Better shoppers do not guarantee better returns. Last year, across 1,528 Standard Shopping accounts we analyzed at Optmyzr, the four days from Black Friday through Cyber Monday brought the best traffic of the quarter, and ROAS still fell by more than a fifth.

Standard Shopping campaigns converted 6.65% of clicks during those four key days. That was the highest rate across all three BFCM timeframes we checked. On top of that, the average order size reached $94.37. But even with all those sales, ROAS tumbled down from 749.46% to 584.43%.

Why? Because the cost for every single click jumped up by 46.24%, while conversion rates only squeezed out a tiny 6.80% bump.

Nothing in that data says the sale went badly. It says the targets and budgets were built for a different week. That is a preparation problem, and almost all of it is decided before the sale starts.

This checklist is built around three moments: six weeks out, two weeks out, and the last 48 hours. For each one you get what to do, what to leave alone, and which jobs you can hand to automation.

Your Black Friday 2026 timeline at a glance

Why does ROAS fall on the best shopping days of the year?

Because click prices rise faster than shoppers improve. We split the BFCM period into three windows (the seven days before Black Friday, the four days from Black Friday to Cyber Monday, and the seven days after) and compared them.

Here is the objection you are already forming: “So what? Volume on those four days is worth a lower ROAS.” Often it is.

The 2025 trade was roughly 12% to 14% more revenue per click for 29% to 46% more cost per click, and whether that clears your margin depends on the gap between your conversion value and your gross profit. The costly version is paying more by accident, with a target someone set in September.

Frederick Vallaeys, Optmyzr’s CEO, put the planning rule this way in the report, “Prepare for volatility instead of chasing past efficiency. Holiday click prices spike, so set your BFCM targets based on the realities of the sale days, not your monthly average.”

What happens when your Black Friday plans don’t go as expected?

Even with a solid checklist, Black Friday can throw unexpected problems your way. Budgets run out sooner than planned, campaign performance drops, or tracking issues make it difficult to know what’s actually happening.

Recently, our evangelist, Aaron Levy, spoke to PPC experts Jyll Saskin Gales, Andrew Lolk, Akvile DeFazio, Brian Binder, and Sean Stone to discuss real BFCM experiences, what went wrong, how they responded, and what they’d do differently.

If you’re preparing for Black Friday, their experiences can help you think through problems your checklist might not cover.

Watch the full episode now

Have a question for the experts? You can tag them in the comments under the video and ask them directly.

6 weeks out: Get the fundamentals right

Your conversion setup, budget goals, and product catalog feed control what the automated bidding system can learn. If any of those three things break during the live sale, you won’t be able to fix them quickly.

Fix your conversion values before you touch a bid strategy

Smart Bidding is only as reliable as the conversion values you feed it. Looking back at our December 2025 numbers, PMax campaigns set to Maximize Conversion Value earned $4.30 per click on average, while campaigns using Maximize Conversions earned $2.32, which is roughly 46% less.

Those are different advertisers, so treat the gap as a signal rather than a verdict. The takeaway is that, if your values are unreliable, Maximize Conversions can be a safer baseline than value-based bidding until you fix the value signal. Here are three things to check:

• Every conversion action that should carry a value does carry one.
• The values reflect margin rather than revenue wherever you can manage it.
• No single action is carrying a default value applied across the whole account.

While you are checking your settings, count up all your primary purchase tracking rules.

Bia Camargo, Head of Google at Ecom Nation, pointed out this super common mistake during a recent episode of PPC Town Hall: “I have clients with three primary conversion actions for purchases, and I say, ‘Oh, you are at least double counting your results. So let’s clean this up.’ I have like page views, but they are not actually applying to any campaigns. So these kinds of small changes that you can fix like quick wins that get your data more clean to give the machine clear signals, right?”

Double-counted purchases make every ROAS target in the account wrong by the same factor, and it often goes unnoticed until the target stops matching the P&L.

Next, figure out what each sale is actually worth to your business. Casey Gill, a Senior Search Specialist at Web Savvy, explained why profit margins matter so much on PPC Town Hall,“You can afford a lower return on a higher margin product than you can on a lower margin product.”

