Making the cheapest of three plan tiers the visual hero of a pricing page is supposed to be the safest move in conversion optimization — one real experiment from my portfolio shows it can quietly commoditize the whole page and pull conversion down instead of up.
TL;DR
- A mid-market energy provider tested making the cheapest of three plan tiers the visual anchor on its plan-selection page, on the standard logic that a clearer, lower price reduces friction and speeds up the decision.
- The variant lost. Lift landed at -10% to -5% against control (Exp-047), and the variant was killed.
- The mechanism wasn't confusion or a broken layout — it was anchoring effect pricing working exactly as designed, just against the business. Elevating the cheapest number turned three differentiated plans into a single price comparison.
- The CRO field treats "highlight the cheapest option" as a safe, obvious anchoring play. This result argues the opposite: anchoring on price can collapse a considered, multi-attribute choice into one number, and the number it collapses to is your lowest one.
- The strategic read: anchoring effect pricing doesn't create conversion by itself — it decides which attribute the buyer weighs most heavily. Point it at price on a differentiated menu and you've told the buyer the differentiation doesn't matter.
What the Anchor Actually Signals
| Anchor placement | What it tells the buyer | When it helps | When it backfires |
|---|---|---|---|
| Cheapest tier | "This is the reference price" | Single-product, price-led categories | Multi-tier menus — collapses the decision into a price contest |
| Middle tier | "This is the intended default" | Three-tier subscription and plan menus | Tiers with weak feature differentiation |
| Premium tier | "This is the ceiling — everything else is a discount off it" | High-consideration or aspirational purchases | Price-sensitive, commodity-perceived categories |
The Decision at Stake
The page in question was a plan-selection card for a mid-market energy provider — three pricing tiers presented side by side, the moment where a shopper who has already decided to switch providers picks which plan to actually buy. That page carries an outsized share of the acquisition funnel's commercial weight: get the tier choice right and the provider locks in a customer at a defensible margin; get it wrong and the provider either loses the sign-up entirely or wins it at a price that doesn't clear the cost to serve.
The team's working assumption was standard practice: on a page with three options, make the lowest price the most visually prominent, because a lower, clearer number reduces the cognitive tax of comparing plans and should convert more browsers into buyers. It's intuitive, cheap to implement, and exactly the kind of change most optimization checklists wave through without a second look. That's precisely why it needed to run as an experiment rather than ship on assumption — a change this "obviously safe" was going onto a page tied directly to revenue per acquisition, not just top-line conversion rate, and the two don't always move together.
Why Anchoring Cuts Both Ways
Anchoring is one of the most replicated findings in behavioral science. Amos Tversky and Daniel Kahneman's original 1974 _Science_ paper on judgment under uncertainty showed that people asked to estimate a quantity — even an arbitrary one — will drift their answer toward whatever number was placed in front of them first, whether it's a spun wheel or an unrelated reference point with no logical connection to the estimate. Pricing pages exploit the same mechanism deliberately: whichever number a buyer processes first becomes the mental yardstick every other number on the page gets measured against.
The industry's standard use of that finding on a pricing page is to anchor high — show the premium tier first, or make it visually dominant, so the mid-tier reads as a discount by comparison. That's a well-established, useful application of anchoring theory. What this experiment tested was the inverse: anchor low, on the theory that a cheaper reference point removes friction rather than reframing value.
The research on multi-option choice sets predicts why that inversion is risky. Itamar Simonson and Tversky's 1992 work on context effects and extremeness aversion (_Journal of Marketing Research_) showed that when people evaluate items in a set, the relationships between the options carry as much decision weight as the options' individual merits — buyers don't evaluate a plan in isolation, they evaluate it against whatever the set has told them to notice. When the cheapest option is made the anchor, the set tells the buyer to notice price, and once price is the anchor, every plan above the cheapest one reads as a markup rather than as more value. The three tiers stop being three products aimed at three different needs and start being one product at three price points — a race the cheapest tier always wins, at the cost of the average order value of everyone who would otherwise have picked a higher tier.
The Methodology Choice
The design was intentionally narrow: a single-variable change to visual hierarchy on the plan card — which tier's price draws the eye first — with everything else on the page (copy, feature lists, tier structure, layout) held constant. That discipline mattered because the hypothesis was specifically about anchoring, not about clarity or design polish in general; a broader redesign would have made it impossible to attribute any movement in conversion to the anchor itself.
