What is the decoy effect?
A worse third option can change which of two others people choose. The research supports it on tables of numbers and finds it much weaker when products are shown or tasted.
Answer to the question
What is the decoy effect?
The decoy effect is the finding that adding a clearly worse third option can shift which of two other options people choose, though not reliably in every setting. The worse option loses to one neighbour on every listed feature. Huber, Payne and Puto reported it in 1982. It is best supported where options appear as numbers and is much weaker when products are shown or tasted.
Key takeaways
- Huber, Payne and Puto (1982) showed that adding an asymmetrically dominated option can raise the chance that people choose the option that beats it, which many standard choice models rule out.
- Ariely's Economist menu reports 84 choosing print and web with the decoy present and 32 without it. The figures are unkeyed to any paper and no sample size is quoted here.
- Simonson (1989) treated the effect as choice based on reasons: an option that beats another on every feature is easy to justify.
- Yang and Lynn (2014) reported 11 reliable effects in 91 attempts across 23 product classes, with chance-level effects when options had verbal descriptions or pictures.
- Bateman, Munro and Poe (2008) found a dominated option shifting choices and later valuations in choice experiments for goods with no market price.
The decoy effect is the finding that a worse third option can change which of the other two people pick. It is best supported on tables of numbers.
How Does a Decoy Move Choice?
The paper, by Huber, Payne and Puto, appeared in the Journal of Consumer Research in 1982. Its authors called the added option asymmetrically dominated: one existing option beats it on every dimension, the other does not. The added option itself rarely gains. Adding it raised the chance that people chose the option that beat it. That breaks a rule most choice models rely on, which says a new option can only take share from the options already present. The authors noted that the result suggests product line strategies that might not otherwise be intuitively plausible. The effect is weaker when products are shown or tasted, as a later section shows.
An invented example shows the structure. Take two subscriptions, one cheap and limited, one dearer and generous. A third that costs almost as much as the dearer one and offers noticeably less is the decoy. Almost nobody takes it. Its job is to make the dearer plan look like the sensible choice, and the decoy effect is the measured shift towards that plan. A menu holds a decoy only if some option loses to a neighbour on every listed feature while the other neighbour does not beat it.
The best-known demonstration opens Dan Ariely's Predictably Irrational (2008). An advertisement for The Economist offered three subscriptions: web only at $59, print only at $125, and print and web together at $125. The print-only option is the decoy, since the combination costs the same and includes more. Ariely allows that the print-only line may have been a typographical error. He ran the menu in two versions. The experiment is the book's own and is not keyed to a published paper, and no sample size is quoted here because it could not be checked.
| Menu shown | Web only | Print only | Print and web |
|---|---|---|---|
| All three options | 16 | none implied | 84 |
| Print only removed | 68 | not offered | 32 |
With the decoy present, 84 of the group chose the combination. With it removed, the combination fell to 32 and web only rose from 16 to 68. Nobody was reported choosing the print-only option, so its presence moved the whole split. The effect itself is not Ariely's: the chapter's bibliography lists Huber, Payne and Puto, and Simonson, among its readings, while presenting the menu as his experiment.
Why Does a Worse Option Work?
The standard explanation is about reasons. Itamar Simonson argued in 1989 that when people cannot fix the exact value of their options, they favour the option backed by the best reasons. An option that beats the decoy on every feature arrives with a ready reason. The abstract of his paper reports that the attraction effect and the related compromise effect were both stronger when people expected to justify their decisions to others, and that choosing the dominating or the compromise option went with more difficult decisions. The compromise effect, which [The Compromise Effect on the Wine List](https://thewhyweproject.com/articles/the-compromise-effect-on-the-wine-list) takes up, is the finding that an option gains share when it sits in the middle of a set.
Haste does not remove the decoy effect. When Jonathan C. Pettibone varied the time allowed for a decision across four levels in 2012, both the decoy effect and the compromise effect were larger as time pressure increased.
Does It Survive Outside Numbers?
Sybil Yang and Michael Lynn reported 91 attempts to produce the effect, across 23 product classes and 73 decoyed choice sets. Only 11 gave reliable effects, fewer than their statistical power led them to expect. When options were distinguished by meaningful verbal descriptions or pictures, the effects shrank substantially and were at chance levels. Their paper appeared in the August 2014 issue of the Journal of Marketing Research, in the same issue as Frederick, Lee and Baskin's "The limits of attraction".
