Every year, sometime around the first week of November, the same thing happens across dozens of advertising accounts. Someone opens Ads Manager, checks the calendar, sees that Black Friday falls on Friday, November 27, and increases the daily budget from $100 to $300. No new creative, no different audience, no consideration of margins: just more fuel in the same engine, based on the idea that people buy more in November, so the more you spend, the more you make. Three weeks later, that same account has spent three times as much, brought in maybe half as many additional results, and seen its cost per lead climb to a level that, if someone had announced it in September, nobody would have accepted.
The problem isn’t Black Friday, and it isn’t the budget either. The problem is the idea that Q4 can be bought when it arrives, by entering the auction on the same day everyone else does, with the same tools everyone else has, and with one key difference: less preparation. A media buyer who knows their craft doesn’t try to beat the auction on the day it is most crowded. They arrive with an advantage that was built beforehand, made up of audiences that already know the brand, creatives whose performance is already understood, and data that the algorithm has already processed.
This article explains how to build that advantage, month by month, starting with something almost nobody considers: Black Friday doesn’t begin in November, or even in September. It begins in August.
Black Friday Starts in August
In August, many business owners shut down their campaigns, convinced that nobody is buying and that every dollar spent is a dollar wasted. Purchase intent does indeed decline across many industries, but this is exactly where the point gets missed: when a significant portion of advertisers leaves the auction, the auction becomes less crowded, and in many accounts the cost of reaching a thousand people drops significantly. On Google, you can sometimes see days when the platform can’t even spend the planned budget simply because there isn’t enough demand or competition. This isn’t a universal rule, because it varies by industry and market, but it happens often enough to deserve a strategy rather than an on/off switch.
Those who continued investing in August, perhaps with reduced budgets and lower thresholds to prevent the platform from wasting money it couldn’t spend effectively, didn’t just buy leads. They bought attention at a discount. They accumulated video views, profile interactions, landing page visits, people who opened a lead form without completing it, and contacts added to a list. In other words, they bought signals. And signals are exactly the raw material the advertising system will need in November, when everything costs more.
In September, the picture changes. Advertisers who had paused their campaigns come back, company budgets ramp up, and demand picks up again. This is the real restart, and from here the market begins its climb toward the late-November peak. Anyone who reaches this point with an audience already built has a tangible advantage over someone starting from scratch, because the people who already encountered the brand in August don’t have to be bought again at full price in November. They need to be converted. And as we’ll see shortly, that difference can be measured in dollars in the cost of every individual lead.
Why CPL Rises in Q4 Even When the Campaign Is Identical
There’s a question every client eventually asks around mid-November: the campaign is the same as October, the creatives are the same, the audience is the same, so why has the cost per lead gone up? The answer lies in a calculation worth keeping in mind because it explains more than any report ever could.
Cost per lead is the result of three factors: how much it costs to show the ad to a thousand people, how many of those people click, and how many of those who click become leads. Put simply, CPL is the CPM divided by the number of leads you generate per thousand impressions, and that number depends on both the click-through rate and the conversion rate. Let’s use some round numbers. With an $8 CPM, a 1% CTR, and a landing page that converts 10% of visitors, every thousand impressions produce ten clicks and one lead, making the cost per lead $8. If the CPM rises to $12 in November and everything else stays the same, the cost per lead rises to $12. Nobody did anything wrong, the campaign is identical, and yet every lead costs 50% more.
Now imagine those impressions aren’t going to strangers, but to people who already saw the brand in August and September. If that warm audience converts at 15% instead of 10%, the cost per lead drops back to $8 even with the same $12 CPM. This is the central point of the entire argument: when the cost of attention rises, the only lever capable of absorbing that increase without sending CPL through the roof is a higher conversion rate. And a higher conversion rate can’t be improvised in November. It is built in the months beforehand.
But does CPM really always go up? This is where it’s worth being honest, because the data tells a less linear story than the one you usually hear. On Google, the pressure is clear. According to smec’s Black Friday 2025 recap, based on nearly 2,000 Google Ads ecommerce accounts across Europe, spending increased 31% year over year and cost per click increased 24%, with 12% more unique competitors participating in auctions. During the same period, average order value fell 6.6% and gross margins eroded throughout November. In other words, advertisers spent more to sell at lower values.
On Meta, the picture is more nuanced. Tinuiti’s Q4 2025 benchmark for the U.S. market found that aggregate Facebook CPM was down 13% compared with the same quarter the previous year, partly due to the growth of lower-cost inventory such as Reels. Yet within that declining average, 44% of the advertisers analyzed saw their own CPM increase. Two clarifications are important: this was a year-over-year comparison, not a September-to-November comparison, and it covered the United States. The lesson, however, applies everywhere. The market average isn’t your account, and an aggregate figure can fall while nearly one advertiser in two pays more.
