Psychology of marketing: a field guide to the evidence

The psychology of marketing is the set of published findings about how people notice, remember, judge and choose, applied to products, prices and messages. This field guide lists the main effects, the study behind each, an example and an honest rating of how well the evidence has held up.

What is the psychology of marketing?

The psychology of marketing is the use of research on attention, memory, social influence and decision-making to shape what people buy and how. It works online and offline: a checkout page and a shop shelf rely on the same mental shortcuts.

I have grouped the effects into six families. For each one you get a one-line definition, the original study, a marketing example and an evidence rating:

• Strong: replicated across many labs or confirmed by meta-analysis.
• Moderate: repeated support, with known limits or disputes.
• Weak or contested: failed replications, a near-zero average, or an active argument among researchers.

Why does marketing psychology need a field guide now?

Because a lot of what marketers repeat comes from single studies that did not survive re-testing. In 2015 the Open Science Collaboration repeated 100 published psychology studies and reported that replication effects were about half the size of the originals.

The second reason is law. Tactics that were once just aggressive are now named offences. India's Central Consumer Protection Authority (CCPA) notified its Guidelines for Prevention and Regulation of Dark Patterns on 30 November 2023, listing 13 banned practices.

Attention and memory: what gets noticed and remembered?

People like what they have seen before, and they remember what stands out. These two effects sit under most brand advertising.

Mere exposure. Repeated exposure to something makes people like it more. Original study: Zajonc (1968). Marketing example: steady, low-cost reach on YouTube or outdoor media so a new brand feels familiar by the time someone compares options. Evidence: strong, with a ceiling. A 2017 meta-analysis of 268 curves found an inverted-U pattern: liking rises with exposure, then falls with over-exposure.

The von Restorff (isolation) effect. The one item that differs from the rest is remembered best. Original study: von Restorff (1933). Marketing example: one highlighted plan on a pricing page, or one coloured pack in a row of white ones. Evidence: strong as a memory effect in the lab. I will cover both effects in detail in a coming article on memory and attention, including distinctiveness.

Social influence: how do other people change what we buy?

We look to other people and to experts when we are unsure. Robert Cialdini's book Influence (1984) made these ideas mainstream; a coming article will go through his principles one by one.

Social proof. People copy what others like them do. Original marketing-relevant study: Goldstein, Cialdini and Griskevicius (2008), where hotel signs saying most guests reused towels beat a plain environmental appeal. Marketing example: "4,200 people in Jaipur booked this week" or star ratings beside a product. Evidence: moderate. A German replication found both message types beat no message, but descriptive norm messages were not more effective than the standard one. A separate cluster article will cover social proof and herd behaviour.

Authority. People defer to perceived experts and figures of authority. Original study: Milgram (1963), the obedience experiments, in which 26 of 40 participants followed orders to the highest shock level. Marketing example: a dermatologist explaining a product, or a named expert byline. Evidence: moderate. Milgram studied obedience, not buying, so the jump to marketing is an inference.

Decision biases: why do people judge prices and offers inconsistently?

People judge value by comparison, not in absolute terms. That makes the first number, the wording and the other options on the page matter. A coming article covers anchoring, framing, decoy and loss aversion in more depth.

Anchoring. An initial number pulls later estimates towards it. Original study: Tversky and Kahneman (1974). Marketing example: showing the MRP before the sale price, or listing the premium plan first. Evidence: strong. Anchoring was one of 10 of 13 effects that replicated consistently in the Many Labs project across 36 samples and 6,344 participants.

Framing. The same facts described differently lead to different choices. Original study: Tversky and Kahneman (1981). Marketing example: "95% fat free" rather than "5% fat", or "save ₹500" rather than "avoid paying ₹500 more". Evidence: strong for risky-choice framing, which the original paper showed with both money and lives.

The decoy effect. Adding a clearly worse option makes the option it most resembles look better. Original study: Huber, Payne and Puto (1982). Marketing example: a medium popcorn priced just below the large, so the large looks like the obvious deal. Evidence: moderate, with a clear limit. Frederick, Lee and Baskin (2014) found it may be restricted to stylised options where every attribute is a number, and it often vanished when people tasted or saw the product.

