Back to Blog

None of what follows is our opinion, and none of it is a single blog post citing another blog post. Every claim below points at a specific paper, with authors, an institution, a journal, and a year, so you can read the original. We have linked each one. Where the evidence is contested, and one of these is genuinely contested, the post says so, because a study is evidence, not scripture.

Myth 1: You have a few seconds to make a first impression

The common version of this is "users decide in the first few seconds." The measured version is far more brutal. In three experiments at Carleton University, Gitte Lindgaard and colleagues showed people web pages for as little as 50 milliseconds and asked them to rate visual appeal. The snap judgments made in 50 milliseconds correlated strongly with judgments the same people made after viewing the page for a full half second.

Fifty milliseconds is a fifth of a blink. It is not enough time to read a headline, parse a value proposition, or notice your offer. It is only enough time to register colour, layout, balance, and whether the thing looks competent. The verdict on your brand is cast before a single word is processed, and the rest of the visit mostly confirms it.

The takeaway for anyone shipping a homepage or an ad: the visual craft is not decoration applied after the "real" work. It is the first thing measured and the hardest to recover from. A cluttered, dated, or off-brand first frame loses the argument before the copy gets to make it.

Source: Lindgaard, G., Fernandes, G., Dudek, C., and Brown, J. (2006). Attention web designers: You have 50 milliseconds to make a good first impression. Behaviour and Information Technology, 25(2), 115 to 126.

Myth 2: More choice sells more

The instinct is that a bigger range gives more people something to buy. Sheena Iyengar of Columbia and Mark Lepper of Stanford tested it in a real store. On one Saturday a tasting booth at an upscale California grocer displayed 6 jams. On another, it displayed 24. A coupon let them track who moved from tasting to actually buying.

The big display pulled a bigger crowd and sold almost nothing. The small display pulled fewer lookers and converted ten times as well. More options got more attention and less action.

Here is the honest part. This exact finding, "choice overload," is the most contested study on this list. A 2010 meta-analysis in the Journal of Consumer Research pooled 63 conditions across 50 experiments and found the average effect was close to zero, with enormous variation between studies. The lesson is not "always cut your range." It is that more choice is not a free win, that it carries a real cost in decision friction, and that whether it helps or hurts depends on the buyer and the context. Anyone who tells you a longer menu, a fatter pricing table, or a wall of options is automatically better is guessing.

Sources: Iyengar, S. S., and Lepper, M. R. (2000). When choice is demotivating. Journal of Personality and Social Psychology, 79(6), 995 to 1006. Contrasted with Scheibehenne, B., Greifeneder, R., and Todd, P. M. (2010). Can there ever be too many options? Journal of Consumer Research, 37(3), 409 to 425.

Myth 3: Sex sells

This one is stated so often it barely gets questioned. In 2015, Robert Lull and Brad Bushman at Ohio State University did the thing that settles arguments: a meta-analysis. They pooled 53 experiments covering 8,489 participants that measured how sexual and violent content affected whether people remembered the brands advertised.

Brand memory was significantly worse when ads ran inside sexual or violent programming, and attitudes toward those brands were less favourable, not more. The proposed mechanism is simple and evolutionary: sex and violence are attention magnets, so they pull cognitive resources toward themselves and away from the brand sharing the screen. The steamy ad gets remembered. The product in it does not.

There is one nuance worth keeping. Memory improved when the ad matched its surroundings, a violent ad inside a violent show, which points at congruence rather than provocation as the real lever. But the headline belief, that sexual content lifts sales for the advertised brand, is not what the pooled evidence shows.

Source: Lull, R. B., and Bushman, B. J. (2015). Do sex and violence sell? A meta-analytic review. Psychological Bulletin, 141(5), 1022 to 1048.

Myth 4: A cheaper price is a neutral lever

Discounting is treated as a pure demand dial: drop the price, sell more, the product is unchanged. Baba Shiv at Stanford, with Ziv Carmon and Dan Ariely, showed the product is not unchanged. They gave people an energy drink marketed to boost mental sharpness, then had them solve word puzzles. One group paid full price. Another bought the identical drink at a discount.

Same drink, same puzzles. The people who paid the discounted price solved meaningfully fewer, and the effect was highly reliable. Price had quietly set an expectation about how well the drink would work, and that expectation shaped the actual result. This is a placebo effect, driven by an expectation the buyers were not aware of.

