Black Friday Results: How to Analyse Your Store's Week
How to analyse Black Friday results: new vs returning customers, discount depth and margin, MER vs platform ROAS, refunds, and 30, 60 and 90-day cohorts.
Wilmer
Co-founder & CEO

Black Friday results arrive as a record revenue figure and a stack of dashboards that each claim the credit. Many stores do not find out until February whether the weekend made money, who bought, and whether those buyers came back. This guide sets out how to analyse a store's Black Friday and Cyber Week results, what to put in place before 27 November so the analysis is possible, and which comparisons with 2025 still hold after Shopify changed how it counts sessions.
The short answer: judge Black Friday on orders, margin and repeat buyers, and use Arktis to read conversions by source
A useful Black Friday analysis compares orders, net sales and contribution margin with the same event days in 2025, splits buyers into new and returning, works out what the discount cost per order, sets total revenue against total ad spend rather than adding up what each platform reports, subtracts refunds as they arrive, and follows the new customers for 30, 60 and 90 days. Shopify sessions and conversion rate are not directly comparable with 2025, because Shopify changed how sessions are measured between 21 and 23 September 2026.
For seeing which UTM campaign and ad platform the weekend's conversions came from, Arktis is the clear winner for an online store. Its Revenue page shows revenue net of refunds, orders and paying customers by the source of the visit each order was matched to, and conversions are counted per UTM campaign and per ad platform. Plans are $49, $149 and $349 a month, and Growth, which adds conversion tracking and session recordings, has a 7-day free trial.
| What to measure | Where to read it | The trap |
|---|---|---|
| Orders and net sales against 2025 | Shopify Analytics, by event day | Black Friday was 28 November in 2025 and is 27 November in 2026 |
| Sessions and conversion rate | Shopify sessions reports, same bot filter | Not directly comparable across the September 2026 update |
| New and returning customers | Shopify New vs returning customers report | Returning means any earlier order, ever |
| Contribution after discount and ads | Your own sheet of costs per order | Revenue can rise while profit falls |
| MER | Total revenue over total ad spend | Platform ROAS figures overlap and can include views |
| Refunds and returns | Shopify sales reports, Arktis Revenue page | Shopify dates a return when it is processed |
| Repeat orders | Shopify Customer cohort analysis | The 90-day read lands on 25 February 2027 |
Key takeaways
Decide before the sale what you will compare, with which costs, against which days of 2025. Afterwards, read orders and net sales first and sessions last. Split revenue into new and returning customers, because a weekend that mostly discounted orders your regulars would have placed anyway is a different result from one that bought new customers. Work out contribution after the discount and the ad spend, set total revenue against total spend, and run the numbers again in January once refunds have landed. The verdict on Black Friday comes from the repeat orders of its new customers, which takes until late February to read.
Before the sale: what to set up
Most of a good post-mortem is decided in the next two weeks. Write down the comparison now: a baseline of two ordinary weeks in early November, and last year's event days matched by name, so Black Friday 2026 on 27 November sits against Black Friday 2025 on 28 November and Cyber Week against the five days from Thanksgiving to Cyber Monday.
Put a cost on each order before the discount muddies it: cost of goods, shipping, payment fees and any packaging, so contribution margin can be worked out per order afterwards. Our guide to contribution margin explains which costs belong in which line.
Keep a log of every promotion with its start and end time, the discount codes and which channel each went to, and keep ad spend per platform per day. Tag every email, SMS and influencer link with its own UTM campaign. The Black Friday tracking checklist covers testing purchase events, click identifiers and consent before the code freeze.
Compare like with like, and leave sessions for last
Shopify's session measurement update rolled out from 21 to 23 September 2026. Sessions now continue across midnight UTC and end after 30 minutes of inactivity, some sessions without a pageview are counted, such as a shopper going straight to checkout from a cart link, and identified bot sessions are filtered out of session reports by default. Shopify says orders, sales and customer counts are not affected, but that conversion rate, add to cart rate, bounce rate and other session-based metrics can change while orders stay flat.
So a conversion rate that rose from 2025 to 2026 may only mean fewer bots were counted. Shopify's advice is to treat data after the update as a new baseline, to compare orders, sales and customers alongside sessions, and, in reports that offer the Human or bot filter, to use the same setting for both periods. Bot classification only exists for sessions from 7 October 2025, so Black Friday 2025 can be filtered, but Shopify notes that the same filter still does not make sessions measured before the update directly comparable. Our article on why Shopify sessions dropped goes through the change in detail.