Over on another episode of PPC Town Hall, Google’s Tetsuo Konno described how a Dutch fashion retailer sent a custom “profit value” into Google that factored in product margins along with return rates. You don’t need to get that complex by October! But you definitely need to spot which popular items have thin profit margins before ad click prices climb higher.

Decide your bid strategy and targets while there is still time to be wrong

Google says Smart Bidding needs about one to two conversion cycles to adjust whenever you update a target. A conversion cycle is the average time between a click on your ad and an order. So, count backward by two full cycles to find your safe deadline for target changes. If shoppers usually take seven days to buy, November 13 is your cutoff date. If they buy on the same day, you have more wiggle room, but you should still set your targets now so the system settles down.

You can set the BFCM target for the days you are actually selling into, using margin rather than a monthly average, and decide in advance what you will do if December repeats 2025: Standard Shopping accounts went into December paying near-November click prices on a conversion rate that had dropped 10.20%.

The learning math matters most for anything new. Jyll Saskin Gales, a Google Ads coach and the author of Inside Google Ads, said on PPC Town Hall that a campaign needs “at least 30 conversions in 30 days. Or if you’re a lower-volume account where that’s not going to happen, then it can take two, three, four months for the thing to learn.”

Nils Rooijmans, a PPC agency founder and Google Ads scripts expert, was blunter about the cost on PPC Town Hall, “A bit of the Google entry tax, just to be able to play, you have to build up that data, and that can easily cost $10,000 or $20,000.”

When you put those two facts together, the core rule is simple. If you launch a brand-new campaign in late November, you end up paying high learning costs right during Black Friday peak hours. So, either get new campaigns launched well in advance, or hold off until January.

Decide where the structure stops

Andrew Lolk, Founder of Savvy Revenue, posted a Black Friday guide where he talks about his rule for changing campaign setups, “October 1st should be the cutoff. November 1st, if you’re really really pushing it. Anything after November should be a no-go for structure changes.”

Anything that needs to learn, such as a new campaign, a new asset group, or a new bid strategy, goes in before the first week of November. Anything after it should be the kind of change that does not reset what Google has already learned.

Clear the feed problems you can see

Disapprovals, missing identifiers, and price or availability mismatches cost the most during a sale, because the products affected are the ones carrying the budget. Bia Camargo’s first step on PPC Town Hall 123 matches ours: “So the first thing is feed health. We organize everything that is not working.”

Run the Shopping Feed Audit on the linked merchant feed. It scores the feed out of 100, flags disapproved products with their reasons, out-of-stock percentage, products advertised in more than one campaign, long titles, missing brand in titles, and missing identifiers, and you can schedule it weekly or monthly so the check keeps running.

You can also edit up to 25 products at a time and push the fixes to Merchant Center as a supplemental feed.

Feed health also includes which products deserve the budget they are getting, and Optmyzr’s Shopping Analysis aggregates product performance over a date range and across the campaigns you choose, so you can see which items drive conversions and which only drive spend.

For a closer look at which attributes are empty or inconsistent, Merchant Feed Analysis, inside Shopping Campaign Management, lists each attribute, how many variations it has, and how many products are missing it. Optmyzr’s newer Feed Optimization tool goes a step further for Google Ads accounts: it publishes edits through a supplemental feed and leaves your primary feed untouched, and its own use cases include seasonal cleanup of thousands of products before a major sales period.

Menachem Ani, who founded JXT Group, highlighted a super important rule to double-check here. He called it rule number one for any ecommerce account on PPC Town Hall,“Make sure that each product is only in one campaign. If you have it in more than one campaign, it kind of fights with each other, and it just reduces efficiency.”

Six weeks out is when you refresh the customer lists and check that remarketing audiences are populating, so you can have a healthy list for Black Friday.

2 weeks out: Build the sale

This is when the work gets visible. It is also when most teams make the mistake that costs them the most, which is building the sale on top of automations calibrated for a normal week.

Submit promotions earlier than Google says you have to

Google’s Merchant Center guidance is to submit promotions at least 24 hours before the effective start time. “Otherwise, the promotion may be delayed by 24 hours.”