The marketing lesson is not "never discount." It is that price is a claim about quality, not just a number on a tag. Reflexive discounting can lower what a customer believes they are getting and, on this evidence, what they actually get out of it. A cheaper price is never neutral.

Source: Shiv, B., Carmon, Z., and Ariely, D. (2005). Placebo effects of marketing actions: Consumers may get what they pay for. Journal of Marketing Research, 42(4), 383 to 393.

Myth 5: Free shipping is just a small discount

The reasoning sounds airtight. Free shipping is a discount equal to the shipping cost, so a big enough discount should do the same job. It does not, because zero is not an ordinary price. Kristina Shampanier, Nina Mazar, and Dan Ariely tested it with chocolate. They offered a Lindt truffle and a Hershey's Kiss and watched what people chose as the prices moved.

The price gap between the two chocolates was identical in both rounds, 14 cents. The only change was dropping the Kiss from one cent to nothing. Demand for it jumped from a quarter of people to more than two thirds. A one-cent cut, because it crossed into free, did what a much larger discount could not.

That is why "free shipping" outperforms an equivalent dollar discount, why a free trial beats a cheap one, and why a bundled free item can move more units than the same value taken off the price. Free is not a number on a spectrum. It is a category of its own, and it triggers demand out of proportion to its cost.

Source: Shampanier, K., Mazar, N., and Ariely, D. (2007). Zero as a special price: The true value of free products. Marketing Science, 26(6), 742 to 757.

How to read all of this

Two cautions, because using research badly is its own kind of myth-making. First, a single study is a data point, not a law, which is exactly why the choice-overload section includes the meta-analysis that pushes back on it. Findings from a lab or a single store do not transfer unchanged to your category, your price point, and your audience. Second, effect sizes matter as much as direction. "Statistically significant" is not the same as "large enough to reorganize your business around."

What the five hold in common is more useful than any one result. Buyers are not the rational calculators that folk-marketing assumes. A first impression is visual and near-instant. More options carry a hidden cost. Attention-grabbing creative can starve the brand it is meant to sell. Price quietly rewrites the experience. And free behaves nothing like cheap. Test these against your own numbers before you bet on them. Just do not accept the folklore versions, because on the evidence, the folklore is wrong.

Frequently Asked Questions

How fast do people really form a first impression of a website?

About 50 milliseconds, which is about a fifth of an eye-blink. In three experiments at Carleton University, Lindgaard and colleagues (2006, Behaviour and Information Technology) found that visual-appeal ratings formed in 50 milliseconds correlated strongly with ratings the same viewers gave after a full half second. The impression is visual and near-instant, formed before any copy is read.

Does offering more product choices increase sales?

Not reliably. In Iyengar and Lepper's 2000 field study, a display of 6 jams converted 30% of tasters to buyers while a display of 24 converted only 3%, despite drawing a bigger crowd. But a 2010 meta-analysis in the Journal of Consumer Research pooled 50 experiments and found the average choice-overload effect was near zero with high variance. The honest read is that more choice is not a free win, it carries a decision cost, and whether it helps depends on context.

Does sex actually sell in advertising?

The pooled evidence says no, at least not for brand recall. A 2015 meta-analysis by Lull and Bushman at Ohio State (Psychological Bulletin), covering 53 experiments and 8,489 participants, found that brand memory was significantly impaired when ads ran inside sexual or violent content, and brand attitudes were less favourable. The provocative content pulls attention away from the brand it accompanies.

Is discounting a price a neutral way to sell more?

No. In Shiv, Carmon, and Ariely's 2005 study (Journal of Marketing Research), people who drank a discounted energy drink solved fewer puzzles (about 4.2) than people who drank the identical drink at full price (about 5.8). Price sets an unconscious expectation of quality that shapes the real experience, so discounting can lower both perceived and actual value.

Why does free shipping work better than an equal discount?

Because zero is a special price, not a point on a scale. In Shampanier, Mazar, and Ariely's 2007 study (Marketing Science), holding the price gap between two chocolates constant at 14 cents, dropping one from a penny to free lifted the share of people choosing it from 27% to 69%. Crossing into free triggers demand out of proportion to the actual saving, which an equivalent discount does not.

Related Reading

• The State of AI Search in 2026, our sourced data digest on how AI answer engines are changing search.
• Performance vs brand marketing, why the split is a false choice and how to divide budget by stage.