The experiment ran for about five weeks, long enough to move past the initial novelty window and cover the natural variation in a household's shopping day, since choosing an energy plan isn't an impulse decision made at the same hour every day. The more consequential methodology choice, though, was measuring the right unit of value: this had to be evaluated on conversion across the full plan mix, not just clicks toward the cheapest card, because a change that pulls volume toward the cheapest tier can look like a win on click-through and still be a loss on blended revenue per visitor. That's the diagnostic most teams stop short of running — it's easy to test whether an anchor moves clicks, harder to test whether it moves the right revenue.
The Result
The variant lost. Lift landed at -10% to -5% against control (Exp-047), consistent enough across the window to call it directionally accurate rather than a fluke on one segment — and the variant was killed. Making the cheapest price the visual anchor measurably reduced the plan-selection page's ability to convert shoppers, and the team reverted to the prior layout rather than iterate on the concept further.
The Assumption This Overturns
This is where the result stops being a single failed variant and becomes a paradigm-shift finding worth naming directly: the CRO field treats "highlight the cheapest option" as one of the safest, most obvious applications of anchoring theory available. It shows up in optimization checklists, in agency playbooks, in the default advice given to any team running a multi-tier pricing page. The assumption is that a lower anchor can only help, because a lower number can only look more attractive.
This experiment is evidence the assumption is backwards for differentiated, multi-tier choice sets. Anchoring on the cheapest price doesn't just make that one tier more attractive — it recalibrates what the entire page is being evaluated on. Once price is the anchor, the differentiation the other two tiers are built on has to compete against a number the buyer has already been told is the reference point. The result isn't a modest lift to the entry tier a business could live with; it's a page-wide conversion loss, because the anchor didn't just move demand between tiers, it changed what buyers were shopping for in the first place. Anchoring effect pricing is not a universally safe lever — it's a lever that reframes the entire choice set, and reframing a differentiated menu toward "cheapest" is only safe when there's nothing more valuable on the page you actually want the buyer to consider.
FAQ
Is anchoring effect pricing always risky on a multi-tier page?
No — anchoring on a premium or middle tier is a well-supported way to make the rest of the menu look like better value by comparison. The risk in this experiment was specific to anchoring on the _cheapest_ tier in a differentiated set, which reframes the whole page around price instead of value.
Does this mean we should never show our cheapest price prominently?
Not universally. It depends on whether the page sells one thing at one price point or a differentiated set of tiers. On a single-product page, a clear low price can reduce friction as intended. On a three-tier plan card, this experiment argues the same move recalibrates the whole comparison around price — a different, and here worse, outcome.
How much should we trust a -10% to -5% result from one experiment?
Enough to kill the variant, which is what happened — the direction was clear-cut for the page and audience it ran on. It should not be trusted to transfer automatically to a different page, category, or tier structure without re-verification; that same discipline is what flagged this as worth experimenting on rather than shipping by default.
What should we test instead if we want to use anchoring on our pricing page?
Test which tier to anchor, not whether to anchor at all. The stronger, better-evidenced pattern is anchoring on the premium or middle tier so the rest of the menu reads as a discount off a higher reference point, rather than anchoring on the floor.
Why does this matter more to a founder or CMO than to a CRO practitioner?
Because the failure mode here isn't a broken button or a confusing layout — it's a strategic assumption about what "reducing friction" means on a pricing page. Any team can implement the anchor; the judgment call is knowing which attribute you want it to make salient, and that call belongs to whoever owns pricing strategy, not just whoever owns the experiment.
Bottom line
Anchoring effect pricing works — that's not in question. What this experiment shows is that it doesn't have one safe direction on a differentiated pricing page: anchor on the wrong tier and you're not removing friction, you're telling the buyer to stop evaluating value and start evaluating price, and the whole page pays for it in blended conversion.
If you're looking at your own pricing page and treating an anchoring change as too obvious to test, that instinct is exactly the pattern this result argues against. I help founders and growth teams design the experimentation programs that catch calls like this one before they ship — get in touch if you want a second set of eyes on what your pricing page is actually testing for.
Evidence sources and free next step
A recent anchoring replication and meta-analysis examines the behavioral mechanism without claiming that the cheapest option must win. Compare the pricing page optimization review and A/B testing examples evidence table before transferring the tactic. Then get started with GrowthLayer free to document the anchor, value metric, and revenue guardrail.