonal decor, the season is your year, and everything below applies twice over. If you sell brake parts or warehouse supplies, Q4 is a normal quarter with more expensive clicks, and your job is to not get dragged into a bidding war you don’t need. And know which season is actually yours. One specialty food store in our book does most of its year around a September holiday: its September ran at 14.8x, and by November it’s a quiet account. Its Black Friday is in September.

251% Increase in ROASLuxury Home Decor Retailer Quadruples ROASSee the case study →241% Increase in ROASLuxury Fashion Brand Triples ROASSee the case study →

Cyber Five paid 17% better: more buyers and bigger baskets, with the ROAS lift from the baskets

Here is what the five days looked like against the same accounts’ yearly daily average. Spend per day was $73,610 against $52,957, up 39%. Revenue per day was up 63%. ROAS was 8.41x against 7.18x. Conversion rate was 2.54% against 2.59%, so flat, and CPC was actually a touch lower at $0.857 against $0.889. Black Friday on its own returned 9.97x. Cyber Monday returned 8.83x, but its average CPC of $1.185 was 33% above the yearly average, which made it the most expensive day of the year.

With spend per day up 39% and CPC slightly lower, clicks and orders each rose by roughly 40%, so there were more buyers. The ROAS lift came from the bigger baskets: orders over the five days averaged $284 against $246, about 15% larger. Black Friday and Cyber Monday themselves did convert better (3.25% and 3.44%), but Thanksgiving and the weekend pulled the five-day rate back to normal. That’s an offer and merchandising story (bundles, thresholds, gift sets), not a bidding story.

It also tells you how to treat one setting Google offers for exactly this week. Google’s help page on seasonality adjustments says they are meant for short events of one to seven days, and only when you expect a major change in conversion rate. Across the five days, our book didn’t see one. If you use an adjustment, keep it to Black Friday and Cyber Monday themselves, and size it from your own account’s history, not from a hopeful guess.

CPCs start rising the first week of October, and the expensive two weeks are Cyber Monday week and the week after

Weekly CPC on the fixed set, against the September 2025 average: the week of October 6 was the first jump, +8%, and October held 3% to 7% up after that. November climbed to +12%, +15% and +16%. Thanksgiving week itself ran only +7%. Then the week of December 1 hit +27% and the week of December 8 +28%, before prices fell back to +3% the week of December 22 and below September by the week of December 29.

My read on Thanksgiving week is that plenty of advertisers cap out or hold budgets flat going into the long weekend, while the gift-buying that happens in early December runs into everyone’s shipping-deadline pushes at once. Either way, the expensive part of the season is Cyber Monday week and the week after, not Black Friday week.

The other surprise is that the auction didn’t get cheaper afterward. On the median account, CPC in June to August 2026 ran at index 106 to 108, above November 2025 (101.5) and December 2025 (102.8). The old advice to bid down for the holidays assumes a spike you can wait out. In 2026 clicks simply cost more all year. For planning, budget October at roughly 8% more per click than September, mid-November at about 16% more, and the first two weeks of December at about a quarter more. How those numbers translate into a monthly figure is in my post on how much a store should spend on Google Ads.

Launch anything new by mid-October, because a November Performance Max won’t be ready in time

Performance Max is like a freight train. It takes a long while to get going, but once it does it’s very difficult to slow down, and if it stops it takes a while to get up to speed again. In my experience a new campaign needs two to four weeks to settle, and it doesn’t really perform until it has seen around 30 conversions in 30 days. Launch one on November 15 and it spends Black Friday learning from the most abnormal traffic of the year. Then there’s the second problem I see every season: the algorithm gets excited by what it sees in late November and overspends into December, when the discount shoppers are gone.

Work backward from November 27. Four weeks of learning plus a week of buffer puts the last sensible launch date around October 23, and I would rather have anything new live by mid-October. That applies to anything that resets learning: new campaigns, switching bid strategies, big target changes, splitting a catch-all into new campaigns. If it isn’t in place by then, it waits until January. How I build the campaign itself is in Performance Max for ecommerce.

What you can change late is the offer, and you should put it where your existing campaigns can pick it up: sale prices in the product feed, Merchant Center promotions, promotion assets, and fresh headlines in the asset groups that are already running. For one store we run, the month-long Merchant Center promotions were clearly worth keeping, while the short flash sales weren’t doing much we’d miss. Schedule as much of this as you can now, so nobody runs a fire drill the week of Black Friday. And check that tracking counts each purchase once before the traffic arrives; a doubled conversion in November teaches the bidding the wrong lesson at the worst time. The feed side of this is in product feed optimization.

Make the offer at least 25% off, and do the break-even math at the sale price

In my experience, shoppers expect at least 25% off on Black Friday. Would you buy from a brand that isn’t giving you 25% that week? I wouldn’t. And free shipping is not a Black Friday Cyber Monday offer; you should be doing free shipping and baking it into the price anyway. The flip side is that most brands discount too often the rest of the year, which trains customers to wait. Black Friday is the one week where everybody is looking for a deal, so make it your biggest discount of the year and keep the other 51 weeks quieter.

If 25% is too deep for your margins, you don’t have to give away 20% to 30% to get attention. A free gift that’s overstock, high margin and cheap to you works well, and so do bundles. Both lift order value, which is exactly where the Cyber Five gain in our data came from.