New versus returning customers
Shopify's New vs returning customers report splits customers by period. A first-time customer is one who placed their first order with the store, and a returning customer is one whose order history already held at least one order. Run it for Cyber Week 2026 and Cyber Week 2025 and compare the share of returning customers.
Two things to watch. Shopify's customer reports are built on each customer's entire order history, so a November new customer who orders again in December shows as a repeat customer when you open a November report later. And returning customers are usually the cheapest orders to win and may well have bought at full price a week later, so a weekend where most revenue came from them may have moved demand rather than created it.
Discount depth and what it did to margin
Adobe's 2026 forecast for US online retail expects discounts of up to 30 percent off listed prices during Cyber Week, against up to 14 percent in early November. A discount takes money straight out of contribution, so it needs more orders than it looks to break even.
Take an invented example. A product sells for $100 and costs $50 to make and ship, leaving $50 of contribution, a 50 percent margin. At 25 percent off it sells for $75, still costs $50, and leaves $25, a 33 percent margin. Each order now earns half as much, so the weekend needs twice the orders to stand still. Break-even ROAS, which is one divided by the margin, moves from 2.0 to 3.0. The break-even ROAS guide has the full table.
Now the week. In an ordinary week the store sells 400 orders at $100, which is $40,000 of revenue and $20,000 of contribution, and spends $8,000 on ads, leaving $12,000. In Black Friday week it sells 1,200 orders at $75, which is $90,000 of revenue and $30,000 of contribution, and spends $24,000 on ads, leaving $6,000. Revenue rose 125 percent and contribution after ads halved. Run this arithmetic per offer and per product line, because a sitewide percentage hits low-margin products hardest.
MER against platform ROAS
Marketing efficiency ratio, or MER, is total revenue divided by total ad spend. In the example above it fell from 5.0 in the ordinary week to 3.75 in Black Friday week. It needs no attribution, which is why it is the number to start with. Our comparison of MER vs ROAS covers the definitions.
Platform ROAS is each ad platform's own claim, and the claims overlap. Under Meta's standard attribution model, an advertiser can choose to credit website purchases within 1 or 7 days of a link click, within 1 day of an impression and within 1 day of a non-link click on the ad. Meta's help centre also says some accounts may still use earlier versions of the click-through and engage-through definitions while the change rolls out, so check the setting on each ad set before comparing with 2025. In a week when many shoppers were going to buy anyway, view-through credit can attach to purchases the ads did not cause.
A simple test: add up the purchase value each platform reports for Cyber Week and set it against your store's own revenue for the same days. If the platforms together claim more than the store took, some of their credit is shared. Our guide to why Facebook Ads and Shopify sales don't match explains the usual causes.
Refunds and returns
Shopify's sales reports define net sales as gross sales minus discounts minus sales reversals, and record returns as negative amounts on the date the return was processed. So a sales report for Black Friday shows the weekend before most of its returns, and those returns appear later on the December and January dates they are processed, not on the date of the original order.
To see what the weekend kept, look at Cyber Week's orders net of the refunds made against them since, and do it again in early January after the main return window. Our guide to ecommerce returns covers when holiday returns arrive and how to measure revenue after refunds.
Cohorts at 30, 60 and 90 days
Shopify's Customer cohort analysis report groups customers by the date of their first order and shows what they bought in the weeks or months after. It can be filtered on the first order's sales channel, marketing channel or product. Shopify also lets you build a customer segment from a first order date range, which is how you isolate the customers whose first order fell in Cyber Week.
Compare that cohort with customers acquired in an ordinary month: the share who ordered again within 30, 60 and 90 days, and what they spent. Shopify's cohort table counts in weeks, months or quarters rather than days, so weekly intervals come closest. Counting from 27 November, those reads fall on 27 December, 26 January and 25 February 2027. If Black Friday customers come back at half the rate of October's, their lifetime value is lower, and the next Black Friday should bid less for them. The LTV calculator turns those repeat rates into a value per customer.
Where Arktis fits
Arktis matches Shopify orders back to a visit and, where that visitor arrived with an ad click identifier, to the ad platform. The Revenue page shows revenue net of refunds, orders and paying customers by the source of the visit each order was matched to, over preset ranges. Because it groups revenue by the date the order was placed, the Last month preset opened in December shows November's orders net of the refunds that have arrived since, and a fully refunded order drops out.