Google also says review, processing, and validation can take up to 72 hours, that promotions do not go through SKU review until their start date and time arrive, and that you should submit during business hours, Monday to Friday.

So it’s time to do the calendar math. Black Friday is November 27 and Thanksgiving is November 26, so the last clean business day before launch is Wednesday, November 25, and a promotion that hits trouble that day has no buffer. We would treat Friday, November 20 as the real deadline for every promotion you plan to run, and test each promo code with a real order at checkout before you submit it.

Also open the Promotions tab in Merchant Center this week. Google has an automated promotions feature that extracts live promotions from your product pages. It is in a closed beta for merchants in the US, France, Germany, Japan, Great Britain, Canada, Australia, and India, and Google says not all websites or products are eligible.

If you see a “Website promotions have been linked” banner and you did not set that up, decide on purpose whether you want it on. You can end individual promotions or remove automatic promotion updates from the banner.

Build sale products as their own campaign if you need separate control

If sale products need their own budget or target, build them as their own Shopping campaign instead of mixing them into your evergreen ones.

In Optmyzr’s Shopping Campaign Management, the Product Selection option limits a campaign to products that match a rule on a feed attribute, such as CustomAttribute0 equal to “sale.”

The campaign builder can also save the campaign paused, so you can build it ahead of time and enable it when the sale starts. The Refresher automation then syncs the campaign with the latest feed data on a daily, weekly, or monthly schedule and, depending on your settings, adds new products or suggests removals.

This is where Menachem Ani’s one-product, one-campaign rule bites. A product that sits in the new sale campaign and still sits in the evergreen one is exactly the overlap he warns about. Exclude it from the old campaign when the new one goes live.

A campaign created in mid-November has no history, and Menachem Ani warns against swapping everything out for that reason, “We like to use our existing campaigns, layer on all the updated ad copy, add creative promotions, build on top of what’s already there. Definitely don’t like swap it all out because if you do, you’re kind of forcing the system to learn a bit new.”

A separate sale campaign is worth it when the products need their own budget or target. If they do not, update the campaigns that already exist.

Lift the budgets that are capping you, and ramp them

If a campaign is limited by budget and earning well, it is sitting out auctions you want.

Andrew Lolk recommends fixing budget caps ahead of time rather than tweaking them during the sale, “If you are limited by budget, make sure you raise the budgets a little bit before because as of right now, you’re being limited in the amount of auctions you can get into when you’re limited by budget and you don’t want to walk into Black Week with that limitation on. So, make sure you increase them way in advance and just use your target [ROAS] to manage your spend instead.”

He also advises ramping up gradually rather than making one big swing. A seven-day ramp from now is easier to read, and easier to reverse, than one big change on November 24.

Use seasonality adjustments rarely, and Promotion mode with care

Google describes seasonality adjustments as ideal for short events of one to seven days and says they may not work as well beyond 14 days. That makes sense on paper, but it leaves out what happens during the biggest shopping week of the year!

We studied it. We compared accounts that set a seasonality bid adjustment during BFCM with accounts that did not, across 2022, 2023, and 2024.

Accounts without an adjustment saw CPC rise about 16% to 17% during the period. Accounts with one saw CPC rise 32% to 36.7%. Smart Bidding detects the BFCM spike on its own, and our conclusion was to let it handle the macro trend and keep adjustments for brand-specific events it has no history for, like a one-off flash sale or an email-only promotion.

Google also has a newer option. Promotion mode lets you schedule a temporary change to your ROAS tolerance plus extra daily budget for a set window. Ginny Marvin described it as a feature that ends on its own, “there’s no need to remember to come back to turn it off,” and the window runs three to 14 days.

PPC News Feed’s September 2026 report says it is in beta for Search and PMax campaigns and that baseline ROAS targets cannot be changed while a promotion is active, so set them first.

Audit the automations you already have

Navah Hopkins brought up a great point while hosting PPC Town Hall for Optmyzr. She was talking about sudden tariff shifts, but her advice fits holiday sales just as well, “Go check your automations, your scripts, your roles, every single one of them, I promise you, they are not tuned for the market we’re about to hit. You might have ROAS goals that are too aggressive. You might have bid thresholds that are out of whack.”

