Anchoring Bias

What anchoring bias means

Anchoring bias is the cognitive tendency to rely too heavily on the first piece of information encountered when making a decision. That initial information - the anchor - sets a reference point, and all subsequent judgements are made by adjusting away from it, usually insufficiently. Even when the anchor is arbitrary, irrelevant, or deliberately misleading, it exerts a powerful gravitational pull on the final decision.

The concept was identified by psychologists Daniel Kahneman and Amos Tversky in 1974 as part of their broader research into the heuristics people use when making judgements under uncertainty. Their experiments demonstrated that even random numbers - generated by spinning a wheel in front of participants - influenced estimates on completely unrelated questions. People who saw a high number on the wheel gave higher estimates; people who saw a low number gave lower ones. The anchor had nothing to do with the question. It shaped the answer anyway.

This is what makes anchoring so unsettling. It works even when you know it’s happening. Even when the anchor is transparently arbitrary. Even when you’re actively trying to ignore it.

How anchoring bias works in practice

The mechanism is deceptively simple. You encounter a number, a price, a claim, or a starting position. That figure lodges in your mind as a reference point. Everything you evaluate afterwards is measured against it - not against an objective standard, but against the anchor.

Why the first number wins

When someone asks you to estimate the population of a city, your answer will be influenced by whatever number you encountered most recently - even if it was in a completely different context. If you’ve just been thinking about a number in the thousands, your estimate will skew lower. If you’ve been thinking about millions, it will skew higher. The anchor doesn’t need to be related to the question. It just needs to be present.

This happens because the brain doesn’t start from zero when making estimates. It starts from whatever is available and adjusts from there. The problem is that the adjustment is almost always insufficient - we don’t move far enough away from the anchor. Psychologists call this “insufficient adjustment,” and it’s remarkably consistent across cultures, contexts, and levels of expertise. The effect gets stronger under decision fatigue - a tired brain leans even harder on whatever number it was given first, because generating a fresh estimate is genuinely more expensive than nudging the anchor a little.

Anchoring in negotiations and salary discussions

Anchoring is one of the most powerful forces in any negotiation. The party that sets the first number - the opening offer, the initial asking price, the first salary figure mentioned - has a structural advantage, because all subsequent discussion orbits around that anchor.

In salary negotiations, this is why employers often ask candidates for their current or expected salary before making an offer. The candidate’s number becomes the anchor, and the offer will adjust from it. If the anchor is low, the offer will be low - regardless of the role’s market value. This also explains why negotiation experts consistently advise making the first offer rather than waiting: the anchor you set shapes the entire conversation.

Anchoring bias in consumer decisions

Retailers and marketers have built entire pricing strategies around anchoring, and you encounter them daily without necessarily recognising them.

How pricing exploits anchoring bias

The “was £99, now £49” format is pure anchoring. The original price serves as the anchor, making the current price feel like a bargain - even if the product was never worth £99 and £49 is its actual market price. The anchor doesn’t need to be real. It just needs to be visible.

Restaurant menus often place an expensive item at the top of the list - not because they expect many people to order it, but because it anchors the perception of everything below it. After seeing a £45 steak, a £22 pasta dish feels reasonable. Without the steak, the same pasta might feel expensive.

The same principle operates in property markets, car dealerships, and subscription services. The “premium” tier exists partly as an anchor that makes the “standard” tier look like good value. The decoy option - a deliberately unattractive choice designed to make another option look better by comparison - is a close relative of anchoring, working through the same mechanism of relative judgement.

Anchoring bias in the workplace

Professional decision-making is riddled with anchoring effects, from budgeting to project planning to performance evaluation.

How anchoring distorts planning and budgets

When teams estimate how long a project will take or how much it will cost, the first estimate mentioned tends to anchor all subsequent discussion. If someone opens with “I think this is a six-month project,” the conversation will orbit around six months - perhaps adjusting to five or seven, but rarely questioning whether the right answer might be two months or eighteen.

This connects to the planning fallacy, where people systematically underestimate time and cost. If previous projects set the anchor and those projects were themselves underestimated, the anchor perpetuates the error. Each new estimate inherits the optimism of the last.

In performance reviews, anchoring shows up when a manager’s initial impression of an employee - formed early in the review period or even at hiring - anchors their assessment of everything that follows. The halo effect and anchoring often work together here: a strong first impression creates both a positive halo and a high anchor against which all subsequent performance is judged.

Anchoring bias in politics and public opinion

Political communication uses anchoring strategically. By opening with an extreme position, a negotiator or politician can shift the entire frame of discussion so that the eventual compromise lands closer to their actual goal.

How extreme anchors shift the centre ground

This is how the Overton window moves. If a politician proposes an extreme policy, the public debate anchors around that extreme. The “moderate” counter-position - which might previously have been considered bold or unusual - now feels like a reasonable compromise by comparison. The anchor hasn’t made the extreme position more popular, but it has made positions closer to it feel more acceptable.

The framing effect is closely related. Anchoring determines the starting point; framing determines the direction. Together, they shape not just what people think about an issue but the range of positions they consider thinking about.

In polling and surveys, the order and wording of questions can anchor responses. Asking “do you think the government spends too much on welfare?” before asking about specific programmes will anchor responses differently than asking the same questions in reverse order. The anchor shapes what feels like a reasonable answer.

Why anchoring is so difficult to counter

Anchoring persists because it operates beneath conscious awareness and resists correction even when you know about it. Studies have shown that experts are just as susceptible to anchoring as novices in many contexts. Judges give different sentences based on randomly generated numbers. Estate agents give different property valuations based on the listed asking price. Doctors make different diagnostic estimates based on initial information. Expertise doesn’t eliminate the bias - it just changes what you’re anchoring to.

The availability heuristic feeds anchoring: the most readily available number becomes the default reference point, whether or not it’s the most relevant one. Confirmation bias then protects the anchor - once a reference point is established, we selectively notice information that supports it and discount information that suggests it’s wrong.

Loss aversion adds another layer. Once you’ve been anchored to a high value - a high salary, a high price, a high estimate - anything below that anchor feels like a loss, even if the lower figure is objectively reasonable. The anchor doesn’t just set a reference point; it sets an expectation, and falling short of an expectation activates the psychology of loss.

How to counter anchoring bias

The most effective defence is independent research before entering any situation where anchoring might operate. If you’re negotiating a salary, research market rates before you hear any figures. If you’re buying a house, assess comparable sales before you see the asking price. If you’re estimating a project timeline, build your estimate from first principles before hearing anyone else’s.

First principles thinking is the direct antidote to anchoring. Rather than adjusting from someone else’s starting point, you build your own assessment from the ground up. This doesn’t eliminate anchoring entirely - once you’ve heard a number, you can’t fully unhear it - but it significantly reduces its influence.

Independent evaluation also helps. Forming your own judgement before exposure to others’ opinions means you arrive at the discussion with your own anchor rather than adopting someone else’s. In group settings, having people write down their estimates privately before sharing them prevents the first speaker’s number from anchoring the entire room.

Anchoring bias is one of those cognitive shortcuts that works well enough in low-stakes situations - the first price you see at the supermarket is probably in the right ballpark - but becomes genuinely dangerous in high-stakes decisions where the anchor has been placed deliberately. Recognising that the first number you hear is not special, just first, is the beginning of thinking more clearly about everything that follows.