Strategy ⚡ Low

Anchoring Bias

The first number users see becomes the reference point for every number that follows.

Pricing design Negotiation Perceived value Discount framing Salary negotiation Numerical judgment

01 — TL;DR

Two sentences.

Anchoring Bias — first documented by Amos Tversky and Daniel Kahneman in their 1974 paper Judgment under Uncertainty: Heuristics and Biases — is the cognitive phenomenon in which the first numerical value encountered in a judgment context disproportionately influences all subsequent estimates and evaluations, causing people to adjust from the initial anchor insufficiently rather than reasoning from first principles. For product designers, this bias is simultaneously a design constraint and a strategic tool: whenever numbers appear in an interface — prices, discounts, feature counts, estimated savings, plan comparisons — the order and context of their presentation shapes user perception independently of the numbers' objective values.

Irrelevant anchors still work

Tversky and Kahneman's famous wheel-of-fortune experiment: participants who saw a random 10 estimated African UN membership at 25%; those who saw a random 65 estimated 45% — even when told the number was clearly arbitrary. The cognitive system does not evaluate whether a prior number is logically relevant before using it as an anchor. A large, impressive number shown anywhere before the price — total users, total value delivered, time saved — can anchor the subsequent price evaluation.

Trigger phrase

When you are presenting a price, a discount, a value claim, or any numerical comparison — and the order and framing of the numbers you show will determine whether users perceive the value as high or low, independent of what the number actually is.

02 — When to Use

Apply this when…

→ A pricing page presents multiple plan options — the highest-priced plan presented first anchors users to a higher reference point, making lower plans feel like good value

→ A discount is being presented — showing the original price first anchors to the higher value, making the discount appear more significant

→ A product's value proposition involves large numbers — users saved, time reduced, revenue generated — where presenting an impressive number before the price creates a favorable price-to-value anchor

→ A free trial or freemium conversion is being designed — users who have experienced the product's value before seeing the price are anchored to a higher perceived value

→ A salary or pricing negotiation interface is being designed — the party who names the first number gains the anchoring advantage, a design decision with direct financial consequences

→ Estimated savings, projected returns, or cost comparisons are presented alongside a product price — the larger comparative number anchors the evaluation

When NOT to apply it

Skip anchoring when it would require false or inflated reference numbers — artificially high "original prices" that were never genuine are both ethically problematic and legally regulated in the UK, EU, Australia, and many US states. Skip it when users are expert buyers who anchor to market-rate benchmarks rather than to presented numbers — sophisticated B2B and procurement professionals override anchors with their own references. And skip it in trust-critical contexts — financial advisory, healthcare cost estimators, legal fees — where anchor manipulation damages the professional relationship the product depends on.

03 — How It Works

The mechanism

Numerical judgment under uncertainty is not performed from first principles — it is performed by adjusting from the most available reference point. When asked a question like "Is this price reasonable?", the cognitive system searches for a reference point and adjusts from it. The first number encountered in the context serves as that anchor — the most available value in working memory when adjustment begins. The adjustment is invariably insufficient: research consistently shows people stop adjusting once they reach what feels plausible, and "feels plausible" is determined relative to the anchor, not to any objective standard. In Tversky and Kahneman's experiments, mean estimates differed by 20–40% depending on the anchor alone — a difference with enormous commercial consequences when the number is a price.

Decide which number should be the first one users see

Every interface where numbers appear has a first number. That number is the anchor whether you designed it to be or not. The first design decision is which number you want anchoring everything that follows — the highest tier, an impressive user count, a cost-of-alternative figure, or whatever number best serves perceived value for your pricing decision.

Anchor prices high by presenting the most expensive plan first

Presenting plans ascending from cheapest to most expensive anchors users to the lowest number — making the middle tier feel expensive. Presenting descending from most expensive to cheapest anchors users to the highest number — making the middle tier feel affordable. A/B tests across SaaS consistently show higher average revenue per user with descending price order because the anchor does the work before deliberate evaluation begins.

Use comparison anchors — the cost of the alternative

A B2B SaaS anchored to "the cost of one bad hire" ($50,000) creates a powerful reference for a $500/month subscription because the user relates it to a genuine cost in their domain. The anchor must be in the user's perceived reference class — close enough to the actual decision that it feels relevant, not a stretch the cognitive system dismisses as unrelated.

Anchor value before price, not price before value

Default instinct is price visibility — show the price prominently and early so users "know what they're paying for." Anchoring research suggests the opposite: establish a high value anchor before the price. A product tour demonstrating impressive outcomes before pricing, a landing page that leads with "saves you X hours per month" before the subscription cost, a case study showing ROI before the fee — all create anchors that make the subsequent price evaluation more favorable.

A/B test anchor variations and measure willingness-to-pay

Different anchor conditions should produce different selection rates and different average revenue per user. Set up A/B tests that vary the first-number on pricing pages — plan order, value anchors, comparison anchors — and measure the effect on plan distribution, conversion, and ARPU. Anchoring effects are typically large enough to detect in modest traffic volumes, making this one of the higher-leverage experiments in pricing-page optimization.

The anchor must be real

Anchoring is one of the most commercially exploited biases in digital commerce — and one of the most regulated. Inflated "original prices" manufactured to create discount perceptions exist on a spectrum from legitimate design (genuine previous prices) to consumer-protection violations (fake references). UK Consumer Protection from Unfair Trading Regulations, the EU Omnibus Directive, and most US state laws require reference prices to reflect genuine pricing for a meaningful period. Beyond legality, modern shoppers use price-history tools (Camelcamelcamel, price-tracking extensions) that make manufactured discounts increasingly detectable — and the trust damage when users discover manipulation typically outweighs the short-term lift. Anchor must be real.

