The winter sales, measured
Did the winter sales add revenue, or only move it?
The first week of the sales is the busiest of January. The weeks shoppers waited for it, and the February it emptied, are quieter. Measure what the sales actually added, on your own sales file, with the season held constant.
Invitation only while in beta. Tell us who you are and we get in touch.
Weekly sales, with and without the winter sales
December to February, sample retailerAn illustration on one of our sample markets. On your file the line without the sales is estimated from your own years, and every number comes with its confidence interval.
Four ways the winter sales flatter themselves
The report says the first week of the sales was the best of the winter. All four of these sit inside that number, and none of them is new demand.
The weeks before
Between Christmas and the first day of the sales, shoppers who know the date stop buying at full price. The last week of December sinks, and the sales collect it.
The weeks after
What was bought in the sales is not bought again in February. The lift of January is partly a hole in the month that follows.
A low month anyway
January sells less than December every year, sales or not. Against December the sales look like a loss, against an ordinary January like a win. Neither is the measure.
Everyone opens the same morning
The start is set by the region in Italy and by the state in France, so every competitor discounts on the same day. The lift you measure is yours, net of theirs.
From the booked lift to the real one
Start from what the weeks of sales sold over the line, then take back what the weeks around them lost. On the sample retailer, of the booked lift was new demand. The rest was sales moved inside the winter.
That is before the discount, which was paid on every sale of the period, full-price buyers included. The margin question starts from the net figure.
Where the booked lift went
sample retailer, 2027The path in tea
Three analyses, on the same file, each answering one part of the question.
-
Seasonality & Event Impact
Declare the sales, the weeks before them and the weeks after them as events of their own. Each gets its effect with its 95% interval, measured with the calendar held constant. Windows that overlap are flagged. How it works
-
Promo Elasticity
With a column for the discount depth, the lift per point of discount comes out net of the weeks after. It needs at least three promotions in the file, the summer sales included, and says so. How it works
-
Contribution Decomposition
If media ran for the sales, the discount column enters the model next to media, price and the season, so the January weeks are split between them instead of all going to the discount. How it works
What your file needs
Two sales seasons at least
Daily or weekly rows over two years, three is better. The summer sales count too.
date week
The number to explain
Sales, orders, units. One column, the same on every row.
sales orders
The discount
Average discount depth per week, and media spend if it ran. Optional, and each one is a part of the answer.
discount social_spend
The sales dates are yours to enter, from your region’s calendar, and the Data Scout tells you before the run if the file is too thin for the question.
The incremental effect of the sales, in a paragraph
The incremental effect of the winter sales is the difference between what you sold and what you would have sold in the same weeks without them. The second number is never observed, so it has to be estimated: from the years you have, with the trend, the season and the weekday read at the same time as the sales. The weeks around the sales belong to the measure, because a date everyone knows in advance changes when people buy before it changes how much they buy.
What comes out is three numbers with their intervals: the lift in the weeks of sales, the loss in the weeks before, the loss in the weeks after. Their sum is what the sales added. When the interval around it crosses zero, the honest answer is that the file cannot tell, and that is an answer too.
How the event model reads a calendar, and where it misleads: Seasonality and event impact, explained.
How do I measure the impact of the winter sales?
Against the sales the same weeks would have made without them. January is a low month anyway after Christmas, so a lift over December is not the measure; the weeks before, when shoppers stop buying at full price, and the weeks after, when what they bought in the sales is not bought again, are part of it.
Is the fixed start date a problem?
It helps and it hurts. The start is set by the region in Italy and by the state in France, the same for every competitor, so the date is known in advance and your file can be dated exactly. It also moves from year to year with the weekday, which is why a year-on-year comparison of the first week misleads.
Does a bigger discount mean a bigger lift?
Usually, and how much is the question Promo Elasticity answers: the lift per point of discount, net of the weeks after. It needs a column with the discount depth and at least three promotions in the file.
Are the sales profitable?
The measure here is volume. The discount was paid on every sale of the period, the ones that would have happened at full price included, so a positive net lift can still lose margin. Knowing the net volume is the first half of that sum.
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.
Plan the next sales on the net, not the first week.
Drop the file, measure the sales, 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