That’s why a media buyer’s job isn’t to predict whether CPM will rise. It’s to be ready for the moment when the cost of acquiring the next customer rises, whatever the reason: more competitors in your industry, more budget concentrated on the same audience segments, or more aggressive promotions raising the bar for what customers expect.
Do Remarketing and Retargeting Still Exist in 2026?
If a warm audience is the key to Q4, the next question is inevitable: can you still do remarketing in 2026? After years of increased privacy restrictions, iOS updates, cookie banners, and increasingly automated algorithms, many advertisers are convinced that retargeting is dead. It isn’t. It has changed, and anyone still thinking about it using 2019 frameworks risks looking for a tool that no longer exists while missing what has replaced it.
Remarketing a few years ago was essentially a list. The pixel recorded who visited the website, the platform built an audience of people who had visited in the previous 30 days, and the advertiser showed them a dedicated ad, often with a discount. That list is much smaller and much less reliable today: some users reject cookies, some browse on devices that limit tracking, and some simply aren’t visible to the platform. If you open a custom audience of website visitors and find it is smaller than it was a few years ago, you’re not necessarily doing anything wrong. You’re seeing the actual boundaries of what the platforms can still observe.
At the same time, the way platforms use audiences has changed. On Meta, with Andromeda and Advantage+ configurations, a custom audience no longer functions simply as a fence containing the campaign. It acts more like a suggestion, a signal telling the algorithm where to start while giving it the freedom to go beyond that audience when it finds people more likely to convert. On Google, old remarketing lists have been incorporated into first-party data segments, which still allow advertisers to reach people who visited a website or took specific actions, including through Search campaigns.
So where does effective retargeting live today? In first-party data, the data the brand owns or that is generated within platforms without depending on cookies. People who watched a significant portion of a video, those who interacted with an Instagram profile or Facebook page, people who opened a lead form, newsletter subscribers, CRM contacts, existing customers, and phone numbers that entered a WhatsApp conversation. These are more robust audiences because they don’t depend on a pixel successfully firing in the right browser at exactly the right moment. Customer lists, of course, require a proper legal basis under GDPR: uploading a database to Meta or Google without the appropriate consent isn’t a shortcut, it’s a risk.
There’s also a second, subtler change. Retargeting doesn’t necessarily need to be a separate campaign. The old cold, warm, hot architecture, with three separate campaigns and divided budgets, can fragment data and slow learning in many accounts today. The value isn’t in rebuilding those structures, but in two different things. The first is giving the platforms better signals through Conversions API, Enhanced Conversions, and clean tracking. The second is building a different creative strategy for people who already know the brand. Someone who has seen you ten times doesn’t need you to explain who you are. They need a reason to decide now.
September: Build Before It Gets More Expensive
September is the month when a media buyer should think like an investor rather than a salesperson. The goal isn’t simply to generate leads, even though leads are necessary. It’s to accumulate knowledge: about the account, the audience, the creatives, and customers’ decision-making timelines. Everything you learn in September costs less than it would cost to learn in November, and in November there won’t be enough time to learn it anyway.
By the end of the month, some answers should already be on the table. Which hooks actually stop the scroll in the first few seconds? Which messaging angles generate leads, and which generate qualified leads, which are not at all the same thing? Which landing page converts best, and on which device? Which offer is perceived as interesting and which feels like just another discount? And above all, how much time typically passes between someone’s first interaction with the brand and conversion?
That last question is the most underestimated, yet it determines the entire Q4 calendar. If a customer takes an average of three weeks from seeing their first ad to making a purchase or requesting a quote, then the people who need to convert during Black Friday week have to be reached for the first time in early November, and the campaigns reaching them should already be stable before then. Anyone who doesn’t know their conversion timeline turns campaigns on when everyone else does, only to discover in December that November sales were seeded too late.
September is also the right month to grow engagement audiences. Video campaigns that accumulate views, content that generates profile interactions, lead magnets that bring people onto a list: these activities rarely show an immediate return in that month’s report, but in November they become the pool from which you can draw at a lower cost than advertisers who never built that pool.
October: The Month That Decides Black Friday
If September is about building, October is about choosing. It’s the last month when auctions are still relatively accessible and the first month when the decisions you make have a direct impact on November’s results. It’s not business as usual, but it’s not an early Black Friday either. It’s the month of pre-scaling.
The first activity is creative testing. October is the time to compare different angles, formats, openings, and isolate the ones that actually work, with enough data to trust the result. The reason is simple: every test costs more in November, and every significant change to a campaign that is performing well risks sending it back into the learning phase during the very week when it should be at peak performance. If you’re still wondering on November 20 whether video A or video B performs better, you’ve already lost part of your Black Friday.
