The 3-day sale: how a D2C brand booked 86% of a month in one weekend

In October 2026, a streetwear client of ours ran a sitewide sale that lasted exactly three days — and booked 86% of the entire previous month's store revenue inside the window. Daily revenue across the sale window ran at roughly three times the prior month's average, and return per rupee of ad spend came in at nearly three times the prior month's level. (Full context, including how the account is measured, is in the Deadbear case study.)

Nothing about it was lucky. Here is the playbook, piece by piece.

1. A hard window, loudly stated

Three days, dates in the creative, no quiet extensions. The urgency is the product: a buyer who has been circling a full-price purchase for weeks needs a reason to decide now, and a visible deadline is that reason. Brands that extend their sales teach their audience that deadlines are fiction — and their next sale pays for the lesson.

2. A proven skeleton, not a new build

The campaign was a duplicate of the structure that had already worked in the previous festive season: same objective, same bidding, broad delivery. Launching a novel structure into a three-day window means spending day one — a third of the sale — in learning. Duplication carries the account's accumulated signal into the window instead.

3. An offer you can say in one line

Sitewide, up to half off, three days. No codes to remember, no category exclusions to explain in the comments. Every word of creative then works on desire instead of terms and conditions.

4. A cart-threshold bonus

An extra discount above a cart value just beyond the typical order turned single-item orders into two-item orders. Deep-discount windows compress margins; raising the average order value inside the window is how the economics stay healthy.

5. Multiple creative routes, broad delivery

Four distinct creative treatments of the same offer — bold type, product photography, colourway variants — ran simultaneously into a broad audience. Modern delivery systems reward exactly this: give the machine genuinely different ads and let it find the buyer for each.

6. Honest measurement, afterwards

The sale's return was reported next to the evergreen baseline, never instead of it. Sale economics are real — the revenue banks — but they are not the account's new normal, and a report that annualises a spike is a report that gets someone fired in March. Spend was read from the platforms, revenue from the store, and the window was judged as a window.

What transfers to your brand

The specific offer is the least transferable part. What transfers: the hard time-box, the duplicated structure, the one-line offer, the order-value mechanic, creative diversity into broad delivery, and a measurement habit that keeps the spike and the baseline in separate columns. If your sales keep underperforming, the gap is usually in one of those six — not in the discount percentage.