Q4 attribution checkup: what to check before Black Friday

Run your Q4 attribution checkup in October: Black Friday 2026 is Friday 27 November and Cyber Monday is 30 November, and a tag fixed after the first peak order only helps the post-mortem. In October, a broken tag is a small fix. In Cyber Week, it costs the data you most wanted.

This is the quarterly checkup, timed for the peak. Think of it as the dental checkup, not the emergency appointment: most of what follows will turn out to be fine, and October is the last month when finding out is cheap.

What should I check in my attribution before Black Friday?

Six things, in this order. None of them needs a new tool.

The October checklist:

1. Campaign tagging for the promotion. Decide the Black Friday campaign name now, and give every promotional link the same utm_campaign value on paid, email and affiliate links alike, so the whole promotion reads as one row in GA4. Build the links in the UTM builder and check which channel each one lands in before anything is scheduled. An email written in a hurry the night before is how an untagged link gets out.
2. One UTM scheme across paid, email and affiliates. Use the same UTM parameters conventions everywhere, with mediums GA4's default channel rules recognise, such as email for email and affiliate for affiliates. A link GA4 cannot match to any channel lands in Unassigned, and a paid link with the wrong medium can land in an organic channel. The UTM coverage calculator shows how much of your traffic arrives tagged today, and the Black Friday number nobody checks covers why that matters most in a peak week.
3. How many days your multi-channel buyers take. In GA4, open Advertising, then Attribution paths, pick your purchase key event, and read days to key event for paths with two or more channels (how to measure the lag). Count back from 27 November by that many days: that is the date by which spend meant to open Black Friday paths has to be live. If your number is two weeks, the date is Friday 13 November. If your lag is longer than the peak, spend first switched on in Black Friday week reaches buyers who will order after it.
4. Discount depth against break-even ROAS. A deeper discount raises the ROAS a channel has to clear before it makes money, and it does so before the first ad runs (the loss you book in advance). Enter the order value and gross margin you will have after the discount, with your shipping, fees and returns, in the break-even ROAS calculator, and use the result in place of last year's target.
5. What a holdout can answer in a peak week. Decide it now, before the peak, using the next section.
6. What to leave alone the week before. Agree the freeze in October, so nobody argues about it in Black Friday week.

Every ad platform will grade its own homework on Black Friday, inside its own window, and each one will find that it did well. Tags you control, a scheme you wrote down and a lag you measured are how you walk into December with numbers that are not theirs.

What can a holdout tell you in a peak week?

A holdout test switches a channel off in some regions and compares total orders there with the regions where it stayed on. In a peak week that comparison still works, because both groups live through the same week. It can tell you what the channel added to orders in that week, under those prices.

What a peak-week holdout cannot tell you:

1. What the channel does in an ordinary week. Peak buyers come for the discount, and some had decided before the ad reached them, so a peak-week result describes the peak.
2. Anything measured against October. The comparison is between regions in the same week. Last month is not a baseline for Black Friday.
3. The buyers the channel reached before the test began. Switching the ads off does not undo the paths they opened in early November, and on a long lag some of those buyers order during the test in both groups. The holdout sees that week's ads, not the channel's whole contribution.
4. Anything at all, if the week is too short. The holdout test planner shows how many days your order volume needs, and how many orders and how much revenue the held-out regions may give up. If it needs more days than the peak lasts, the peak cannot answer the question.

A holdout that should inform the Black Friday budget has to finish before the peak starts, and the last day to start a holdout test before Black Friday does that count.

What should I leave alone the week before Black Friday?

From Friday 20 November, change nothing that decides how a sale is counted. A change in that week lands inside the post-mortem, and nobody will be able to separate it from the peak.

Leave these alone from 20 November:

1. GA4 settings. Key events, channel groups, attribution settings and data filters.
2. Your UTM scheme. New links, yes. New conventions, no.
3. Tags and consent. The tag manager container, the consent banner, the checkout scripts and the theme apps that touch them.
4. Ad platform attribution settings. A different window in peak week makes this year's numbers incomparable with last year's.
5. Tests. A peak-week holdout decided in October runs as planned. Nothing new starts.

If something breaks anyway, fix it, and write down what changed, with the date and the hour, so the post-mortem can read the before and the after separately.

Where a read fits

Later, if you would rather not work through paths and channels by hand before the peak, a causal attribution read like Causality Engine's is built for that October slot. It takes one GA4 Attribution paths export, where 12 months or more of history reads best, and shows what each channel caused next to what last-click gave it, with Direct split back to the channels that sent those buyers. Every channel gets a data-health score from 0 to 100 and a next step, and the read shows your paths to purchase, with receipts that check every number against the export. It reads what GA4 recorded, so the tags, the discount and the freeze stay on your list. It takes 1 to 2 minutes and costs EUR 99 once per upload, excluding VAT, with a full refund within 30 days, no questions asked. The quarterly attribution checkup sets out how a read fits the quarter. The first finding is free: your browser works out how many days buyers who saw two or more channels take to convert, the number step 3 needs, and the file never leaves your machine.

Source, accessed 28 September 2026: Default channel group (Google Analytics Help).