The second activity is building lists of people who want to be notified. Early access, waitlists, exclusive access to offers before the public launch: the format varies depending on the business, but the logic remains the same. Every person who joins an email or WhatsApp list in October is someone you can reach in November without going through the advertising auction at its most expensive point. There is, however, a new factor to account for. As of October 1, 2026, Meta changed the WhatsApp Business pricing model, and service messages beyond the monthly free allowance are now charged. Building the list remains a smart choice, but it needs to be done with a clear understanding of how many messages you’ll send and how much it will cost to manage the conversations that follow.
The third activity is checking tracking. Conversions API on Meta, Enhanced Conversions and Consent Mode on Google, correctly configured events, deduplication between pixel and server: these checks are boring, but discovering a tracking error in November means optimizing the entire seasonal peak based on incorrect data. Alongside tracking, you need to prepare automations, CRM flows, email sequences, and segmentation for the contacts that will come in over the following weeks.
There’s one sentence that captures the meaning of October better than any checklist: November isn’t the time to learn. It’s the time to scale what you’ve already learned.
November: Increasing the Budget Isn’t Enough
November arrives, and with it comes the most common temptation: increase the budget. It seems obvious because demand is rising, so it feels natural to spend more to capture it. The problem is that spending more doesn’t mean getting more in the same proportion. The way to see this is to stop looking only at average cost per lead and start looking at marginal cost.
Average cost per lead is the number that appears in reports: total spend divided by total leads. Marginal cost per lead is something else, and it’s much more useful for making decisions. It’s what the additional lead costs you when you increase the budget. Let’s use intentionally simple numbers. With $100 a day, an account generates leads at $18 each, or roughly 5.5 leads. Increasing the budget to $150 raises the average cost to $19 and brings the number of leads to almost eight. Those additional $50 brought in just over two leads, at roughly $21 each. At $200, the average cost reaches $23, but the number of additional leads compared with the previous step is less than one. Those additional $50 bought leads at roughly $62 each. At $300, the average cost reaches $31, a figure that may still look manageable, but the final $100 produced just one additional lead, at a marginal cost of around $100.
Looking only at average cost, the move from $18 to $31 might look like a manageable deterioration. Looking at marginal cost reveals that part of the budget is buying leads at five times the original price. The right question in November isn’t whether Black Friday converts. It’s how much margin the last lead you’re buying leaves you, and whether that margin still exists.
Several practical rules follow from this. The first concerns timing. Campaigns intended for Black Friday should launch two or three weeks in advance so they can exit the learning phase before the peak arrives. On Meta, an ad set needs to accumulate a sufficient volume of optimization events, on the order of 50 per week, before stabilizing, and launching one on the Monday of the peak week means asking it to learn precisely when learning is most expensive.
The second concerns how you increase the budget. Algorithms interpret spending changes as signals about demand: if a budget triples overnight, the platform tries to spend it, and to spend it, it may buy into more expensive auctions as well. This is the same mechanism that, in August, led to lowering thresholds on days when the platform couldn’t spend efficiently, viewed from the opposite direction. Gradual increases spread across the weeks leading up to the peak give the algorithm time to find new opportunities without rushing into the most expensive ones.
The third concerns creative. During Black Friday week, winning creatives should be left to work. You can introduce variants in a controlled way, but completely overhauling ads that are performing during the highest-volume days is the fastest way to lose stability precisely when you need it most.
The fourth concerns Google. For short, predictable events such as Black Friday, Google Ads offers seasonality adjustments, which tell automated bidding strategies that the conversion rate will change significantly during a specific window. They are designed for short periods, ideally no longer than a week, and should be used carefully: they communicate an expected change in conversions to the algorithm, but they don’t replace the work of managing budgets and targets.
Don’t Run Black Friday? You’ll Pay for It Anyway
So far, this reasoning seems to apply primarily to ecommerce businesses, but there is a category of companies that experiences Black Friday without participating in it at all. Aesthetic clinics, professional practices, local services, B2B lead generation companies: they don’t run promotions, they don’t have a shopping cart to fill, and yet they see their cost per lead rise in November just like an ecommerce business does.
The reason is that it’s the same auction. A woman between 35 and 55 living in Milan who follows beauty content is being targeted at the end of November by every cosmetics brand, every beauty retailer, and every fashion ecommerce company with a budget to spend during those days. The aesthetic clinic trying to generate consultation requests is buying the same impression, at the same time, at a price inflated by competitors selling completely different products. And often, that same person is thinking more about Christmas gifts during those days than booking a treatment.
The risk for these businesses is twofold. The first is economic: leads become more expensive during a month when the willingness to request a quote may actually decline. The second is decision-making: the client sees the cost per lead rise, assumes the campaigns have stopped working, and asks to turn them off, even though the problem isn’t the campaigns. It’s the calendar.