Conversions per UTM campaign and per ad platform come from matched orders and the conversion events the tracker records, and Arktis syncs Meta campaign spend once a day. The Models view compares five rule-based models, first-touch, last-touch, linear, time-decay and position-based, per ad platform, so you can see how each platform's share of conversions changes when the first or the last touch gets the credit. The AI Assistant Traffic panel counts sessions and conversions from ChatGPT, Perplexity, Gemini and similar assistants, a channel where Adobe expects traffic to US retail sites to rise 130 percent year on year this season.
Where Arktis does not fit
Shopify orders arrive from the day the store connects, with no bulk import of earlier orders, so a store connecting this month has no 2025 Black Friday in Arktis to compare with. Orders that cannot be matched to a visit appear under Direct, and orders completed inside another app's checkout usually arrive unmatched. Arktis identifies visitors per browser and does not know who saw an ad without clicking it, so it cannot check view-through claims, only show what the click-based view says.
Get the weekend into your data
Connect before the sale so orders from the first teaser onwards reach Arktis. To see Cyber Week revenue by source net of refunds, start the 7-day Growth trial, or compare plans on the pricing page. The Cyber Week 2026 calendar has the dates, and Black Friday Facebook ads covers the Meta side of the plan.
Sources
Shopify Help Center: changes to sessions and conversion rate in Shopify Analytics, 21 to 23 September 2026 rollout, what changed, bot classification from 7 October 2025, orders and sales unaffected, comparison advice, accessed 2 November 2026
Shopify Help Center: Customers reports, first-time and returning customer definitions, entire order history caveat, Customer cohort analysis and first order date segments, accessed 2 November 2026
Shopify Help Center: Sales reports, net sales and sales reversals, returns recorded on the date processed, accessed 2 November 2026
Meta Business Help Centre: about attribution models and attribution settings, click-through, view-through and engage-through windows and the rollout note, accessed 2 November 2026
Adobe: 2026 holiday shopping forecast, US online retail, Cyber Week discounts of up to 30 percent, early November discounts and AI traffic growth of 130 percent, published 28 September 2026
Frequently Asked Questions
How do I analyse my Black Friday results?
Compare orders, net sales and contribution margin with the same event days in 2025, split revenue between new and returning customers, and set total revenue against total ad spend. Then subtract refunds as they arrive and follow the new customers for 30, 60 and 90 days. Leave sessions and conversion rate for last, because Shopify changed how it measures sessions in September 2026.
Can I compare Black Friday 2026 conversion rate with 2025 in Shopify?
Not directly. Shopify's session measurement update between 21 and 23 September 2026 changed how sessions are counted and filters identified bots by default, so conversion rate can move while orders stay flat. Shopify advises comparing orders, sales and customers instead and treating post-update data as a new baseline.
Why is platform ROAS higher than my store revenue suggests during Black Friday?
Each ad platform counts the purchases it can claim under its own attribution settings, and the claims overlap. Meta's standard settings can include purchases within 1 day of an ad impression, which can attach to purchases during Black Friday that shoppers would have made anyway. Comparing total store revenue with total ad spend, the marketing efficiency ratio, avoids that double counting.
When should I re-check Black Friday revenue for returns?
Shopify records a return on the date it is processed, so a sales report for the Black Friday dates shows only the returns processed on those dates, and most returns against its orders land later. Look at Cyber Week's orders net of the refunds made against them in early January, after the main holiday return window. A report grouped by order date, such as the Arktis Revenue page, shows that directly.
How long does it take to know if Black Friday customers were worth it?
Follow the customers whose first order fell in Cyber Week for 30, 60 and 90 days, which from 27 November 2026 means reads on 27 December, 26 January and 25 February 2027. Compare their repeat rate and spend with customers acquired in an ordinary month. Shopify's Customer cohort analysis report groups customers by first order date for this.
Free tools for this topic
From the blog
Try Arktis free for 7 days
Attribution tracking, session recordings, and AI automation in one platform. Set up in 2 minutes, free for 7 days.
Start 7-day free trialWritten by
Wilmer
Co-founder & CEO
Wilmer leads product strategy at Arktis, focusing on privacy-first analytics and attribution tracking for e-commerce brands.