So run the arithmetic on your own rules. If a rule pauses anything above a cost per conversion threshold, and click prices run much higher than the week before, that rule can fire across your best products on Black Friday morning. Either raise the threshold for the sale window, or run the rule so it only reports.

In Optmyzr’s Rule Engine, a strategy with an “include in report” action never changes anything, and the Send Notifications automation option emails the matches without applying them (the Help Center says this option is available on request), so you can use that to test your automations before firing any changes.

Set up the alerts now

Two weeks give you enough time to find out which alerts are noise. Three kinds are worth setting.

1. Budget Monitors track spend against a target and notify you when pacing runs over or under, and when spend crosses 50, 75, 90, and 100% of the target. Monitoring and alerting is the default. Pausing and auto-adjusting budgets are optional and have to be switched on.
   

   
   

2. KPI alerts cover the metrics that matter on sale days. They can fire when a KPI moves more than a percentage you set, or drops below a number with an allowed deviation, at the account, campaign, bid strategy, campaign type, product, or merchant feed level.
   

   
   

3. Feed alerts come in two forms. Optmyzr sends an automatic Feed Anomaly alert when more than 50% of the products in your feed are disapproved or out of stock, which is a last-resort signal. For something different, you can use the KPI alert to create one at the Feed level on Merchant Feed Product Disapproval, with a maximum disapproval percentage or a maximum number of disapproved products.

The last 48 hours: Freeze the edits

By now, every change you make is one you cannot measure before the sale starts, and one you can only half-undo. The job is to freeze what can be frozen and to be precise about what cannot.

What is safe to touch, and what is not

Check that the freeze actually held

A freeze is a promise, and people break promises on busy days. Ask for the recent account change history and look for anyone who touched a bid strategy, target, or budget. If you work through the Optmyzr MCP, you can ask Claude to list recent changes for the account, filtered by campaign or change type. Platform change history is most reliable for roughly the last 30 days, which is exactly the window you need.

Decide who watches what

Agencies already know the failure mode: three people look at the same dashboard, nobody owns the decision, and the client emails at 2 p.m.

Menachem Ani has seen worse, “I’ve seen everything from like a promotion that was like approved but then like suddenly came back as disapproved and you know all the pro everything we’re doing is predicated on the success of that promotion. So I’ve seen that I’ve seen like account suspensions come through like while you’re having the turkey at Thanksgiving dinner.”

Someone has to be looking when that happens, and that someone needs to know it is their job. Write the ownership down before the sale. The table below is our suggested starting point.

If your team runs this across many accounts, Optmyzr’s Account Blueprints can turn the table into a runbook: they schedule one-time or recurring task series, assign each task to a person or a role, and work across multiple ad accounts.

Write the if-then playbook now

The temptation on the day is to react to a number. The antidote is to decide in advance what each pattern means. We would start with these four:

• CPC up, revenue per click up about as much. The auction is hot and the shopper is good. Do nothing.
• CPC up, revenue per click flat. Inspect auction pressure and your competitors before you change budgets or targets.
• Conversion rate down, CPC stable. Check landing pages, stock, and delivery promises first. This is a site problem until proven otherwise.
• Order value down, conversion rate stable. Your promotional mix is doing something bidding cannot fix.

During the sale: what to check, and when

Most of the sale is a monitoring job. The bigger risk is reacting to a non-problem. A reasonable cadence:

Optmyzr’s Sale Day Command Center puts this on one screen. It pulls from Google Ads, Merchant Center, and Google Analytics, compares against the same sale event last year, and auto-refreshes every 30 minutes. You can share it by secure link with clients for your Google Ads accounts.

For the feed owner’s hourly check, the Product Performance view shows how many products are disapproved, which revenue-driving products are among them, and which strong performers from the last 90 days have gone out of stock.

When a campaign is capped and earning, the Optmyzr MCP can help you decide how much to add. Ask Claude how much budget you can safely shift to top performers, and it will read your campaign data and propose a reallocation. It does not apply anything. Every change the MCP proposes becomes a Change Request that a person with Apply access reviews.