The answer isn’t to shut campaigns off. It’s to shift the strategy. Move part of lead generation into October, when the same impression costs less. Allocate a larger share of the November budget to warm audiences, which tend to withstand rising costs better. And for many service businesses, build a Christmas-related offer, such as gift cards for treatments or consultations, turning an unfavorable period into a sales opportunity. The takeaway is simple: if your industry doesn’t do Black Friday, Black Friday still needs to be part of your media plan.
Black Friday: Acquisition and Monetization Are Two Different Jobs
Once you reach the peak week, which this year runs from Friday, November 27 through Cyber Monday on Monday, November 30, it’s worth taking a step back and looking at the entire system. This is where the most common evaluation mistake happens: judging every channel on its own.
Every advertising tool has a different job. Meta prospecting campaigns create new demand by putting the brand in front of people who weren’t looking for it. Google Search captures demand that already exists, from people actively searching for a product or service. Retargeting, email, WhatsApp, and CRM monetize the demand built over the previous months. Brand searches capture some of the demand generated elsewhere, perhaps from a Reel someone saw three weeks earlier or word of mouth from a satisfied customer.
Looked at individually, these channels create a misleading picture. Retargeting will always seem the most efficient because it works with people who are already convinced. Brand search will look miraculous because it captures demand that someone else created. Prospecting will look the most expensive because it does the hardest work without necessarily getting credit for it in attribution. And the business owner who cuts prospecting because it costs too much may discover a few weeks later that retargeting and brand search have stopped performing too, because nobody was feeding the pool they depended on.
That’s why in Q4, more than at any other time, the number to watch is the aggregate result. MER, or Marketing Efficiency Ratio, compares total revenue with total marketing spend without relying on individual platform attribution, which tends to count the same sale multiple times. Total divided by total. It’s less precise when evaluating a single ad, but much more honest about the company’s overall result, and it’s the number that actually tells you whether Black Friday generated profit or simply generated revenue.
From Black Friday to Christmas
The risk after Cyber Monday is to consider the game over. In reality, everything that happened in November is the starting point for December. Every person who bought during Black Friday has entered the customer list, generated a conversion signal that helps the algorithm find similar people, and can be reached with a complementary offer or an invitation to come back. Anyone who visited without buying is a warm audience for Christmas campaigns, with a different message: no longer the discount, but the gift, on-time delivery, or a solution for someone who is running late.
For lead generation businesses, the logic is similar, although the timing is different. Leads collected in October and November shouldn’t be left to cool off in the CRM. They need to be worked in December and, in many cases, January, when auctions often become less crowded again and people’s attention shifts from holiday purchases to projects for the new year.
Seen this way, Black Friday stops being an isolated event and becomes an accelerator, accumulating customers, data, and audiences that can make the following month more efficient. Those who treat it as one extra sales day capture only part of its value. Those who treat it as the final step in a process that started in August also capture its effects in December and January.
Three Checks to Run on Your Account Today
If you’re reading this article in September, the advantage is still entirely yours to build, and there are three checks worth running immediately, before any creative plan or budget forecast.
The first concerns your warm audiences. How large are your engagement, video-view, and website visitor audiences from the last 90 days, and how much have they grown compared with July? If they’ve remained flat, August was a missed opportunity and September needs to make up for it.
The second concerns tracking. Are Conversions API, Enhanced Conversions, and Consent Mode actually working, and are the platform numbers reasonably aligned with those in your CRM or business management system? It’s much better to discover a discrepancy now than in mid-November.
The third concerns timing. What is the average time between first contact and conversion in your account, and what does that mean for the date your Black Friday campaigns need to launch?
None of these three questions requires increasing the budget by a single dollar. And that’s precisely what separates a Q4 that gets taken by surprise from one that has been prepared: the difference isn’t visible on Black Friday itself. It’s visible in the weeks when almost nobody else is thinking about it yet.
Frequently Asked Questions About Black Friday Ads Strategy
When should Black Friday campaigns launch?
Campaigns that need to convert during Black Friday week should launch at least two or three weeks in advance so they can exit the learning phase before the peak. The preparation work starts much earlier, though: September is for building audiences and data, October is for testing creatives and offers and checking tracking.
Does retargeting still work in 2026?
Yes, but no longer as a list based solely on the pixel. The strongest retargeting today is built on first-party data: video and profile interactions, CRM and customer lists, email contacts, and WhatsApp contacts. On increasingly automated platforms, custom audiences function more as signals for the algorithm than as rigid boundaries.
Does CPM always increase in November?
Not uniformly. Market data shows significant cost increases on Google during Black Friday, while aggregate Meta averages can actually decline year over year, with many individual advertisers still paying more. What matters is the performance of your own account and industry, not the market average.
How much should I increase the budget for Black Friday?
There is no percentage that works for everyone. The right reference point is marginal cost per lead or acquisition: increase the budget as long as the additional lead or sale remains compatible with your margins, and do it gradually so you don’t push the platform into buying the most expensive auctions.
