A pricing page can win by making the price harder to notice, not easier to justify.
That's the finding from experiment Exp-053, and it's the kind of result that makes a growth team uncomfortable before it makes them money. Most pricing CRO training says the same thing: be transparent, reduce friction around the number, build trust by putting price up front. Exp-053 says that's a heuristic, not a law — and shows what happens when a team runs the exception instead of assuming the rule.
TL;DR
- Exp-053 examined reducing the visual salience of pricing on a plan-selection page — not redesigning or improving how the price was shown, but making it less prominent.
- The experiment was timed deliberately around a seasonal window when market rates were running at a multi-month high — the number itself was working against the business, not for it.
- The lower-salience variant won, with a lift range of 10% to 20%, one of the stronger margins across this experiment set, and the approach was scaled for that seasonal window.
- The result challenges a near-universal assumption in pricing CRO: that transparency and prominence are always the right call. They're conditional on whether the number helps or hurts the pitch at that specific moment.
- The judgment worth hiring for here isn't "experiment with the price display" — it's recognizing when market context makes the standard playbook the wrong one to run.
| Factor | Conventional pricing-page instinct | What Exp-053 examined |
|---|---|---|
| Market condition | Rarely treated as a design input | Seasonal rate peak, deliberately targeted |
| Design lever | Improve how the price is presented | Reduce how much attention it draws |
| Underlying belief | Transparency always builds trust | Transparency helps only when the number itself helps |
| Result | — | Winner, 10%–20% lift (Exp-053) |
The decision behind the experiment
This ran on a plan-selection page at a large energy retailer, in the months leading into a seasonal peak in market rates. Anyone who has sold a variable-cost product through a seasonal cycle knows the pattern: rates climb for reasons entirely outside the company's control — weather, wholesale markets, regulatory pass-through — and the number a prospective customer sees on the page gets worse at exactly the moment conversion needs to hold up.
The standard playbook in that moment is to explain the number: add context, comparison points, reassurance copy, maybe rate-lock messaging. All of that treats the price as something to be justified. The alternative hypothesis behind Exp-053 was that justification wasn't the right frame at all — that when the number itself is a genuine headwind, the more effective move is to reduce how much visual weight it carries on the page, not argue for it harder.
That's a different kind of decision than most pricing-page experiments settle. It isn't "which price format converts better." It's "should this page even be built around the price this month, given what's happening in the market." That's a strategy question before it's a design question, and it doesn't get asked if an experimentation program treats every pricing experiment as interchangeable with every other one — same page, same lever, whatever week happens to be open on the roadmap.
Why salience moves behavior
Framing effects are among the most replicated findings in behavioral economics. Kahneman and Tversky's prospect theory (_Econometrica_, 1979) established that people evaluate outcomes relative to a reference point and weight losses more heavily than equivalent gains — and that changing how a number is presented, without changing the number itself, changes the decision people make. Their follow-up work on the framing of decisions (_Science_, 1981) showed the same underlying choice can flip entirely depending on whether it's framed as a gain or a loss.
Price framing is a direct descendant of that finding, and it has a more specific cousin in salience theory: the work by Bordalo, Gennaioli, and Shleifer (_Quarterly Journal of Economics_, 2012) formalizes how attention gets pulled toward attributes that visually stand out, and how that attention share distorts the weight a decision-maker gives them relative to their actual importance. Applied to a plan-selection page: a price rendered with high visual weight gets weighted more heavily in the decision than its true relevance to the customer's outcome would justify. When that number is elevated for reasons a customer might find reasonable in the abstract (seasonal market conditions) but unpleasant in the moment (the bill just went up), high salience doesn't build trust — it anchors the entire decision to the worst part of the offer.
This is the mechanism Exp-053 leaned on: not hiding information, but declining to amplify the one detail most likely to trigger a loss-framed reaction at a moment when that reaction was avoidable. Reducing salience doesn't deceive anyone — the price stays fully disclosed, accurate, and exactly where a customer expects to find it. It simply stops competing for attention with the parts of the page that could actually earn the decision: plan terms, service reliability, the reasons to choose this provider over whatever's open in the next comparison-shopping tab.
The methodology call: timing as part of the design
The easiest version of this experiment to run is context-free: reduce price salience on the plan-selection page, whatever week is next on the calendar, see what happens. That's the default most teams reach for, and it would have produced a noisier, harder-to-interpret result — because in a normal-rate month, the price isn't a threat. It might even be a selling point if the retailer is genuinely well-priced relative to the market. Running the experiment in a random week bakes a coin-flip on market conditions into a hypothesis that has nothing to do with chance.
The senior call was to hold the experiment until the seasonal window when market rates were already elevated across the category — a period lasting several weeks — and run it specifically then. That reframes the question from "does hiding the price work" (unstable, because the answer depends entirely on context that wasn't controlled for) to "does reducing price salience work when the price is actively working against you" (stable, because it's tied to a condition that recurs on a predictable cycle every year). That distinction is the difference between a one-off result and a piece of institutional knowledge a business can act on every time the seasonal pattern repeats.
