Charm pricing
Also known as: Odd-even pricing, Just-below pricing, Psychological pricing
Prices ending in 9 feel meaningfully cheaper than the round number above.
What it means
Charm pricing is the practice of setting a price just below a round number — most often ending in 9, like $4.99 instead of $5.00 — so it feels meaningfully cheaper than it is. It works through two routes that usually reinforce each other. The first is perceptual: shoppers read a price left to right and encode the leading digit ($4-something) before the rest, so the figure anchors low. The second is associative: after decades of exposure, a 9-ending itself signals a discount or a keen price. The same cue can also whisper 'low quality,' which is why prestige sellers often avoid it. It is one of the most pervasive tactics in retail pricing, but its effect on actual purchases is small and conditional — it reliably changes how cheap a price feels far more than how much people buy.
Two mechanisms that can pull apart
The nine-ending works through two routes that usually reinforce each other but sometimes conflict. The first is perceptual: shoppers read a price left to right and encode the leading digits before the rest, so $2.99 is filed near 'two-something' rather than rounded up to three. Thomas and Morwitz showed this only happens when the leftmost digit changes — $2.99 versus $3.00 shifts perception, but $3.59 versus $3.60 does not. The second route is associative: after decades of exposure, a '9' ending itself signals a discount or a keen price. These can diverge, because the same cue that whispers 'cheap' can also whisper 'low quality,' which is why the tactic is not universally safe.
What the field evidence shows
Anderson and Simester ran three mail-order field experiments, printing the identical item in different catalog versions sent to matched customers. A nine-ending raised demand every time, and in one test a dress at $39 outsold the same dress at $34 — a higher price drawing more sales. The lift was largest for items customers had not seen before and shrank when a 'Sale' cue was already present, implying the ending substitutes for information the shopper otherwise lacks. Strulov-Shlain later estimated the bias structurally from retail scanner data on thousands of products: consumers reacted to a one-cent rise above a 99-ending as if it were more than a twenty-cent rise. He also found many chains under-use nine-endings relative to what that response would justify, leaving money on the table.
How large the effect really is
Averaged across studies the effect is real but modest, and it lands more on belief than on behavior. A 2024 preregistered meta-analysis by Troll and colleagues pooled sixty-nine studies and hundreds of effect sizes. Just-below prices improved a product's price image by a small margin and made prices much more likely to be underestimated, a moderate-to-large effect. Yet the average lift in actual purchase decisions was small, and there was essentially no effect on perceived quality. The practical reading is that nine-endings reliably shift how cheap something feels but translate into extra sales only under the right conditions: inexpensive, unfamiliar, frequently bought items where price is the deciding factor and the shopper is not deliberating carefully. Outside those conditions the boost is easy to overstate.
Where it backfires
Because a nine-ending signals a bargain, it can undercut the very products that trade on prestige. Luxury brands, fine dining, and professional services often move to round numbers precisely to avoid the discount association; a $60 wine and a $59.99 wine send different messages about the bottle. Round or unusually precise prices carry their own weight: in negotiations and high-value sales, a specific figure like $362,500 can anchor more credibly than a suspiciously tidy $360,000. The tactic is also not free of scrutiny. Unit-pricing rules, rounding at the till, and fuel-duty conventions exist partly because regulators recognize that a fraction of a cent nobody can actually pay is doing psychological work rather than reflecting a real cost.
Examples
A sweater priced at $49.99 sells better than at $50, and sometimes better than at $45.
Fuel is priced at 149.9 a litre. The tenth of a penny that nobody can actually pay exists so the sign reads one-four-nine rather than one-five-zero.
An estate agent lists at £299,950 rather than £300,000. The £50 costs the seller nothing and keeps the house inside every buyer's 'up to 300' search filter.
A mobile game prices its coin packs at $0.99, $4.99, and $49.99. Each tier sits one visual digit below the round number above it, softening the felt cost of each purchase.
A luxury watchmaker prices a new model at a round $12,000 rather than $11,999, since a nine-ending would read as a markdown and clash with the exclusivity the brand is selling.
First described in Long retail practice; Anderson & Simester (2003).
Key references
- Troll, E. S., Loschelder, D. D., Frankenbach, J., & Friese, M. (2024). A meta-analysis on the effects of just-below versus round prices. Journal of Consumer Psychology, 34(2), 299-325. doi.org/10.1002/jcpy.1353
- Strulov-Shlain, A. (2023). More than a penny's worth: Left-digit bias and firm pricing. The Review of Economic Studies, 90(5), 2612-2645. doi.org/10.1093/restud/rdac082
- Thomas, M., & Morwitz, V. (2005). Penny wise and pound foolish: The left-digit effect in price cognition. Journal of Consumer Research, 32(1), 54-64. doi.org/10.1086/429600
- Anderson, E. T., & Simester, D. I. (2003). Effects of $9 price endings on retail sales: Evidence from field experiments. Quantitative Marketing and Economics, 1(1), 93-110. doi.org/10.1023/A:1023581927405