Loss aversion. Losses weigh more than equal gains. Original study: Kahneman and Tversky (1979), prospect theory. Marketing example: "Don't lose your 2,000 reward points" rather than "Use your points". Evidence: strong on average, but disputed in scope. Gal and Rucker (2018) argued that losses do not, on balance, loom larger than gains. A 2024 meta-analysis of 607 estimates then put the average loss aversion coefficient at 1.955, meaning losses count roughly twice as much. The fair reading is that loss aversion is real but depends on the context.

Choice architecture: does the way options are set out change the choice?

Yes, and sometimes more than the options themselves. Defaults are the strongest tool here; choice overload is the most overrated. Both belong in a coming article on nudges, defaults and behavioural economics.

Defaults. People tend to stick with the option chosen for them. Original study: Johnson and Goldstein (2003), "Do Defaults Save Lives?", which compared organ donor consent in opt-in and opt-out countries in Europe and found far higher consent where donation was the default. Marketing example: annual billing pre-selected, or a gift wrap add-on left unticked. Evidence: strong in field data, undecided in pooled studies. A 2022 reanalysis of a large nudge meta-analysis found that after correcting for publication bias, the evidence is undecided for "structure" interventions such as defaults. Note too that a pre-ticked paid add-on can be "basket sneaking" under India's dark patterns guidelines.

Choice overload. Too many options can stop people choosing at all. Original study: Iyengar and Lepper (2000): in a grocery store, 60% of passing shoppers stopped at a 24-jam display against 40% at a 6-jam one, but nearly 30% of those who stopped at the small display bought jam, against 3% at the large one. Marketing example: trimming a pricing page from six plans to three. Evidence: weak as a general rule. Scheibehenne, Greifeneder and Todd pooled 50 experiments with 5,036 people and found a mean effect of virtually zero, with large differences between studies. Simplify because it helps people compare, not because "more choice kills sales".

Reciprocity, scarcity, commitment and ownership: what drives the yes?

These four effects shape the moment of commitment. They are also the ones most easily pushed into manipulation.

Reciprocity. People feel obliged to return a favour. Original study: Regan (1971), where participants given a soft drink by a stranger bought more raffle tickets from him, whether or not they liked him. Marketing example: a free sample in a beauty store, or a genuinely useful free tool before a sales call. Evidence: moderate. The finding is classic and widely cited, but it comes from small lab studies.

Scarcity and urgency. Things seem more valuable when they are scarce. Original study: Worchel, Lee and Adewole (1975), where cookies from a nearly empty jar were rated more desirable, especially when supply had just dropped. Marketing example: a real limited batch, or a festive offer that genuinely ends on Diwali. Evidence: moderate. The legal line matters more than the effect size: fake countdown timers and false stock claims are named in the FTC's 2022 staff report on dark patterns. A coming article covers what works in scarcity and what is now illegal.

Commitment and consistency. Once people agree to something small, they tend to agree to something bigger. Original study: Freedman and Fraser (1966), the "foot-in-the-door" technique. Marketing example: a free quiz or a ₹1 trial before the full subscription. Evidence: moderate.

The endowment effect. People value what they own more than the same thing they do not own. Original study: Kahneman, Knetsch and Thaler (1990), where people given a mug asked far more to sell it than others would pay. Marketing example: free trials, home try-ons, and "build your own" configurators. Evidence: moderate, contested. Later economics work found it shrinks under some conditions, such as with experienced traders.

Emotion and storytelling: do feelings sell better than facts?

Often, yes, especially over the long run.

Narrative transportation. People who get lost in a story come away holding beliefs that fit the story. Original study: Green and Brock (2000). Marketing example: a founder story or a customer's before-and-after told as a short film rather than a list of features. Evidence: moderate. The effect has held up in later work, but not every experiment in the original paper supported the proposed mechanism.

Emotional advertising. Les Binet and Peter Field's 2013 analysis of the IPA Effectiveness Databank, The Long and the Short of It, argued that emotional, brand-building campaigns drive bigger long-term effects than rational ones, and advised balancing "head and heart". Evidence: moderate. The databank holds award entries, so the findings are correlational, not experiments. A coming article covers emotion, storytelling and brand memory; another covers colour, sensory and neuromarketing claims, which have even thinner evidence.

Where did marketing psychology go wrong? The replication crisis

Several famous effects failed when independent labs repeated them under stricter conditions. The pattern is consistent: small samples, flexible analysis and a publication system that rewarded surprising results.