It also means the result doesn't generalize the way pricing-page wins often get misread as generalizing. This isn't "always reduce price salience" — it's "reduce price salience when the number is a genuine headwind, and leave it alone when it isn't." An experimentation program that can't hold that conditional will misapply this result the first time rates come back down.
The result
The lower-salience variant won — one of the stronger margins across this experiment set, with a lift range of 10% to 20% (Exp-053). The business scaled the approach for that seasonal window: reduced pricing salience while the market made the number a genuine headwind, reverted to the standard presentation once rates normalized.
That reversion matters as much as the win did. The result was never "hide the price permanently" — it was "reduce its visual weight for as long as the seasonal math says the number is working against the business, and not a day longer." Treating this as a fixed template instead of a conditional one would misread what the evidence actually showed.
The assumption this breaks
Pricing CRO carries a near-default assumption in most playbooks: transparency wins, prominence builds trust, and any instinct to de-emphasize price reads as manipulative or, at best, a short-term trick that erodes trust over time. That assumption is usually right, which is exactly why it's dangerous to apply without checking — it gets used as a rule instead of a heuristic, and rules don't ask whether the specific number, at the specific moment, is helping the pitch or hurting it.
Exp-053 is evidence against the universal version of that rule. Prominence isn't a virtue on its own — it's a virtue when the thing being made prominent helps the case, and a liability when it doesn't. A price that reflects real value at a fair moment deserves the spotlight. A price that reflects a seasonal spike outside the company's control deserves to sit quietly in its expected place on the page while the rest of the offer does the persuading. Treating "always be transparent" as a rule instead of a conditional is the kind of category error that costs conversion every time market conditions shift — which is exactly why this kind of context-aware read is worth building into an experimentation program rather than leaving to instinct.
FAQ
Doesn't reducing price salience hurt trust with customers?
No information was hidden or removed — pricing stayed fully disclosed, accurate, and exactly where a shopper would expect to find it. The variant changed how much visual weight the number carried, not whether it was present. That distinction matters: this is a salience experiment, not a disclosure experiment, and the two get confused more often than they should.
Would this pattern hold outside of energy pricing?
The mechanism — salience shaping how heavily a number gets weighted in a decision — is general, and shows up anywhere pricing is visible and comparison-shopped: insurance, subscription tiers, usage-based SaaS. What's specific to this case is the trigger condition: an externally driven, temporary spike in the number itself. Any category with a seasonal, cyclical, or market-driven cost swing is a reasonable candidate for the same logic. A category with stable, self-determined pricing usually isn't a fit.
How do you know this wasn't just a fluke of that particular season?
One experiment in one seasonal window is directionally accurate evidence, not a permanent rule — which is exactly why the takeaway isn't "hide pricing forever," it's "treat pricing salience as a lever conditioned on market context, and confirm the specific condition before assuming it repeats." That's a more defensible claim than either "always show price prominently" or "always hide it," and it's the standard I'd hold any single-market result to before scaling it further.
What's the actual skill being demonstrated here, beyond this one experiment?
Recognizing that the market moment itself is an input to the experiment design, not noise to average out. Most pricing experiments get built and launched without asking whether the current month is even a fair one to run the experiment in. The judgment to hold the experiment for the right window, and to define success conditionally instead of universally, is the harder and more transferable skill — and it's the one that doesn't show up in a results dashboard.
Why not just improve the pricing presentation instead of reducing its salience?
Because the hypothesis here was specifically about whether the number itself was a headwind, not about how well it was presented. Improving presentation assumes the price is a neutral or positive factor that just needs better framing. Exp-053 was built to examine the opposite assumption — that no amount of polish fixes a number that's actively working against the business at that moment — and the evidence backed it.
Bottom line
Pricing transparency is a good default, not a law. Exp-053 shipped a lift of 10% to 20% by treating price salience as a variable to manage against market conditions rather than a principle to maximize on every page, every week — and the discipline to time the experiment around the one seasonal window where the hypothesis actually had a fair hearing is the part of this result most worth hiring for.
If you're running a pricing page, or an entire experimentation program, where the standard playbook keeps producing flat or negative results, the fix is rarely "run more variants." It's usually a market-context read the current program isn't built to make. I design experimentation programs — including the judgment calls about what to run, when, and why — for teams that want evidence-backed pricing decisions instead of best-practice guesses. Work with me to build one for your team.
Evidence sources and free next step
Baymard's research on price and discount presentation offers a public comparison point for salience and clarity. Use the pricing page optimization review and case-study evaluation guide to separate a contextual exception from a rule. Then try GrowthLayer free to record the market condition, primary metric, and sunset rule.