Black Friday 2026: how to prepare your online store with data

During Black Friday 2025, Brazilian ecommerce processed 16.5% more orders than the year before. Revenue, however, grew only 7.8%. Run the numbers and you reach the headline almost nobody wrote: every order got cheaper. More boxes leaving the warehouse, more shipping, more customer service, and less money per order.
That changes how you should prepare for Black Friday 2026, which falls on November 27. If you plan around revenue alone, you can break your sales record and still close the month with worse margins than October. If you plan with data, you start now, seven weeks out, and lock in discounts, inventory and ad budget before the media auctions heat up.
How to prepare an online store for Black Friday with data: in October, review last year's Black Friday by day, hour and product; size inventory from your real sales curve; calculate break-even ROAS with the discount already applied; and start warming up your audience. In November, track an hourly dashboard with a handful of metrics and pre-agreed actions.
What Black Friday 2025 data says about 2026
According to Confi.Neotrust, Brazilian ecommerce generated R$ 10.19 billion between Thursday and Sunday of Black Friday 2025, up from R$ 9.39 billion in 2024, across 21.5 million orders. Divide one by the other and the average order value lands near R$ 474. The same math for 2024 gives roughly R$ 509. That is a drop of almost 7%.
Two more details matter for planning. First, Thursday outsold Friday itself, which breaks the historical pattern. Second, Saturday fell about 10% year over year. The event is spreading out, and a deal that goes live Friday at 8 a.m. is already late for part of the audience.
For anyone selling online, that means three things:
- Order volume tends to grow faster than revenue, so cost per order (shipping, packaging, support) weighs more.
- The useful window runs Thursday to Monday, not just Friday.
- Margin, not revenue, decides whether the event was worth it.
For the broader market picture, see our post on ecommerce data in Brazil in 2026.
Step 1: review your last Black Friday
Before setting any goal, open your November 2025 data and answer four questions. Which days and hours concentrated sales? Which products sold, and which sat still despite the discount? How did the week's average order value compare with October's? And which channel brought profitable orders, not just orders?
The question about products that sat still is usually the most useful. Almost every store has that item that got 30% off and sold the same as in a normal week. A discount without demand only takes margin from people who would have bought anyway.
Also separate new sales from pulled-forward sales. If December 2025 sales dropped compared with December 2024, part of Black Friday was the same customer buying earlier and cheaper. The weekly report won't show that; only the month-over-month comparison will.
What if you don't have reliable history?
New store or messy data? Use the average of October's four weeks as a baseline and work with scenarios: conservative (2x the average week), likely (3x) and optimistic (4x). The multiplier is an assumption, not research, and it helps you size inventory and support without guessing blind. If last year's campaigns had no UTM tags, fix that now: our guide to UTM campaign tracking covers the minimum standard.
Step 2: size inventory from your real sales curve
Running out of stock mid-Black Friday is worse than not selling: you paid for the click, the customer arrived and found "sold out". The fix is to calculate coverage per SKU, at least for the 20% of products that drive most of your revenue.
Take a home goods store. A cookware set sells 30 units in a normal October week. During last year's Black Friday week, it sold 4.2 times that. The projection for the week is 126 units, plus a 20% safety buffer, so around 150. If the supplier needs 30 days to deliver, the order must go out by the end of October. November is too late.
The reverse logic also applies: a product with high stock and low turnover is the natural candidate for an aggressive discount. You turn idle capital into cash instead of discounting the item that would sell on its own.
Step 3: calculate break-even ROAS with the discount applied
This is where most online stores lose money without noticing. The Black Friday ROAS target is often the same as the rest of the year, but the discount changes the whole equation.
Here is a simulated example. A product sells for R$ 200, with R$ 100 in cost of goods and 15% of the price going to taxes, fees and subsidized shipping (R$ 30). That leaves a contribution margin of R$ 70, or 35%. Break-even ROAS is 200 ÷ 70 = 2.86. Below that, every ad-driven sale loses money.
Now apply a 20% discount. The price drops to R$ 160, cost of goods stays at R$ 100, and fees become R$ 24. Margin falls to R$ 36, or 22.5%. The new break-even ROAS is 160 ÷ 36 = 4.44. A 20% discount raised the minimum return your campaign must deliver by 55%.
