Decoy Pricing
Decoy pricing introduces a strategically inferior option to push buyers toward a target tier. Here's how the asymmetric-dominance effect works, with the math and realistic share-shift benchmarks.
Decoy Pricing
Adding a deliberately worse option so a target option looks like the obvious pick by comparison.
Decoy pricing is a tactic where you introduce a third option engineered to be inferior to your target tier on the dimensions buyers care about, while being similar in price. The decoy isn't there to sell — it's there to anchor judgement. Buyers compare the two options closest in price, see one clearly dominates, and choose it without doing the harder cross-comparison against the cheapest tier.
The behavioural mechanism is the asymmetric-dominance effect: people struggle to compare items on multiple attributes, but they easily pick a 'dominating' option when one is offered. Subscription-box and SaaS pricing pages lean on this constantly — the awkward middle tier you'd never recommend is often the decoy, not a mistake.
The canonical example is the cinema popcorn menu: small €4, large €7, medium €6.50. The medium is the decoy — almost as expensive as the large but visibly smaller. Most buyers upgrade to large, and revenue per transaction jumps. Remove the decoy and large-popcorn share collapses.
In e-commerce the same pattern shows up in subscription cadence (monthly vs quarterly vs annual), bundle size (1 / 2 / 3 SKUs), and shipping options. A well-designed decoy doesn't lie or trick anyone — every option is genuinely available — it just makes the trade-off easier to resolve in favour of the tier you actually want to sell.
Lift_target = Share_target_with_decoy - Share_target_without_decoy
Lift_target
Decoy lift
Percentage-point change in the target tier's share of choices when the decoy is added.
Share_target_with_decoy
Target share, decoy present
Proportion of buyers choosing the target tier when the decoy is included in the menu.
Share_target_without_decoy
Target share, decoy absent
Baseline proportion choosing the target tier on a two-option menu.
An apparel subscription tests two pricing layouts: a control with Basic (€19/mo) and Premium (€39/mo), and a variant adding a Standard decoy at €35/mo with fewer items than Premium.
Premium share (control, 2 options): 32%
Premium share (variant, decoy present): 51%
→ Lift_target = +19 percentage points
The decoy made Premium look like obvious value: nearly the same price as Standard, materially more product. Revenue per visitor rose ~37% even though the decoy itself sold to almost no one.
Two conditions have to hold for the lift to materialise. First, the decoy must be clearly dominated by the target on at least one salient attribute — same price, less value, or same value, higher price. Second, the price gap between decoy and target must be small enough that the comparison feels lopsided. A €35 decoy next to a €39 target works; a €35 decoy next to a €69 target doesn't.
Typical share shifts when a decoy is added to a two-option menu
| Vertical / offer | Target share without decoy | Target share with decoy | Share point lift |
|---|---|---|---|
| Apparel subscription (monthly vs annual) | 28% | 47% | +19 pp |
| Beauty bundle (2-SKU vs 4-SKU) | 35% | 52% | +17 pp |
| Electronics warranty (1yr vs 3yr) | 22% | 41% | +19 pp |
| SaaS tiers (Starter vs Pro) | 31% | 44% | +13 pp |
| Shipping (standard vs express) | 18% | 27% | +9 pp |
Lifts cluster in the 10-20 percentage-point range when the decoy is well-constructed. Lower lifts usually mean the decoy is too obviously useless (buyers ignore it entirely and revert to two-option logic) or the target was already winning. This is one of the easier wins to validate with an A/B test — the effect size is large enough that you don't need huge traffic to reach significance.
Frequently asked questions about decoy pricing
Only if the decoy is fake or the dominated tier is hidden. When every option is genuinely available and accurately described, you're just structuring a menu — buyers are still free to pick any tier. The decoy reduces choice friction rather than removing choice.
Anchoring uses a single high reference price to recalibrate perceived value across a menu. Decoy pricing is about asymmetric dominance — a specific third option that's worse than the target but similar in price, designed to make the target the obvious comparison winner. Both fall under pricing psychology and often appear together.
Directly adjacent to the target tier, with the decoy on the left or middle so the target reads as the upgrade. Most SaaS pricing pages put Basic → Decoy → Target → Enterprise, with the target visually highlighted ('Most popular').
No — and often it shouldn't. Healthy decoy share is 2-8% of buyers. If the decoy is selling 20%+, it's not dominated enough; tighten the value gap. If it's selling 0%, it may be too obviously useless and the lift will weaken.
Yes. Bundle size (1 / 2 / 3 units), warranty length, and shipping speed all behave the same way. Any choice with two or more attributes the buyer trades off is a candidate.
Run a two-arm split: control with your current two tiers, variant with the decoy inserted. Primary metric is revenue per visitor, not conversion rate — adding a tier can flatten conversion while raising AOV. Watch target-tier share and decoy share separately so you can confirm the mechanism is working.
Two failure modes. A decoy that's too close in value steals from the target instead of pushing buyers up. A decoy that's too obviously bad insults the buyer and creates a 'they're trying to trick me' reaction — especially in higher-consideration purchases. Test, don't deploy blind.
Less reliably. On a narrow screen, buyers often see one tier at a time and lose the side-by-side comparison the effect depends on. Mobile-optimised pricing pages need either a sticky comparison toggle or a condensed feature matrix to preserve the asymmetric-dominance contrast.
Three or four named tiers is the practical ceiling. Five or more options trigger choice overload — buyers abandon the comparison entirely and default to the cheapest or leave the page. If you need to show many configurations, hide them behind a 'compare all plans' link.
AI is useful for generating hypotheses — analysing your current tier-share data to suggest where a decoy might shift behaviour — but the answer is always 'test it'. The effect size depends on your specific buyers and price points, and only an A/B test gives you the real lift.
Test ideas before you ship them
Run unlimited A/B tests, attach hypotheses to outcomes, and build a searchable archive of what works — and what doesn't.