Two experienced advertisers share a key warning about the very first day. Andrew Lolk warns against spending money too aggressively on Friday because blended ROAS looks good early on: “The later into Black week that you get the more people from your funnel is going to buy. And if you spend heavily on Friday because your blended ROAS allows it, you’re paying for customers who basically would have bought on Sunday anyway.”

On Menachem Ani’s side, his rule for the whole weekend is simple: “Really what you want to do on Black Friday and even through the whole weekend is watch spend by the hour. You can break down the reporting like that and just make sure that you’re pacing well.”

The week after: do not cut every budget because the sale is over

The week after Black Friday is the worst of the three windows, and most plans leave budgets alone through it. In our 2025 data, Standard Shopping CPC was still 10.21% above its pre-Black-Friday level a week later, and daily conversion value was 17.51% below it. High prices and low demand at the same time.

PMax looked different: its CPC had already fallen 1.68% below the pre-Black-Friday level. So review the two campaign types separately, and use your own post-sale numbers before you cut.

Menachem Ani puts it this way: “You’re going to want to lower your bids and budgets very quickly as soon as the sale ends or as soon as the promotion tapers off because otherwise the system will overspend that it won’t convert.”

Our data supports that instinct for Standard Shopping, where click prices stayed high while value fell, and supports it less for Performance Max, where click prices had already come down. So the best strategy is to cut where your own numbers show the demand is gone.

When it is over, pause the sale campaigns in bulk. Optmyzr’s Rule Engine can find every enabled campaign with a naming convention such as “Black Friday 2026” and set its status to Paused, which is one more reason to name them consistently now.

Pull last year’s numbers this week

The most useful thing you can do after reading this is to pull your own last-year numbers for three windows (the week before Black Friday, Black Friday through Cyber Monday, and the week after) and write down CPC, conversion rate, and revenue per click for each. Every decision above gets easier once you can see your own version of the table. If you use the Optmyzr MCP, you can ask for campaign performance by day across those three windows in a single prompt.

New to Optmyzr? Start a 14-day free trial of Optmyzr and set up the Budget Monitors, alerts, and Sale Day Command Center before the two-week mark. Already a customer? Open those tools from the Help Center links above.

Frequently asked questions

When should I stop making changes before Black Friday?

Stop changing bid strategies and targets at least two conversion cycles before the sale, because Google says Smart Bidding needs one to two conversion cycles to adjust to a new target. Stop restructuring and launching new campaigns by the two-week mark. In the last 48 hours, touch only price and availability updates, disapproval fixes, and budgets on campaigns that are earning.

Should I use seasonality adjustments for Black Friday?

For the mainstream retail event, our 2022 to 2024 study says no. Accounts that used them saw about double the CPC inflation and a larger ROAS decline than accounts that did not, while Smart Bidding picked up the BFCM spike on its own. Use them for brand-specific events Google has no history for, like a flash sale or an email-only promotion, and keep them to the one-to-seven-day window Google recommends.

How early should I submit Merchant Center promotions for Black Friday?

Google’s minimum is 24 hours before the start time, and a late submission may be delayed by 24 hours. Review and validation can take up to 72 hours. With Thanksgiving on November 26, we would submit every Black Friday promotion by Friday, November 20.

How long does Smart Bidding take to adjust to a new target?

Google says Smart Bidding will adjust to a new target over the next one to two conversion cycles. A conversion cycle is roughly the time it takes for a click to turn into a purchase. Count back two cycles from Black Friday to find your last safe day to change a target.

How much will CPC rise during Black Friday and Cyber Monday?

In our 2025 data, Standard Shopping CPC rose 46.24% and Performance Max CPC rose 29.15% from the week before to Black Friday through Cyber Monday. In our 2022 to 2024 study, CPC rose about 16% to 17% for accounts without a seasonality adjustment, measured from the Wednesday before Black Friday to the Wednesday after Cyber Monday against a baseline week earlier in November. The windows differ, so use the same window for your own comparison.