There is another way to read it: to generate the same R$ 70 of margin as one full-price sale, you need almost two discounted sales (70 ÷ 36 = 1.94). That is exactly the pattern Black Friday 2025 showed in aggregate, with more orders and a lower ticket. If these calculations are new to you, our posts on contribution margin and on break-even ROAS and realistic targets walk through the formulas.
Two decisions follow. First, set discounts per product, not a single store-wide percentage. An item with a 60% margin can take 25% off; an item with a 25% margin can't even take 15% off with paid media behind it. Second, set your campaign ROAS targets for the week using the new number. Optimizing to October's ROAS during Black Friday means paying to sell.
Step 4: build your warm-up funnel before the auctions heat up
During Black Friday week, everyone competes in the same Meta Ads and Google Ads auctions, and CPMs usually rise. That's why the audience you build in October and early November is worth more than the one you try to buy on the day.
- Owned list: a "notify me" or early-access page that captures email and WhatsApp. It's the channel whose price doesn't spike on Friday.
- Remarketing audiences: people who viewed product pages or abandoned carts in the last 30 to 60 days. Feed this audience with content now.
- Offer tests: use October to test creatives and offer formats (free shipping, tiered discounts, gifts). During the week, only scale what has already proven itself.
And remember the 2025 signal: Thursday was strong. Having a warm list and an offer ready to release to subscribers on Wednesday or Thursday can capture the buyer who won't wait for Friday.
Which metrics should you track on Black Friday?
A dashboard with 25 charts on Black Friday is a dashboard nobody reads. What works is a small set of metrics, refreshed hourly, each tied to an action agreed in advance.
| Metric | What it signals | Pre-agreed action |
|---|---|---|
| Campaign ROAS vs. break-even ROAS | Whether ads are making or losing money | Below break-even for 3 hours straight: cut budget |
| Site conversion rate | Checkout, pricing or stock problems | Sharp drop: test checkout immediately |
| Stock of top 20 SKUs | Stockout risk | Below 20%: pause that item's ads |
| Average order value per hour | Whether discounts are shrinking baskets | Falling: switch on free shipping above a threshold |
| Cumulative estimated margin | The day's real result | Compare with target at noon and 6 p.m. |
The "below break-even for 3 hours" rule is a starting point, not a law. Low-volume stores need longer windows, because a few sales can make ROAS swing wildly from one hour to the next.
What about after Black Friday?
The December review is what prepares you for 2027, and it's the one almost everyone skips. Three questions are enough. Did the week's total margin beat a normal week plus the sales pulled forward from December? How many new customers bought again within 60 days? And which discounted products didn't increase volume? A customer who only shows up at 30% off has a very different LTV from one who came back in January at full price.
Frequently asked questions
When is Black Friday 2026?
Black Friday 2026 is on Friday, November 27. Since Thursday outsold Friday in Brazil in 2025, it makes sense to treat November 26 to 30 (through Cyber Monday) as your sales window.
How much did Brazilian ecommerce make on Black Friday 2025?
According to Confi.Neotrust, R$ 10.19 billion from Thursday to Sunday, up 7.8% on 2024, across 21.5 million orders. Orders grew 16.5%, faster than revenue, which points to a lower average order value.
When should you start preparing your store for Black Friday?
In October. Supplier orders often take weeks, and discount and ROAS decisions need to be ready before the November warm-up. If you start during Black Friday week, all you can do is react.
What is the minimum ROAS for Black Friday?
There's no universal number. Your minimum ROAS is your break-even ROAS recalculated with the discount: discounted price divided by the contribution margin left over. In this post's example, a 20% discount moved break-even from 2.86 to 4.44.
How do you know if a product is worth discounting?
Check whether the product responded to discounts in past events and how much margin it has. If volume didn't rise with past discounts, or the margin can't absorb the discount plus ad spend, choose another mechanic, such as a gift or free shipping above a threshold.
Black Friday is decided in October
The day itself is execution. The outcome, profit or just revenue, is set by October's choices: which product gets a discount, how much stock to buy, what ROAS to accept. The answers live in your Meta Ads, Google Ads, GA4 and store platform data, scattered across four different places. Sherlok connects those sources and lets you ask plain questions like "which products sold best with a discount last Black Friday, and what was their margin?", returning a ready analysis while there's still time to decide.
Your AI agents already know what to do in your account.
Turn them on free, no credit card.
- No credit card
- Read-only access
- Revoke it any time