04 — Real Example

Salesforce's pricing page and the high-anchor tier strategy

Salesforce's pricing page for Sales Cloud is a textbook application of price anchoring. The page presents four tiers — Starter Suite, Pro Suite, Enterprise, and Unlimited — but visual emphasis and "recommended plan" highlighting consistently draw attention first to the Enterprise and Unlimited tiers, priced at $165 and $330 per user per month respectively. By the time a prospective buyer looks at the Pro Suite at $100 per user per month, they are evaluating it against an anchor of $165–$330 — making $100 feel affordable by comparison, regardless of what the buyer might have considered reasonable before arriving at the page.

The page also employs comparison anchoring: each tier's feature list details capabilities whose combined market value — if purchased separately — would substantially exceed the tier price. This creates a second anchor pull alongside the price-tier anchoring. The combined effect is a pricing page that makes every tier feel like reasonable value relative to two simultaneous anchors: the higher tiers above it in the sequence, and the market value of the features within it.

B2B SaaS · Pricing page · Salesforce

High-tier anchoring makes mid-tier appear affordable before deliberate analysis begins

Salesforce's consistently high average contract values and low price sensitivity among customers are partly attributable to this systematic anchoring architecture — the page is designed to make the numbers feel right before any deliberate price analysis begins. Pro Suite at $100/user/month evaluated against an Enterprise anchor of $165/user/month feels like good value. The same $100 presented after a Starter anchor of $25/user/month would feel expensive. Same plans. Same prices. Different anchor. Different outcome.

$165–$330 anchors · mid-tier feels affordable

05 — In the Wild

Test yourself & see real examples

No examples yet — be the first.

Spotted a pricing page, discount presentation, or value proposition where the first number you saw changed how you evaluated every number that followed — or one where the price felt arbitrary because no anchor had been established? Submit a screenshot and annotate what you see. Every approved example gets attributed to you.

✓ Reviewed before publishing ✓ Your name on every example you submit ✓ Violation or fix — both welcome

Seen Anchoring Bias applied deliberately or missed in a real product? Help grow the evidence base.

06 — Common Mistakes

Where teams go wrong

✗ Using fake original prices as anchors. Crossed-out prices manufactured to create discount perceptions — rather than reflecting real previous prices — are ethically problematic and legally regulated. UK CPRs, the EU Omnibus Directive, and many US state laws require reference prices to reflect genuine pricing over a meaningful period. Price-history tools make manufactured discounts increasingly detectable, and the trust damage when users catch it dwarfs any short-term lift.

✗ Presenting the cheapest plan first and anchoring to low value. Ascending price order (often chosen for "user friendliness") anchors users to the lowest number. A $9/month plan seen first makes $49/month and $99/month feel expensive. A/B tests across SaaS consistently show higher average revenue per user with descending price order — teams defaulting to ascending may be optimizing for the appearance of simplicity while leaving significant revenue on the table.

✗ Neglecting to anchor value before showing price. The instinct is price visibility — show it prominently and early. Anchoring suggests the opposite: establish a high value anchor before the price appears. A product tour demonstrating impressive outcomes before the pricing page, a landing page leading with "saves you X hours per month" before the subscription cost, a case study showing ROI before the fee — all create value anchors that make the subsequent price evaluation more favorable.

✗ Anchoring to the wrong reference class. A B2B SaaS anchored to "the cost of one bad hire" ($50,000) creates a powerful anchor because users relate it to a real cost in their domain. The same tool anchored to "the market cap of the Fortune 500" is a large number that does not connect — the anchor is irrelevant enough that users discard it. The anchor must be in the user's perceived reference class — close enough to the actual decision to feel like relevant context.

✗ Assuming sophisticated users are immune. Tversky and Kahneman demonstrated that even participants told the anchor was random still showed anchoring effects. Awareness of the bias does not eliminate it. Real estate professionals in Northcraft and Neale's research showed anchoring effects despite having complete property information. Users who consider themselves rational still anchor — design accordingly.

07 — Variations & Related Principles

Connected ideas

Anchoring Bias is one of the most commercially important cognitive biases in product design. Its closest relationships are with the other cognitive biases and design principles that govern numerical perception and value judgment.

The most important pairing is Anchoring Bias with Framing. Anchoring determines the reference point against which a number is evaluated; framing determines how the comparison between the anchor and the evaluated number is presented — as a saving, a cost, a value, or a loss. A price anchored to a high original value is made more compelling by loss framing ("don't pay $299 when you can pay $99") than by gain framing ("save $200"), because loss aversion amplifies the anchor's impact. Together, anchoring and framing provide the complete toolkit for numerical presentation design.

08 — 10-Min Exercise

Run it right now

⏱ 10 minutes · Solo · No prep

The First Number Audit

Navigate to your product's pricing page, checkout page, or the primary page where users encounter your price for the first time. Write down every number you see, in the order you see them (left-to-right, top-to-bottom).

1. The first number on your list is your current anchor. Ask: is this the number you want anchoring all subsequent numerical evaluation? Is it a large number that makes your price look small, or a small number that makes your price look large? Is it a competitor price, an original price, a savings claim, or your lowest tier?

2. Identify any large, impressive number related to your product that is not currently on this page but could legitimately appear before the price: total users, average customer savings, hours saved per month, cost of the alternative, ROI achieved by existing customers. Each is a potential high-value anchor.

3. Write a version of the page's opening numerical statement that places the most impressive legitimate number before the first price the user sees. Compare this to the current anchor.

4. If the reordering changes which number appears first and the new first number is larger and more favorable to price perception, you have identified an anchoring opportunity worth testing in the next pricing-page iteration. And verify the anchor is real — not a number manufactured for the effect.

10 minutes