Christmas, measured
Did the Christmas campaign sell, or did Christmas?
December would have been your best month anyway. Measure what the campaign added on top of the season, and what January gave back, on your own sales file.
Invitation only while in beta. Tell us who you are and we get in touch.
Weekly sales, with and without the Christmas campaign
December to February, sample retailerAn illustration on one of our sample markets. On your file the line without the campaign is estimated from your own years, and every number comes with its confidence interval.
Four ways a Christmas campaign flatters itself
The report says December beat November by a third. Most of that third would have come without a single spot, and none of these four is the campaign.
The ordinary week
A December week compared with a November one gets the whole season as a gift. The campaign is measured against December without it, not against the rest of the year.
Black Friday in the same report
Cyber Monday falls in the first week of December. Its sales belong to Black Friday, and a December total that includes them pays the campaign for another event.
January
Some of what December sold, January would have sold: purchases brought forward, gift cards spent later. The weeks after belong to the measure.
Last year is not a baseline
Christmas falls on a different weekday every year, a Friday in 2026 and a Saturday in 2027, and the number of shopping days before it changes with it.
From the report to the campaign
Start from what the campaign weeks sold over an ordinary week, then take off what the season brings every year and what January gave back. On the sample retailer, of the December uplift was the campaign. The rest was Christmas.
Whether that paid depends on what the campaign cost, and that is the number to set next to the spend, not the one in the December report.
What December was made of
sample retailer, 2026The path in tea
Three analyses, on the same file, each answering one part of the question.
-
Seasonality & Event Impact
The year read as a profile, month by month, with the national holidays, Christmas Day among them, loaded from the calendar. What the season brings is measured before anything else gets credit. How it works
-
Contribution Decomposition
The campaign spend by channel enters next to the season, promotions and price, so the December weeks are split between them instead of all going to the campaign. How it works
-
Lag & Carryover
How long the December spend kept working: the adstock and its half-life, so the January weeks are read with the campaign’s tail in them. How it works
What your file needs
Two Christmases at least
Daily or weekly rows over two years, three is better. One December is one observation.
date week
The number to explain
Sales, orders, units. One column, the same on every row.
sales orders
What you spent
Media spend by channel, week by week, in euro or GRPs. Promotions and price too, if they moved.
tv_grp social_spend
National holidays load in one click for Italy, Germany, the UK and the US, and the Data Scout tells you before the run if the file is too thin for the question.
Christmas incrementality, in a paragraph
The incremental effect of a Christmas campaign is the difference between what December sold and what the same December would have sold without the campaign. The second number is never observed, and it is far above an ordinary week, so it has to be estimated from the years you have, with the season read alongside the spend. The season is not the campaign’s rival: it is its baseline.
What comes out is the campaign’s contribution week by week, with its interval, and the January weeks it borrowed from. Set against what the campaign cost, it says whether next December should spend more, the same, or less.
How the model reads a calendar, and where it misleads: Seasonality and event impact, explained.
How do I measure the impact of a Christmas campaign?
Against what the same weeks would have sold without it, which is far above an ordinary week. December rises every year without anyone spending a euro, so the campaign gets credit only for what sits above the season, and the January weeks that sell less because December sold more are part of the bill.
Why not compare December with November?
Because the difference between them is mostly the calendar. On the sample retailer the campaign weeks sell €192k more than three ordinary weeks would; €136k of it is the season, which comes every year. A comparison that cannot separate the two credits the campaign with Christmas.
What about Black Friday in the same file?
It is a different event and is measured as one. Declared on its own dates, Black Friday takes its weeks and the Christmas campaign takes its own; when two windows largely overlap, tea flags it, because two events on the same days cannot be told apart.
How many Christmases do I need?
Two at least, three is better: one December is one observation of the season and the campaign together. With a single year the model cannot tell how much of December would have come anyway.
What does it cost?
During the beta, nothing: every run is measured and no balance moves. After the beta, every plan includes a monthly token allowance: 150 on Free, 3,000 on Pro, 6,000 on Plus, 20,000 on Business. What is left of it expires when the month renews; tokens you buy never expire. The plan never limits which questions you can ask.
Next December, fund the campaign, not the calendar.
Drop the file, separate the season from the spend, keep the number that holds. Free while in beta, by invitation.
Join the betaLearning
A report says what happened. An analysis says why.
Most of what lands on a marketing desk is reporting. Useful, and a different job. The blog tells one case a day where the difference decides the answer.
Sales were up 11% in April, TV ran six weeks. What happened, by channel, side by side.
4.2 of those points came from TV, 5.1 from Easter moving. What caused it, and what the next euro does.
When a budget optimiser says move 60% into one channel
Rolling cross-elasticity: checking whether a rivalry is stable
