The Anchoring Effect
The first number you hear isn’t neutral. It’s a gravitational field
1. The Number That Shouldn’t Matter
A group of people spin a wheel. The wheel is rigged — it lands on either 10 or 65.
Then they’re asked: “What percentage of African countries are in the United Nations?”
The people who spun 10 guessed, on average, 25%.
The people who spun 65 guessed, on average, 45%.
The wheel was random. Everyone knew it was random. It had no logical connection to the question whatsoever.
It didn’t matter.
The number they’d just seen pulled their estimate toward it. A meaningless anchor reshaped a factual judgment — not because people were careless, but because this is how the mind works.
This is the anchoring effect. And it operates in salary negotiations, courtrooms, pricing strategies, medical diagnoses, and every estimate you’ve made today.
2. What It Is
Anchoring is a cognitive bias first documented by Tversky and Kahneman in 1974, in the same paper that introduced the availability heuristic.
The mechanism is straightforward:
When we make estimates or decisions under uncertainty, we start from an initial value — the anchor — and adjust from there. But we almost always adjust insufficiently. The anchor pulls the final judgment toward it, even when the anchor is arbitrary, irrelevant, or known to be wrong.
The anchor doesn’t have to be credible to be effective. It doesn’t have to come from an expert. It doesn’t have to be in the same domain as the question. It just has to be present — a number in the room before the estimate begins.
And once it’s in the room, it’s very difficult to fully escape.
3. Why the Mind Does This
The current leading explanation involves two mechanisms working together.
Selective accessibility.
When you’re given an anchor, your mind begins — often unconsciously — generating reasons why the anchor might be correct. If the anchor is 65, you retrieve information consistent with 65. This selectively loaded information then influences the estimate. You’re not just starting at 65 and adjusting — you’re building a case for 65 without realizing it.
Insufficient adjustment.
Even when people consciously try to move away from an anchor they know is wrong, they tend to stop adjusting too early — at the first plausible value they reach, rather than continuing to the most accurate one. The adjustment is effortful. The mind stops when it has found something defensible, not something optimal.
Both mechanisms operate largely below conscious awareness. You don’t experience yourself being anchored. You experience yourself reasoning carefully toward a considered judgment. The anchor has already done its work before the reasoning begins.
4. Where It Does the Most Damage
Negotiation:
The first number in any negotiation becomes the anchor for everything that follows. Whoever sets the opening offer shapes the range in which the final agreement will land. This is not a minor effect — studies consistently show that opening offers are among the strongest predictors of final outcomes. The party that anchors first holds a structural advantage, regardless of the merits of their position.
Pricing:
Retailers use anchoring deliberately. The original price crossed out above the sale price. The premium option listed first. The suggested donation amounts on a form. Each is an anchor designed to pull spending upward. The “original” price doesn’t need to be real to be effective — it just needs to be visible before the decision is made.
Legal judgments:
In one study, experienced legal professionals were asked to determine a sentence after rolling a pair of dice — rigged to land on either 3 or 9. Those who rolled 9 recommended sentences averaging eight months longer than those who rolled 3. These were trained professionals making consequential judgments. The dice moved them anyway.
Medical diagnosis:
The first diagnosis considered for a patient becomes an anchor that shapes how subsequent information is interpreted. Symptoms that fit the initial hypothesis are weighted more heavily. Symptoms that don’t fit are explained away. This is a significant driver of diagnostic error — not incompetence, but the gravitational pull of the first frame.
Get Brian Kim’s stories in your inbox
Join Medium for free to get updates from this writer.
Performance evaluation:
The first piece of information about a candidate — a school name, a previous salary, a first impression — anchors everything that follows. Subsequent information is interpreted through the lens of the initial anchor rather than evaluated independently. The résumé at the top of the pile has a structural advantage over the one at the bottom.
Self-imposed anchors:
You set a goal. The goal becomes an anchor. You hit it and stop — even when continuing would have been better. Or you set a target price for an investment, and that number shapes when you sell, independent of what the actual analysis would suggest. The anchor you set for yourself constrains you as surely as one set by someone else.
5. How to Work Against It
Name the anchor explicitly.
The first defense is awareness. When a number enters a conversation before an estimate or decision, name it: “That number is now an anchor. Let me be deliberate about not letting it set the range.” Naming the mechanism doesn’t eliminate it, but it reduces its pull.
Generate your own estimate before hearing others.
In any situation where you’ll be receiving an anchor — a price, a valuation, an assessment — form your own view first. Commit it to paper if possible. An anchor has less power over an estimate that already exists than over one that hasn’t yet formed.
Consider the opposite.
One of the most effective deanchoring techniques is to deliberately consider why the anchor might be too high — or too low. If the anchor is pulling you toward a high estimate, generate reasons it should be lower. This counteracts the selective accessibility mechanism by loading the opposite case into working memory.
Start from a different place.
Rather than adjusting from the anchor, try building the estimate from first principles — from the base rate, from comparable cases, from the underlying data. Arrive at a number through a different path, then compare. The anchor loses power when it’s not the starting point.
Be deliberate about anchors you set for others.
If you’re leading a negotiation, setting a budget, or framing an evaluation, recognize that the numbers you introduce first will shape what follows. This is a responsibility as much as a tactic. The anchor you set determines the gravitational field everyone else is working in.
6. Where This Fits in the Series
- Occam’s Razor → Prefer simple explanations
- Hanlon’s Razor → Prefer benign interpretations
- Dunning-Kruger → Prefer humility about what you know
- First Principles → Prefer foundational truth over inherited assumption
- Second-Order Thinking → Prefer the full consequence over the convenient one
- Inversion → Prefer the question that runs backward
- The Map Is Not the Territory → Prefer the territory over your model of it
- Survivorship Bias → Prefer the full sample over the visible survivors
- Confirmation Bias → Prefer the evidence that challenges you
- Sunk Cost Fallacy → Prefer the future over the weight of the past
- Availability Heuristic → Prefer the data over what’s loudest in your memory
- Anchoring Effect → Prefer the estimate you built over the number you were handed
7. One Question for This Week
Identify a number that’s currently shaping a decision — a budget, a valuation, a salary, a target, a price.
Then ask:
“Did I arrive at this number — or did it arrive first, and I’ve been reasoning around it ever since?”
The anchor feels like a starting point.
It’s actually a conclusion, written before the thinking began.