Whatever the offer, redo your break-even math at the sale price. Take a $100 product that costs you $50. At full price, every $1 of ad spend needs to bring back $2 to break even, a 2.0x ROAS. At 25% off, the price is $75 and the margin is $25, so break-even rises to 3.0x. At 40% off it’s 6.0x.

This is the check that keeps Black Friday profitable. The dashboard shows 8x, everyone says the week went great, and nobody checks that a 40% discount put the floor at 6x. At the 8.41x Cyber Five return across our fixed set of accounts, a 25% offer is comfortable. A store that runs at 3x all year and then discounts 40% is buying revenue at a loss. My post on what a good ROAS is walks through the margin math in more detail.

Loosen your ROAS targets before the rush, then carry promotions into December with care

Target ROAS and target CPA should be viewed as budget tools, not performance tools. Set your target ROAS too high (or your target CPA too low) during the one week demand is really there and you’re kneecapping your own impressions. If a client wants 13x, I’ll often set the target around 9x. Too tight a target chokes volume, and the account usually lands above the target anyway. For Black Friday, I ease targets down a notch one to two weeks before, not the morning of, so the bidding has time to adjust.

That sounds like it contradicts the break-even section, but they are two different numbers. The target is a throttle that decides how much traffic the campaign is allowed to buy. Break-even is the floor you judge the week against afterward. Loosen the throttle, but keep it above the break-even you worked out at the sale price.

After Cyber Monday, a lot of advertisers exclude the sale period from their conversion data so the algorithm snaps back to normal. I’m wary of that. You want to carry your Black Friday promotions into December and keep winning customers, and revenue backs it: December’s revenue index was 107, still above the yearly average, even with the two most expensive weeks of CPC. Efficiency doesn’t, though. December ROAS was 6.80x against the 7.18x yearly average, so keep the tail on a short leash, watch the marginal return and trim as the shipping cutoffs pass. Then pull back in January. On the fixed set, January CPC dropped to index 88 and ROAS rose to 7.50x from December’s 6.80x, which makes it a good month to rebuild efficiency.

One more place the season leaks: your brand name. Branded searches jump in November and competitors know it. Protect your name with an exact-match brand campaign on a small budget, which I cover in should you bid on your own brand name.

The Christmas-tree audit: judge the season on the last dollar, not the total

One seasonal store that lives and dies by the holidays doubled its holiday spend and got 50% more revenue. Spend rose 104%, revenue 50% and orders 29%. ROAS fell from 9.16x to 6.71x, which still looked healthy on a dashboard, but the extra holiday dollars returned about 4.4x against roughly 9x on the base. And much of the “growth” wasn’t the ads at all; it came from a price increase of around $40 in average order value.

The extra dollars returned far less than the first ones. That is diminishing returns, and across our book the pattern hides, because bigger accounts post higher ROAS (a median 6.84x at $10,000 to $50,000 a month against 2.13x under $2,000), but that is successful stores earning bigger budgets. Inside one account, in one season, the last dollar is the one that tells the truth, and in the holiday weeks it is buying the most expensive clicks of the year.

What I’d propose to a store like that is the same sequence I’d give any giftable store: work out break-even ROAS from contribution margin, measure the season against one neutral attribution source instead of each platform’s own report, split brand from non-brand so the brand clicks stop flattering the total, pull back hard in the shoulder months, and use Meta catalog retargeting in the gift window, where it earns its keep. We sequence the work by risk, not by what’s most exciting.

The Black Friday Google Ads plan for 2026, in order

1. First: decide the offer and the floor. At least 25% off, or a free gift or bundle that lifts order value, and a break-even ROAS worked out at the sale price, not the list price.
2. By mid-October: freeze the structure. Any new campaign or bid-strategy change goes live now or waits until January. No new Performance Max after about October 23.
3. Check tracking before the traffic arrives. Purchases counted once, as the primary conversion, with add-to-carts and page views set to secondary.
4. Load the offer where running campaigns can use it. Sale prices in the feed, Merchant Center promotions scheduled early, promotion assets and new headlines in existing asset groups.
5. Budget for the CPC curve. Roughly +8% per click in October, +16% by mid-November and about +25% for the first two weeks of December. Add budget in steps, only where a campaign is limited by budget while hitting its target.
6. Loosen targets a notch one to two weeks out. Keep them above your sale-price break-even. Use a seasonality adjustment only for Black Friday and Cyber Monday, sized from your own history.
7. Protect the brand with exact match. Small budget, tight match types, and an eye on who else shows up on your name that week.
8. Carry promotions into December, then pull back. Keep them running while the marginal return holds, cut back in January, and judge the season on your books and your marginal return, not the platform’s total.

If you do one thing this week, pull last year’s monthly revenue out of your ad account and your store, and put November next to March and May. If November wasn’t clearly your best month, plan this Black Friday as a strong week inside a normal quarter and spend accordingly. If it was, lock in the structure by mid-October. If you’d like an outside read on the plan before the freeze, that’s what our ecommerce ads team does every fall. Either way, the prepared account is the one that gets to enjoy the bigger baskets when Cyber Five arrives.

Frequently asked questions