The night sale is a very common promotion in ecommerce, and almost nobody measures what it does beyond the evening itself. We took 9 night sales apart, hour by hour, to see when the money actually comes in and what happens to the days around it.
A night sale is a short, sharp promotion: it opens in the evening, closes the next day, and offers a discount that does not exist the rest of the time. In the US and UK it is often called a flash sale or a 24-hour sale. The logic is simple. When there is a clock running, a customer who has been putting off a purchase decides now.
Everyone who runs an online store knows a night sale "works". The question we ask before every one is how much, when, and at what cost. A night sale gives a discount on every order that comes in, including orders that would have come anyway, so it matters what it really adds.
We took 9 night sales we ran for online stores between July and October 2026: 1,296 orders during the sale hours themselves. For each sale we compared revenue during the sale hours with exactly the same hours, on the same weekday, in the weeks before. A week that had another short promotion in it was left out of the comparison. Revenue is the total of paid orders, after cancellations, the same way every Preneuriat dashboard counts it.
At the median, a night sale brought in 10.6 times the revenue of the same hours in a normal week.
The range is very wide: from 2.4x to 30x. In our experience two things explain most of the gap. The first is how many people know the sale is happening: the email list, a WhatsApp or SMS community, and paid media that starts before the doors open. The second is how quiet the store is on a normal evening. A store that sells little in those hours sees a high multiple even on a modest number of orders.
So the multiple alone does not tell you whether the sale was good. It tells you how different the night was from routine.
A simpler way to grasp the size: how many ordinary days of revenue came in during one night.
At the median, one night brought in as much as 9 ordinary days, and made up 20% of all revenue in the month it happened. In the strongest sale it reached 32% of the month.
That is power, and it is also a risk. A store that leans on a night sale month after month to hit its target is teaching its customers to wait for the discount. A night sale works best as an event, not a habit.
The name is misleading. Break the revenue down by hour, and most of the money does not come in at night.
There are two waves. The first arrives two or three hours after opening, around 9 pm, when people are home and pick up their phones. The second arrives the next morning, as the sale is about to close. In between, from midnight to 7 am, only 12% of the revenue came in at the median.
The morning wave is no smaller than the evening one. The last three hours of the sale brought in 25% of the revenue at the median, and the first three 18%. People who decide at the last minute decide in the morning, over coffee.
A night sale is really two sales: an evening when people discover it, and a morning when they realise it is ending.
We expected a strong discount to make people buy more per order. It barely happened. The average order during the sale was just 6% higher than a normal order, at the median. In 7 of the 9 sales it moved less than 10% either way.
In other words, a night sale brings more buyers, not buyers who buy more. If a bigger basket is also the goal, it has to be built into the offer, for example a discount that grows with the basket or a second item at a lower price. What works and what does not is in our article on how to increase average order value.
The common fear is that a night sale steals sales from the days before and after it: people wait for it, and afterwards it goes quiet. In our data, that is not what happened.
In the 72 hours before opening, revenue was 21% above the same hours in a normal week at the median, and it fell by more than 5% in only 2 sales. In the 72 hours after closing it was 44% higher, and in 6 of 8 sales it stayed above normal.
A night sale is a short promotion that opens in the evening and closes the next day, with a discount that does not exist the rest of the time. Elsewhere it is often called a flash sale or a 24-hour sale. In the night sales we measured, revenue during the sale hours was 10.6 times the same hours in a normal week, at the median.
What matters is that the sale includes a morning: the last three hours brought in 25% of the revenue at the median, more than the first three.
At the median, one night brought in as much as 9 ordinary days and made up 20% of the month it happened in. The range is wide, from 2.4x to 30x the same hours in a normal week. In our experience it depends mostly on how many people know the sale is happening, and on how quiet the store is on a normal evening.
In our data, no. In 6 of 8 sales, revenue in the 72 hours after closing was above normal, 44% higher at the median. This is an observation, not proof: those days also include paid media, reminders and sometimes an extended sale.
Every growth playbook says the same thing: find your hero products and put everything behind them. We opened 31.5 thousand orders to see what heroes actually do. They carry the revenue, open the door to new customers and fill bigger baskets. They do not, on their own, bring customers back.
Daniel Hirsh8 October 202610 min readMost advice on raising average order value is a list of tactics. We opened 42,570 orders from online stores to see what the basket actually looks like, and which levers the data really supports.
Daniel Hirsh29 September 202610 min readEcommerce is easy to define and harder to run. This guide covers what it means, how an order travels from the first click to the doorstep, the main types of ecommerce, what a business needs behind the store, and what 42.6 thousand real orders show about how online stores make and grow their revenue.
Daniel Hirsh8 October 20269 min readEvery Preneuriat client gets a live dashboard: store revenue, all media spend, real ROAS, updated all day. The first call goes both ways.
Check the fit