That paper argues the effect may be confined to stylised products in which every attribute is a number, such as a toaster oven with a durability rating of 7.2. It does not typically occur when people taste a drink, or when even one attribute is shown perceptually, as with hotel rooms whose quality is conveyed by a photograph. The authors' explanation is that perceptual representations do not support the comparisons the effect depends on. Huber, Payne and Puto replied in the same issue under a title urging honesty about the attraction effect, defending it, a dispute that [The Option You Were Never Meant to Buy](https://thewhyweproject.com/articles/the-option-you-were-never-meant-to-buy) follows in full.
Support is strongest where attributes are numbers. Ian Bateman and his co-authors Alistair Munro and Gregory L. Poe took the decoy effect into choice experiments and contingent valuation, the methods used to value goods that have no market price. An asymmetrically dominated option significantly affected choices between the other options in the same set, and the shift carried through to the valuations people then gave those options. Those valuations are the numbers such surveys exist to produce.
The demonstrations and the failures may differ in what they make salient. Krol and Krol argued in 2019 that a pictorial setting makes the similarity between decoy and target salient, while an abstract one makes their inferiority salient, and that inferiority is what the decoy effect depends on. In an incentivised eye-tracking study, the transfer of attention towards the undesirable option, and the effect itself, tracked inferiority rather than similarity, which fits an effect that shows up in numeric tables and fades with pictures.
Common questions
What is the decoy effect?
The decoy effect is the finding that adding a clearly worse third option can shift which of two other options people choose, though not reliably in every setting. The worse option loses to one neighbour on every listed feature. Huber, Payne and Puto reported it in 1982. It is best supported where options appear as numbers and is much weaker when products are shown or tasted.
Is the decoy effect the same as the compromise effect?
They are related but separate findings. In the decoy version, the added option is beaten by one neighbour on every feature. In the compromise effect, a middle option gains share because it sits between two extremes. Simonson (1989) studied both under the heading of choice based on reasons, and Pettibone (2012) tested both under time pressure.
Does the decoy effect work on real products?
The evidence is divided. Yang and Lynn (2014) found 11 reliable effects in 91 attempts and chance-level effects when options had verbal descriptions or pictures. Frederick, Lee and Baskin (2014) report it typically failing for tasted goods. Bateman, Munro and Poe (2008) found it in valuation surveys where attributes were numbers.
Sources
Ariely, D. (2008). Predictably Irrational: The Hidden Forces That Shape Our Decisions. HarperCollins. The Economist figures are the book's own, unkeyed to a published paper; the sample size is not quoted because it could not be checked.
Huber, J., Payne, J. W., & Puto, C. (1982). 'Adding asymmetrically dominated alternatives: violations of regularity and the similarity hypothesis.' Journal of Consumer Research, 9(1), 90-98.
Simonson, I. (1989). 'Choice based on reasons: the case of attraction and compromise effects.' Journal of Consumer Research, 16(2), 158-174. doi:10.1086/209205.
Pettibone, J. C. (2012). 'Testing the effect of time pressure on asymmetric dominance and compromise decoys in choice.' Judgment and Decision Making, 7(4), 513-523. Both decoy effects were found to be larger as time pressure increased. doi:10.1017/s1930297500002849.
Yang, S., & Lynn, M. (2014). 'More evidence challenging the robustness and usefulness of the attraction effect.' Journal of Marketing Research, 51(4), 508-513. doi:10.1509/jmr.14.0020.
Frederick, S., Lee, L., & Baskin, E. (2014). 'The limits of attraction.' Journal of Marketing Research, 51(4), 487-507.
Huber, J., Payne, J. W., & Puto, C. P. (2014). "Let's be honest about the attraction effect." Journal of Marketing Research, 51(4), 520-525. doi:10.1509/jmr.14.0208.
Krol, M., & Krol, M. E. (2019). 'Inferiority, not similarity of the decoy to target, is what drives the transfer of attention underlying the attraction effect.' Journal of Neuroscience, Psychology, and Economics, 12(2), 88-104.
Bateman, I. J., Munro, A., & Poe, G. L. (2008). 'Decoy effects in choice experiments and contingent valuation: asymmetric dominance.' Land Economics, 84(1), 115-127.