The Anchoring Effect

​Read online​

---

In the 1970s, Tversky and Kahneman¹ ran a simple experiment.

They asked people:

"What percentage of African countries are in the UN?"

But before answering, participants had to spin a wheel with numbers from 0 to 100.

The wheel was rigged to land on either 10 or 65.

Then came the important question:

"Is the actual percentage higher or lower than that number?"

After that, participants gave their best estimate.

Here’s what happened:

• People who saw 10 guessed 25% of African countries are in the UN.
• People who saw 65 guessed 45% of African countries are in the UN.

Same question. Different anchor.

The number from the wheel (which was totally random) pulled their answer in its direction.

This wasn’t just a one-off. Ariely, Loewenstein, and Prelec² tried something similar 30 years later.

This time, they showed students everyday products: a cordless keyboard, a bottle of wine, a box of chocolates.

Before bidding, students had to write the last two digits of their social security number.

Then they were asked:

"Would you pay more or less than that number?"

After that, they gave their actual maximum price.

Students with high social security numbers were willing to pay twice as much as students with low numbers.

We're anchored by the initial price we're exposed to.

Which is why menus in descending price order lead to 4% higher revenue.

And why this ridiculous message boosted Snickers sales:

---

This is known as anchoring

Inside the Nudge Vaults you'll find 41 more insights specifically about this bias.

The Vaults won't cost you £999 per month.*

In fact, you can preview your first 50 insights for free →

Merry Christmas folks, I hope you had a fantastic celebration. — Phill​

*Yes, that was a blatant use of anchoring.

P.S. My good friend Louis Grenier has just published an incredible YouTube series where he locks 3 freelancers in a cottage for a weekend to fix their marketing. Watch it here.​