← All posts

23 Mar 2026 · checkout · 2 min read

Where the 70 percent cart abandonment number comes from

Quote it in a meeting and nobody blinks. Around 70 percent of online shopping carts get abandoned. It appears in pitch decks, agency audits, and every checkout redesign brief written since about 2012. Few people can say where it comes from.

The source is Baymard Institute, a Copenhagen research firm that has run checkout usability studies since 2009. Their headline figure is an average of 49 separate studies, and it currently sits at 70.19 percent. That tidy average is doing a lot of quiet work.

What the average is made of
  1. 70.19%the aggregate average, as Baymard currently publishes it
  2. 49separate studies behind the average
  3. 56–81%the range across those individual studies
  4. ~6 in 10US abandoners who say they were never buying in the first place

Baymard Institute, Cart Abandonment Rate Statistics, updated continuously.

Look underneath and the tidy number dissolves. Individual studies run from the high 50s to the low 80s depending on the year, the market, the device mix, and on how each study chose to define an abandoned cart in the first place. A cart with one item added by a bored commuter counts the same as a loaded cart killed by a $14 shipping fee at the final screen.

Baymard's own survey work explains most of the gap. Asked why they left, about six in ten US shoppers say they were never buying. Browsing. Price checking. Using the cart as a wishlist with extra steps. No checkout redesign reaches those people, and budget spent chasing them is wasted.

The slice you can win back is the remainder, the shoppers who wanted the thing and hit a wall of surprise costs, forced accounts, or a card form that felt wrong. Baymard prices that recoverable slice at around $260 billion a year across the US and EU. If the estimate is off by half, it is still a very large number attached to fixable problems.

Two habits make the statistic useful instead of decorative. Measure your own abandonment with one consistent definition, because the published 70 percent says nothing about your store. Then segment it by traffic source. Paid visitors abandon differently from email subscribers, and a blended rate hides whichever channel is bleeding.