Article: Persuasion Principles for Designers
Reviewed: November 2026 · Method version 2 · How these reviews work
Each factual claim in the article was checked against the source it cites, then challenged by reviewers whose job was to find the strongest evidence against it: replication attempts, meta-analyses, later papers that cite the original work, and research from other fields that study the same mechanism. A separate reviewer weighed both sides, and a script re-checked every quoted source. Where a claim was overstated or misattributed, the article was corrected. The research below supports the article as it now reads.
Cialdini’s own summary names seven shortcuts that guide behavior and asks that they be used ethically.
Cialdini’s firm’s summary page names seven shortcuts and asks that they be used ethically, as the article says.
Six come from the original Influence. Cialdini added unity in Pre-Suasion (2016), and the 2021 new and expanded edition of Influence brings it in as the seventh principle.
Unity first appeared in Cialdini’s Pre-Suasion (2016) and was added to the 2021 edition of Influence. The HarperCollins page was behind a bot challenge, so the edition details were confirmed through Wikipedia’s entries on the book and the author.
Social proof. People look at what others like them do, especially when they are unsure what to do. The previous article covered its modern forms.
Cialdini’s own wording of social proof says it applies especially when people are uncertain. His summary page predates unity, so the unity definition here is the article’s own. Field experiments show gifts raise giving in charity mailings, but a wage gift’s effect on effort faded within hours.
In a restaurant field experiment, diners who received a small piece of chocolate with the check tipped more.
Later field studies (Hilkenmeier and Hoffmann 2021; Malcman et al. 2026) again find that gifts near the end of the meal raise tips.
Tips varied with how much candy was given and the manner in which it was offered.
Hilkenmeier and Hoffmann (2021) found that offering the same gift together with the bill doubled its effect on tips.
In a 1975 study, cookies in scarce supply were rated as more desirable than the same cookies in abundance.
Meta-analyses of commodity theory (Lynn 1991) and of marketing scarcity tactics (Barton et al. 2022) find that scarcity generally raises perceived value, with the type of scarcity that works best depending on the product.
Cookies that became scarce because of high demand were rated higher than cookies that became scarce by accident.
A 2023 meta-analysis in Psychology & Marketing confirms that demand-driven scarcity is more persuasive than restricted supply.
That second detail is the one to carry around. The cookies were identical. Scarcity alone raised the rating, and it rose further when the jar suggested other people wanted them.
Worchel and colleagues found three effects: scarcity alone raised ratings, a change from abundant to scarce raised them more, and demand-driven scarcity beat accidental scarcity. Other people’s demand is one driver among several.
The 2012 Team Cialdini video (11:50) is an animated tour of the original six principles, narrated by Cialdini and Steve Martin; unity is not covered.
Video details confirmed from the YouTube page.
Kaptein and Eckles (2012) found large individual variation: for many participants a strategy’s estimated effect was negative even when its average effect was significantly positive.
A 2025 review of 80 studies (Palm and Tims) also finds that persuasion strategies mismatched to the person often backfire.
A second study repeated the result over three sessions a week apart, and personality measures such as need for cognition explained only a small part of the differences.
Perla et al. (2025) found that tailoring messages to personality has negligible effects on behavior, consistent with personality explaining little of the variation.
This changes how to read a winning test. A scarcity banner that lifts conversion overall may be pushing some people away, and the overall average won’t show it.
Kaptein and Eckles studied repeated influence messages in online experiments, not scarcity banners, so the application to conversion tests is an inference. A 2023 meta-analysis also found that limited-time and limited-quantity scarcity did not raise purchase likelihood on average.
It is also an argument for fewer cues per screen. When Kaptein later tested combinations of influence strategies with Duplinsky, stacking them did not raise compliance, and in a live ad campaign a single strategy beat the combination. Each extra principle is also one more chance to hit someone it repels.
Kaptein and Duplinsky (2013) tested combining influence strategies in three studies, including an online ad campaign, and found that combinations did not raise compliance and that a single strategy beat the combination. They attribute this to interactions between strategies rather than to individual backfires. In other settings, stacking more techniques raised effects: aggressive dark patterns nearly quadrupled sign-ups while mild ones doubled them (Luguri and Strahilevitz 2021), and internet health programs with more behavior change techniques had larger effects (Webb et al. 2010). The advice holds for persuasion cues on one page, not as a general rule.
The Persuasion Knowledge Model holds that people build knowledge of persuaders’ goals and tactics, which earlier persuasion research had ignored, and use it to cope; it shapes attitudes toward the product and the marketer.
Eisend and Tarrahi’s 2021 meta-analysis of 148 papers supports the model’s core prediction, with wide variation across studies.
