The Black Friday Post-Mortem You Cannot Run in January: By January the data behind Cyber Week is a model of itself. Meta's windows have closed, GA4's default retention has expired the events, and nobody remembers what was live on the Sunday. The post-mortem window is the week of 7 December.
Read the full article below for detailed insights and actionable strategies.
The attribution problem
One sale. Four channels. 400% credit claimed.
Reported revenue: €400 · Actual revenue: €100 · Gap: €300
The Black Friday post-mortem has its own deadline, and it is earlier than the January review it usually waits for. Meta's default 7-day click window closes on every Cyber Monday click by 7 December. GA4's default retention deletes Cyber Week event-level data by late January. The exports, the windows and the memory of what was live all decay on a schedule. Run the post-mortem in the week of 7 December, or run it on a model of the data instead of the data.
Octalysis would call the January review an Evanescent Opportunity in reverse: the reward was available in December and quietly stopped being available while everyone was on holiday.
What closes, and when
| What | Closes | Why it matters |
|---|---|---|
| Meta's default 7-day click window on Cyber Monday clicks | Mon 7 Dec 2026 | After this the platform's Cyber Week number stops moving and can be captured as final |
| GA4 event-level data for Cyber Week, default retention | Around late Jan 2027 | Explorations on last peak stop being possible; the aggregated reports remain |
| Your own record of what was live each day | Fades from the first week | Discount depth, email sends, stock-outs and creative swaps are the covariates a read needs |
| Platform definitions | Any date, undisclosed in charts | The book documents Meta's March 2026 click-definition change; the next one will not appear in a year-on-year line either |
The export list for the week of 7 December
- GA4: export the event-level data for 20 November to 7 December, by CSV or by the BigQuery link if it is on. This is the only copy that never expires and the input to any later causal attribution read.
- Each ad platform: export Cyber Week conversions under the default window and under a 1-day click window, side by side. The gap is the "touched earlier" population.
- The store: orders and revenue by day, the fact about money.
- The email and SMS platform: claimed conversions and the attribution window setting that produced them.
- A one-page diary: what discount, what stock, what creative, what send, by day. It takes twenty minutes on 8 December and is unrecoverable by February.
The three numbers to compute while the data exists
Coverage. Attributed conversions divided by orders. The Price of Being Found reports the publisher's own census at 48.6% of sessions with no resolvable source over 17 October 2025 to 2 September 2026, one company's data, not a population estimate. Yours is computable in an hour. The number nobody checks covers it.
Claim ratio. Every platform's claimed conversions, summed, divided by orders. Peak week is when it is highest.
Incremental contribution. If a holdout started by 2 October, its read is due now, against the pre-period fixed in September. If it did not, a causal read on the GA4 export gives per-channel incremental ROAS with intervals from the data you just exported, which is the closest thing to an answer the observational record can supply, and it says so in its interval.
Why January is too late even when the data survives
Even with retention extended and the exports taken, the post-mortem that happens in January answers a different question. In December the question is what to change about the plan that is about to repeat in a smaller form for January sales and in a larger one next November. In January it is a retrospective with no decision attached, and reviews without decisions are the ones that get skipped. The book's cadence is quarterly re-anchoring, with the anchor date on the dashboard, and the first week of December is the anchor for Q4.
What to do this week
- If you have to defend the number: book the post-mortem meeting for the week of 7 December now, with the export list above as its agenda. The meeting will not book itself in the last week of November.
- If you own the budget: decide today that the December read, not the January deck, is what next year's Black Friday budget will be based on, and say so to whoever prepares the deck.
The Black Friday 2026 calendar has the dates before the peak. This one is the date after it.
Dates and default settings as of 9 September 2026. The coverage census and the March 2026 definition change are from The Price of Being Found (Edition 2.10), Chapters 8 and 9, with the book's caveats.
Get attribution insights in your inbox
One email per week. No spam. Unsubscribe anytime.
Key Terms in This Article
Attribution
Attribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
Attribution Window
Attribution Window is the defined period after a user interacts with a marketing touchpoint, during which a conversion can be credited to that ad. It sets the timeframe for assigning conversion credit.
Black Friday
Black Friday is the day after Thanksgiving in the United States. It marks the start of the Christmas shopping season and is a major sales event for retailers.
Causal Attribution
Causal Attribution uses causal inference to determine which marketing touchpoints genuinely cause conversions, not just correlate with them.
Conversion
Conversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
Cyber Monday
Cyber Monday is the Monday after Thanksgiving, known for significant online shopping deals. It consistently ranks as one of the largest e-commerce sales days of the year.
Dashboard
A dashboard is a visual display of key information required to achieve specific objectives. It consolidates data onto a single screen for quick review.
Revenue
Revenue is the total income generated by the sale of goods or services related to a company's primary operations.
Related Articles
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Ready to see your real numbers?
Own the budget? Upload your GA4 export and see which channels drive incremental sales, with confidence intervals, in minutes. Have to defend it? Start with the live demo and take the read to your CFO.
Full refund if you don't see value.
Stay ahead of the attribution curve
Weekly insights on marketing attribution, incrementality testing, and data-driven growth. Written for the person who owns the budget and the person who has to defend it.
No spam. Unsubscribe anytime. We respect your data.
Frequently Asked Questions
When should I analyse Black Friday results?
In the week of 7 December 2026. By then Meta's default 7-day click window has closed on Cyber Monday clicks so platform numbers are final, and GA4's default 2-month retention has not yet expired the event-level data. Export everything that week, because the data decays on a schedule.
What data should I export after Black Friday?
GA4 event-level data for the peak (CSV or BigQuery), each ad platform's conversions under the default and a 1-day click window, orders and revenue by day from the store, the email and SMS platform's claimed conversions with its window setting, and a daily diary of discounts, stock and creative.
Why can't I just run the Black Friday review in January?
GA4's default retention deletes Cyber Week event-level rows by late January, the record of what was live has faded, and platform definitions can change on undisclosed dates. A January review also has no decision attached, whereas a December read still shapes the plan that repeats in January and next November.