• Ego depletion. The idea that self-control runs out like fuel was tested in a pre-registered study across 23 labs with 2,141 participants. The pooled effect was small, with a confidence interval that included zero.

• Behavioural priming. In the classic study, students primed with words about old age walked more slowly. A 2012 replication with infrared timing found no effect, except when experimenters expected one. Flag and currency priming also failed in the Many Labs project linked above.

• Choice overload. As above, a near-zero average across 50 experiments.

• Nudges in general. Once publication bias was corrected, the 2022 reanalysis found no overall evidence that nudges change behaviour, although some individual nudges may still work.

Anchoring, framing and mere exposure came through well. It means one famous study is a hypothesis, not a rule. A dedicated cluster article will list more claims that did not hold up.

Where is the ethics line? Persuasion versus dark patterns

The line is deception and impaired choice. Persuasion gives people a true reason to act now; a dark pattern tricks them or makes the alternative harder than it should be. Three regulators now draw that line in writing.

• India. The CCPA's 2023 guidelines name 13 dark patterns, including false urgency, basket sneaking, confirm shaming, forced action, subscription traps, drip pricing, disguised advertisements and nagging. They apply to platforms, advertisers and sellers. I have written a full guide to India's dark patterns guidelines.
• European Union. Article 25 of the Digital Services Act (Regulation (EU) 2022/2065) bars online platforms from designing interfaces that deceive or manipulate users or materially impair their free and informed decisions. It names examples such as making cancellation harder than sign-up.
• United States. The FTC's September 2022 staff report, Bringing Dark Patterns to Light, groups dark patterns by tactic and lists baseless countdown timers, false low-stock claims and fake "others are viewing" messages.

Real scarcity is lawful; a countdown that resets is not. A real default is lawful; a pre-ticked paid add-on is not.

If an effect only works when the customer misunderstands what is happening, it is not persuasion, it is a dark pattern.

How do you use marketing psychology responsibly?

Pick effects with strong evidence, test them on your own customers, and check each one against the dark patterns lists before it ships. A simple order for the first 30 days:

1. Week one: audit. List every psychological device already on your site, app and store: badges, timers, default ticks, price anchors, review widgets. Mark each one strong, moderate or weak using this guide.
2. Week two: remove the risky ones. Delete any urgency or scarcity claim you cannot prove, and any pre-ticked paid add-on. These carry legal risk and do not need a test.
3. Week three: test the strong ones. Run one A/B test on a well-supported effect, such as price anchoring on the plans page or a sensible default.
4. Week four: decide and document. Keep what moved real sales or sign-ups, not clicks, and record why each device is there.

Where does marketing psychology fit in India?

It fits everywhere, but the examples must be local. MRP anchoring is built into Indian retail. Social proof works best when it names a city or a community the buyer recognises, which matters in markets beyond the metros. Festive deadlines are real scarcity when they are honest. Offline, store layout, music and scent use the same principles, which a coming article on retail atmospherics will cover.

The risk is also local. The CCPA's rules cover every platform, advertiser and seller offering goods or services in India, so a D2C brand's own website is in scope, not only the large marketplaces. Apps with checkouts, subscriptions or add-ons should audit first.

When should you apply these effects, and when should you not?

Apply them when the product is genuinely good and the effect helps people decide faster or more clearly. Take extra care in health, money and children's products, where the rules are stricter. Do not stack many devices on one page: the FTC report notes that dark patterns combined can cause more harm than any one alone.

Who should own marketing psychology in a team?

The growth or conversion lead should own it, with legal review for anything touching price, urgency, consent or cancellation. Copywriters and designers apply the effects. Analysts design the tests and judge results.

My view

My view is that marketing psychology is most useful as a filter, not a toolkit. I have managed a ₹1.2 Cr monthly performance budget across Google, Meta and affiliate networks, and the tests that mattered were the ones that made an offer clearer, not cleverer. Anchoring and good defaults earn their place. Fake timers do not.

The trade-off is real. Honest, tested persuasion is slower than copying a tactic from a viral thread, and some stripped-out urgency devices may cost short-term conversions. I would accept that cost. The legal risk is growing, and customers who feel tricked leave reviews that undo the gain.

Start with an audit this week

Open your own checkout or enquiry form today and list every psychological device on it. Cross out anything you cannot prove or that appears on India's list of 13 dark patterns. Then pick one strong effect from this guide and test it properly.