When the motive was less obvious, people who were the target of the pitch and whose attention was occupied rated the salesperson as more sincere than observers with attention to spare.
In Campbell and Kirmani’s design, the cognitively busy participants were targets and the unbusy ones were observers.
Blatant tactics tend to get spotted and discounted, and they can lower what people think of the company behind them, though they rarely cancel the effect entirely.
Eisend and Tarrahi’s 2021 meta-analysis finds that persuasion knowledge usually leads to less favorable views of marketers but cannot eliminate persuasion effects. Context matters: explicit calls to action hurt in social feeds but helped on shopping platforms.
Subtle ones slip through, especially when people are distracted, which on a phone is often.
Campbell and Kirmani studied face-to-face sales pitches; the link to phone use is the article’s extension.
The restaurant study’s lesson applies: the manner of giving matters, not just the gift.
Strohmetz and colleagues found that both the amount and the manner of giving changed tips. In Falk’s charity experiment, a larger gift raised donations far more than a small one.
A countdown that restarts for every visitor falls apart for anyone who reloads the page, and Kaptein and Eckles’ results suggest some visitors were never going to respond well to that kind of pressure anyway.
Mathur et al. (2019) found deceptive dark patterns, including fake countdown timers, on 183 shopping sites. A 2023 meta-analysis found time-limited scarcity did not reliably raise purchase likelihood.
The research supports one practical test, with a condition attached. Noticing a persuasion attempt usually weakens it. A cue that is true and checkable can survive that, and sometimes gains from it, because looking closer confirms what it claims. A counterfeit cue depends on nobody looking closely, and once someone does, the distrust spreads to the rest of the page.
Isaac and Grayson (2016) found in four experiments that when a persuader uses a credible tactic, noticing the persuasion attempt leads to more favorable judgments, not less. Disclosed default nudges kept their effect in Loewenstein et al. (2015) and Paunov et al. (2018), and a 2025 meta-analysis of 23 publications found that transparency did not reduce nudge effects on behavior. Darke and Ritchie (2007) found that discovered deception breeds distrust that spreads to later advertising. On the other side, meta-analyses show that recognizing a persuasion attempt lowers evaluations on average, and a survey of 406 people found that being aware of manipulative designs did not help them resist. No study compares a true and a counterfeit version of the same cue under high and low scrutiny, so the test holds as a tendency, not a guarantee.
These edits were made on November 12, 2026 after the review.
Claim reviewed: Six principles come from the original Influence; the 2021 new and expanded edition adds unity as the newest principle.
Now reads: Six come from the original Influence. Cialdini added unity in Pre-Suasion (2016), and the 2021 new and expanded edition of Influence brings it in as the seventh principle.
Why: The 2021 edition of Influence does add unity as a seventh principle. But Cialdini first proposed unity in Pre-Suasion in 2016, so crediting it to the 2021 edition hides where it came from. The cited publisher page could not be opened and was checked through other records. Now cites Robert B. Cialdini, Pre-Suasion: A Revolutionary Way to Influence and Persuade, Simon & Schuster (2016).
Claim reviewed: The seven principles work as described: people return favors, act consistently with prior commitments, follow similar others, agree with people they like, follow credible experts, want what is running out, and favor their own group.
Now reads: Social proof. People look at what others like them do, especially when they are unsure what to do. The previous article covered its modern forms.
Why: Six of the seven one-line definitions match Cialdini’s own summary. The social proof line drops his condition that it matters most when people are uncertain, and the cited page has no definition of unity at all. Reciprocity also has known limits: in workplace field experiments a gift raised effort only briefly.
Claim reviewed: The cookies were identical; what moved the rating was the story the jar told about other people’s judgment.
Now reads: That second detail is the one to carry around. The cookies were identical. Scarcity alone raised the rating, and it rose further when the jar suggested other people wanted them.
Why: The demand-versus-accident contrast does show that other people’s judgment matters. But the same study found that plain scarcity, with no story about demand, also raised ratings, and so did a drop from plenty to scarce. Saying the demand story is what moved the rating overstates it.
Claim reviewed: A scarcity banner that lifts conversion overall may be pushing a sizeable group of people away, and the average will never show it.
Now reads: This changes how to read a winning test. A scarcity banner that lifts conversion overall may be pushing some people away, and the overall average won’t show it.
Why: This is a fair inference from Kaptein and Eckles, and the article says ‘may’. But the study used repeated messages in online experiments, not live conversion banners, and it did not report how large the negatively affected group was. ‘The average will never show it’ is too absolute, since breaking results out by segment can reveal some of it.
Claim reviewed: Fewer cues per screen is better because each extra principle is one more chance to hit someone it repels; stacking every principle on one screen means more people a cue backfires on.
