Why Buyers Rarely Pick the Best
In 1992, Itamar Simonson and Amos Tversky ran an experiment that should have ended a thousand arguments about pricing. They offered one group of shoppers a choice between two Minolta cameras: a basic model and a mid-range one. The split was even — 50/50. Then they showed a second group three cameras: the same two, plus a premium model priced well above both. The mid-range camera, now flanked on both sides, took 57% of the vote. Against the cheapest option, its relative share climbed from 50% to 72%. The product had not changed. Its price had not changed. Only its position in the set had changed, and that was enough to reorganize what people wanted.
That is what the word “rarely” in the title rests on: in experiments by Simonson (1989) and Simonson & Tversky (1992), adding an extreme option to a choice set shifted buyers toward the option that became the middle one.
This is the compromise effect, and I think it is one of the most reliable findings in choice architecture, and I still see marketers treat it as a party trick. The lesson is not "add a third tier so the middle one sells." The lesson is harder and more useful: buyers often skip both the best option and the cheapest. They tend to choose the one that looks reasonable — and reasonable is a relationship, not a property. You engineer it by deciding what sits next to it.
Extremes come with an argument against them
A buyer facing three options is not really comparing three products. They are comparing three positions: too little, too much, and the one in between. Each extreme carries a built-in objection. The cheapest option whispers that you are cutting a corner you will regret. The most expensive one whispers that you are overpaying for headroom you will never use. The middle option is the only one that arrives without an accusation attached. Often it is chosen less because it is good and more because the other two have been quietly disqualified by their own positions.
That is why extremeness aversion is so durable. It does not require the buyer to understand the products. It does not even require the products to differ much. It runs on the geometry of the set. And it compounds a problem I rarely see brands diagnose, because they evaluate their offer in isolation — is the Pro tier good? is the price fair? — when the buyer is evaluating a layout. The question the customer is actually answering is not "is this good," it is "is this the sensible one here." If your best, highest-margin tier sits at the top edge of the set with nothing above it, you have placed the option you most want them to buy in the exact spot the mind is built to reject.
The decoy effect is the famous one. It is also the fragile one.
Most people who know one effect from a choice-architecture article know the decoy: add an inferior, asymmetrically dominated option and watch demand shift to the target. It is the tactic behind the apocryphal Economist subscription table. It is also, by a decade of evidence, the one that does not hold up. When researchers moved the attraction effect out of tidy numeric tables and into realistic stimuli — actual images, real products, incentive-compatible choices — it shrank or vanished. Frederick, Lee, and Baskin, and separately Yang and Lynn, published that case in 2014, and a 2025 registered report is still litigating exactly which presentation conditions it survives. The decoy works in a spreadsheet. It struggles in a store.
The compromise effect is the robust sibling. It replicates across product categories, across hypothetical and real-money designs, and across the kind of messy conditions that kill the decoy. If you are going to build strategy on a behavioral finding — and InPhluence builds on the ones that hold up — build it on the middle, not the mirage. In my experience, a real middle beats a fake inferior option, and it does not require you to insult your customer's intelligence with an option engineered to be refused.
The middle is not a safe zone. It is a setting you control.
Here is where the old advice breaks, and where the current research earns its place in your strategy. The convenient story is that the center of any set is a magnet: build three tiers, and the middle collects the buyers. That story is incomplete, and Sharif, Webb, and Bhatia's 2025 research shows why.
In November 2025, Marissa Sharif, Elizabeth Webb, and Sudeep Bhatia published six studies in the Journal of Consumer Research, with a combined sample of 9,377 people, showing that extremeness aversion is not a constant. It depends on how the set is composed. When a choice set contains multiple extreme options, people become more willing to choose an extreme — because the presence of several extremes makes the extreme start to look typical of the category rather than reckless. Fill a menu with high-end options and the high end stops reading as excessive; it reads as normal, and the pull toward the middle weakens.
Read that against the Minolta result and the real principle snaps into focus. The middle does not win because the center is inherently safe. The middle wins because, in most sets, the extremes look like outliers. Change what looks typical and you move the whole distribution of demand. The lever was never the option. It was always the composition of the set — what you show, how many, and at what spread. In my experience, brands spend their energy making one tier better. The leverage is in designing the frame that tier lives inside.
This is a positioning problem, not a pricing-page problem
Collapse this to a tactic for your plans table and you will have missed the larger move. Extremeness aversion governs how a market perceives your entire brand, because a category is also a set, and your brand occupies a position within its visible range.
"The best" is a position at the edge. "The cheapest" is the other edge. Both edges carry the same built-in objection they carry on a pricing page: the premium brand is where you overpay, the discount brand is where you compromise on something that will bite you later. The brand that gets installed as the default — the obvious, reasonable, nobody-got-fired-for-choosing-it option — is the one the market reads as the sensible middle of its category. That position is worth more than superiority, because superiority is an argument you have to keep winning and reasonableness is a judgment the buyer makes for themselves and defends on your behalf.
Which means the strategic questions change. Not "how do we prove we are the best," but "what is the perceived range of our category, and where in it do we sit." Not "how do we justify our price," but "what options are we being seen alongside, and do they make us look like the outlier or the obvious choice." If the set around you is all moderate, your premium position reads as indulgent. If the set around you is crowded with extremes — a category full of loud, expensive, maximalist players — the 2025 finding says your high position stops looking extreme and starts looking like the category standard. The competitive frame is not noise around your strategy. It is an input to it.
The same geometry runs through public discourse
Positioning in a market and positioning in a conversation obey the same rule. In a polarized category — of opinion, of policy, of brand posture — I think the loudest extreme tends to capture attention and tends to miss the default. Attention and adoption are different currencies, and the mechanism that separates them is extremeness aversion: audiences discount the edges of a debate for the same reason shoppers discount the edges of a product line. The voice that gets quoted is rarely the voice that gets trusted to be reasonable.
For a brand making a public argument, this is a discipline, not a comfort. It does not mean splitting the difference or saying nothing. It means understanding that the market will locate you relative to the full spread of voices it can see, and that being the most strident option in a set of strident options is a way to be heard and dismissed in the same breath. Reasonableness, positioned well, is in my view one of the most persuasive stances available — provided the set around you makes it legible as reasonable rather than timid.
What this actually asks you to do
Start by defining the set before you refine the option. Before you improve a tier or sharpen a claim, map what the buyer sees next to it — your other tiers, and your competitors' positions. The target is to make the option you want chosen the one without an edge, flanked on both sides so its position argues for it.
Then audit for orphaned extremes. If the plan you most want to sell is the top of your own ladder, or the brand position you most want to own is the loud maximum of your category, you are fighting the geometry. Give the buyer something to its outside — a genuine premium tier, a more extreme competitor you are happy to be compared against — so your target becomes the compromise rather than the ceiling.
And watch the composition of the whole set, because it is not fixed. A category can drift toward extremes over a cycle, and when it does, the 2025 research says the center loses some of its gravity and positions that once read as excessive begin to read as normal. The brand that reads the drift can move with it. The brand that assumes the middle is a permanent safe harbor will keep optimizing an option while a competitor quietly rebuilds the frame around it.
The option you want them to choose rarely belongs at the edge of what they can see. In my experience, buyers rarely reward the best. They reward the reasonable — and reasonable is something you build, by deciding what stands on either side of it.
If you want to see where your brand sits in the set your buyers actually compare, book a 30-minute call.
