Prospect Theory
Kahneman and Tversky's prospect theory describes how shoppers evaluate gains and losses against a reference point — and why losses sting roughly twice as much as equivalent gains.
Prospect Theory
A behavioral model showing people evaluate outcomes as gains or losses from a reference point, with losses weighted roughly twice as heavily as gains.
Prospect theory, introduced by Daniel Kahneman and Amos Tversky in 1979, is the empirical replacement for expected utility theory. It says people don't judge outcomes on absolute wealth — they judge them as gains or losses relative to a reference point (often the status quo or an anchor price). Losses feel about twice as painful as equivalent gains feel pleasant, and people systematically overweight small probabilities while underweighting moderate-to-large ones.
In ecommerce, it's the foundation under almost every conversion-rate tactic you already use: anchor pricing, free-shipping thresholds, abandonment emails framed around what the shopper is about to lose, and risk-reversal guarantees that neutralise downside.
Prospect theory has three load-bearing ideas: reference dependence, loss aversion, and probability weighting. Reference dependence means a €60 jacket discounted from €100 feels like a €40 gain — but the same jacket at €60 list price feels like nothing. The reference point does the work, not the absolute number.
Loss aversion is the asymmetry between the two. Losing €10 hurts about 2× as much as gaining €10 feels good, which is why a "don't lose your 15% off" exit-intent message converts better than "get 15% off." Probability weighting is the third piece: shoppers treat a 2% chance of a defect almost like a 10% chance, which is why visible review counts and return-policy copy carry so much weight.
v(x) = x^α if x ≥ 0, v(x) = -λ · (-x)^β if x < 0
x
Outcome relative to reference point
Positive = gain, negative = loss, measured against the shopper's anchor (list price, expected delivery time, prior best deal).
α
Gain curvature
Diminishing sensitivity to gains. Empirically ≈ 0.88 — a €100 gain feels less than 10× a €10 gain.
β
Loss curvature
Diminishing sensitivity to losses. Empirically ≈ 0.88 — symmetric with α in the original Kahneman-Tversky fit.
λ
Loss-aversion coefficient
How much more losses sting than gains. Original estimate ≈ 2.25; ecommerce-specific studies range from 1.5 to 3.5 depending on category.
An apparel shopper has a €120 cart. You show them: "Spend €30 more to unlock free shipping (save €9), or pay €9 shipping." Using α = β = 0.88 and λ = 2.25, compare the perceived value of the two frames.
Gain frame: 'save €9': v(9) = 9^0.88 ≈ 6.87
Loss frame: 'pay €9 shipping': v(-9) = -2.25 × 9^0.88 ≈ -15.46
→ The €9 shipping fee is felt as a 15.5-unit loss, while the equivalent €9 saving is felt as a 6.9-unit gain — a 2.25× asymmetry.
This is why free-shipping thresholds outperform discount-on-shipping promotions of the same monetary value. You're not changing the math; you're moving the reference point so the fee disappears from the comparison.
The coefficient λ varies meaningfully by category. Considered, high-ticket purchases (electronics, furniture) show higher loss aversion because the regret cost of a bad decision is larger. Habitual, low-ticket purchases (consumables, beauty refills) show lower λ because the downside is bounded. Knowing your category's coefficient tells you how hard to lean on loss-framed copy versus gain-framed copy.
Estimated loss-aversion coefficient (λ) by ecommerce category
| Category | Typical λ | Reference point shoppers anchor on | Highest-leverage CRO tactic |
|---|---|---|---|
| Consumer electronics | 2.8 – 3.5 | Lowest price seen elsewhere | Price-match guarantee, extended returns |
| Furniture & home | 2.5 – 3.2 | Photographed room expectations | AR preview, free return shipping |
| Apparel (mid-ticket) | 2.0 – 2.6 | Fit and look on similar body | Free returns, sizing reviews, try-before-you-buy |
| Beauty & skincare | 1.7 – 2.2 | Past product that worked | Sample sachets, money-back on first order |
| Consumables / refills | 1.5 – 1.9 | Per-unit price of last order | Subscribe-and-save, lock-in pricing |
| Luxury / one-off gifts | 2.5 – 3.0 | Recipient's expected reaction | Gift wrap, easy exchange, concierge support |
Read the table as a copy-direction guide, not a law. In electronics, where λ runs above 2.8, the conversion gain from removing perceived risk (price-match, 60-day returns) almost always beats the gain from a sharper discount. In consumables, where λ is closer to 1.5, the reverse holds — a clear per-unit saving converts better than a heavy guarantee. Test the direction before committing copy across the catalogue.
Prospect theory FAQ
It's a model of how people actually make decisions under risk, as opposed to how classical economics says they should. The core idea: we don't judge outcomes by their absolute value, we judge them as gains or losses against a reference point — and losses hurt about twice as much as equivalent gains feel good.
Daniel Kahneman and Amos Tversky published the original paper in Econometrica in 1979, and refined it into cumulative prospect theory in 1992. Kahneman won the 2002 Nobel Prize in Economics largely for this work. It sits at the foundation of modern behavioral economics.
Loss aversion is one component of prospect theory, not a synonym for it. Prospect theory is the full model — reference dependence, loss aversion, and probability weighting together. Loss aversion specifically refers to the 2× asymmetry between losses and gains, which is just one piece of the model.
Three plays cover most of it: set explicit reference points (strike-through prices, "was/now" pairs), frame fees as losses to avoid rather than gains to acquire ("don't lose your free shipping" beats "earn free shipping"), and reduce probability-weighted risk with returns policies, reviews, and guarantees.
It's the equation that maps an objective outcome to its psychological value: v(x) = x^α for gains and v(x) = -λ·(-x)^β for losses. The kink at zero — where losses suddenly weigh λ ≈ 2.25× more than gains — is what makes loss aversion mathematically explicit.
People don't treat probabilities linearly. Small probabilities (1-5%) get overweighted — which is why people buy lottery tickets and worry about rare shipping failures. Moderate-to-high probabilities get underweighted, which is why "90% of customers reorder" lands less convincingly than it should.
Yes, but with calibration. The core asymmetry replicates robustly across cultures and product categories. The specific coefficients (λ ≈ 2.25, α ≈ 0.88) are central estimates — real-world values vary by category, stake size, and shopper income. Treat them as starting points, not constants.
Prospect theory is the founding load-bearing model of behavioral economics. Most named effects you'll encounter in CRO — the endowment effect, status quo bias, the certainty effect, framing — are direct consequences of its three mechanisms. See behavioral economics foundations for the wider map.
Strike-through pricing ("€100 €60") creates a reference point at €100, so €60 reads as a €40 gain. Without the strike-through, €60 is just €60. The product, the price, and the margin are identical — only the reference point changed, and conversion typically rises 10-25%.
Yes. Overusing loss-framed copy (countdown timers, "you're about to lose…" everywhere) trains shoppers to discount your urgency signals. And artificially inflated reference prices (illegitimate strike-throughs) get detected and eroded trust costs more than the framing gained. Use the model honestly.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.