Now reads: It is also an argument for fewer cues per screen. When Kaptein later tested combinations of influence strategies with Duplinsky, stacking them did not raise compliance, and in a live ad campaign a single strategy beat the combination. Each extra principle is also one more chance to hit someone it repels.
Why: Pass 4 found no study on stacking cues; pass 5 found one by the lead author of the article’s own source. Kaptein and Duplinsky (2013) ran three studies, including a live online ad campaign, and found that combining influence strategies did not raise compliance and that a single strategy beat the combination. The article’s reason (each extra cue is one more chance to repel someone) is its own, since the authors blame interactions between strategies, and in other settings such as dark patterns and health programs, adding techniques raised effects rather than lowering them. Now cites Kaptein & Duplinsky, Combining multiple influence strategies to increase consumer compliance, International Journal of Internet Marketing and Advertising, 8(1) (2013).
Claim reviewed: Campbell and Kirmani found that when a salesperson’s ulterior motive was easy to see, busy and unbusy people alike used persuasion knowledge; when it was less obvious, people whose attention was occupied rated the salesperson as more sincere.
Now reads: When the motive was less obvious, people who were the target of the pitch and whose attention was occupied rated the salesperson as more sincere than observers with attention to spare.
Why: The abstract supports both halves. One detail is flattened: the distracted people were the targets of the sales pitch and the attentive people were observers, so attention and role were bundled together.
Claim reviewed: Blatant tactics get discounted and cost the brand trust.
Now reads: Blatant tactics tend to get spotted and discounted, and they can lower what people think of the company behind them, though they rarely cancel the effect entirely.
Why: Research on persuasion knowledge does find that obvious tactics are recognized and lower opinions of the persuader. But a meta-analysis shows that recognizing a tactic weakens persuasion without cancelling it, and the cited studies measured a salesperson’s sincerity, not brand trust. One experiment found an explicit call to action helped on shopping platforms. Now cites Eisend & Tarrahi, Persuasion Knowledge in the Marketplace: A Meta-Analysis, Journal of Consumer Psychology (2021).
Claim reviewed: Subtle tactics slip through, especially when people are distracted, which on a phone is most of the time.
Now reads: Subtle ones slip through, especially when people are distracted, which on a phone is often.
Why: The first half follows Campbell and Kirmani: when the motive was subtle, distracted targets were less likely to see it. The claim that phone users are distracted ‘most of the time’ has no source, and we found none.
Claim reviewed: The restaurant study’s lesson is that the manner of giving matters as much as the gift.
Now reads: The restaurant study’s lesson applies: the manner of giving matters, not just the gift.
Why: The study shows that the way candy was offered changed tips, and later studies confirm that timing can double a gift’s effect. But no study compares the manner of giving with the size of the gift, and other field work shows size matters a lot, so ‘as much as the gift’ is not established.
Claim reviewed: A countdown that restarts for every visitor collapses the first time someone reloads, and Kaptein and Eckles’ results suggest some visitors were never going to respond well to it anyway.
Now reads: A countdown that restarts for every visitor falls apart for anyone who reloads the page, and Kaptein and Eckles’ results suggest some visitors were never going to respond well to that kind of pressure anyway.
Why: Fake countdown timers are documented on real shopping sites, and Kaptein and Eckles show that some people respond badly to influence strategies in general. But their study did not test countdown timers, and ‘collapses the first time someone reloads’ assumes people reload and notice. Now cites Mathur et al., Dark Patterns at Scale, Proceedings of the ACM on Human-Computer Interaction (2019).
Claim reviewed: The research points to one test: a true cue keeps working when people notice it, while a counterfeit cue depends on nobody looking closely.
Now reads: The research supports one practical test, with a condition attached. Noticing a persuasion attempt usually weakens it. A cue that is true and checkable can survive that, and sometimes gains from it, because looking closer confirms what it claims. A counterfeit cue depends on nobody looking closely, and once someone does, the distrust spreads to the rest of the page.
Why: Pass 4 found no study behind the ‘noticed cue’ test; pass 5 found several that bear on it. Isaac and Grayson’s four experiments show that when a persuader uses a credible tactic, noticing the persuasion attempt makes people judge it more favorably, not less; openly disclosed default nudges kept their effect in Loewenstein’s and Paunov’s experiments and in a 2025 meta-analysis; and Darke and Ritchie show that discovered deception breeds distrust that spreads to later messages. The caveats are real: on average, recognizing a persuasion attempt still weakens it, and people who recognize manipulative designs say they still cannot resist them, so ‘keeps working’ overstates and the test holds as a tendency rather than a rule. Now cites Mathew S. Isaac & Kent Grayson, Beyond Skepticism: Can Accessing Persuasion Knowledge Bolster Credibility?, Journal of Consumer Research (2016).