██████╗ ██╗██╗ ██╗███████╗██╗ ██╔══██╗██║╚██╗██╔╝██╔════╝██║ ██████╔╝██║ ╚███╔╝ █████╗ ██║ ██╔═══╝ ██║ ██╔██╗ ██╔══╝ ██║ ██║ ██║██╔╝ ██╗███████╗███████╗ ╚═╝ ╚═╝╚═╝ ╚═╝╚══════╝╚══════╝ ███████╗████████╗██████╗ ███████╗███████╗████████╗ ██╔════╝╚══██╔══╝██╔══██╗██╔════╝██╔════╝╚══██╔══╝ ███████╗ ██║ ██████╔╝█████╗ █████╗ ██║ ╚════██║ ██║ ██╔══██╗██╔══╝ ██╔══╝ ██║ ███████║ ██║ ██║ ██║███████╗███████╗ ██║ ╚══════╝ ╚═╝ ╚═╝ ╚═╝╚══════╝╚══════╝ ╚═╝
# Plain-text mirror. Same words as the human page, minus everything a browser needed.
Emotional marketing works. Most of the case studies used to prove it do not.
Five statistics travel with the five most-quoted emotional marketing campaigns, one bolted to each. I checked all five against the publications they are credited to. Four of the five do not exist. The Guardian never reported that Old Spice sales rose 125 per cent. Adweek never reported that “Shot on iPhone” lifted iPhone sales by 30 per cent. Marketing Week never reported Coca-Cola’s 2 per cent figure; the Wall Street Journal did, about a different country and a different year. And Nike’s financial reports, cited by name every time the claim is made, do not attribute a single dollar of revenue to the words “Just Do It”.
I run Pixel Street, a design and branding studio in Salt Lake, Kolkata. We design for Coca-Cola, ITC and Marico, so one of the brands carrying an invented number is a client of ours. These figures get repeated in decks and case studies for years at a time without anyone opening the source. That is roughly how carefully campaign statistics get read, including by the people repeating them.
Emotional marketing works. Most of the case studies used to prove it do not.
Every one of these numbers fails the same way. A real measurement of something narrow gets rounded up, quietly re-scoped onto something much bigger, and then handed the name of a publication that never ran it. Dove’s figure started life as a sales claim and now travels as a social-media engagement claim, about a campaign launched in 2004, before there was much social media to engage with. Coca-Cola’s figure is real, but it belongs to the American launch in 2014 rather than to the Australian first year in 2011, and Coca-Cola itself refused to confirm it.
[IMG: A real, narrow measurement gets rounded up, re-scoped onto something bigger, then handed a masthead that never ran it. Old Spice body wash sales were documented up 11% over twelve months, 55% over three months and 107% over the previous month alongside a coupon; what circulates is a 125% rise credited to The Guardian, which never published it.]
So what follows is the campaigns and the mechanics, with every figure traced to whatever is actually documented and the invented arithmetic named where it sits. If you are building a campaign on one of these numbers, you are building it on nothing.
Emotional marketing is a strategy that aims for a feeling first and a product second, on the basis that the feeling is what gets remembered and repeated. Rather than describing what a thing does, these campaigns try to produce happiness, sadness, anger or recognition, and let the brand attach itself to that.
The honest version of the rationale is narrower than the one usually given. People do not decide purely on specifications, and a brand that produces a feeling is easier to recall than one that produces a feature list. That is defensible. What is not defensible is the popular escalation of it into a percentage, most often some version of “90 to 95 per cent of purchasing decisions are subconscious”. That figure is a misquote, and I explain what it actually says at the bottom of this page.
Emotional marketing is not confined to one channel. It shows up in a film, a product description, a package and a shop window. What it needs is consistency, because the feeling is built by repetition and destroyed by contradiction.
Two kinds of emotion get used. Primary ones such as joy, sadness and fear are close to universal. Secondary ones such as nostalgia are built by culture, which is why a campaign that works in Melbourne can land flat in Kolkata. Five approaches account for most of what you will see.
Nostalgia borrows a happy memory and puts the brand next to it. It works on people who lived through the period and, if the styling is good enough, on people who did not. Vintage packaging and throwback edits are the cheap version. The expensive version is a brand that actually was there, which is an advantage you cannot buy.
Some brands align with a cause and borrow the anger or urgency attached to it. This is the highest-variance play on the list. It works when the brand has a real position and something to lose by taking it. It fails, loudly, when the stance is decorative, and the failure is usually worse than never having spoken.
Fear of missing out is the fear that other people are having a better time than you, and social media industrialised it. Brands convert it into urgency: limited runs, countdowns, drops. It is effective and it is corrosive, because a customer who buys from anxiety does not feel warm about the brand afterwards.
Aspirational marketing shows someone reaching a goal with the product in frame. It runs on excitement and possibility rather than on fear. The risk is the gap between the promise and the product: the wider it is, the shorter the campaign’s useful life.
This one sells the customer a version of themselves rather than a version of the product: more confident, more capable, better looking. It is the oldest move in advertising and the one most likely to age badly, because the ideal being projected is always the ideal of a particular decade.
A four-way chart circulates under this heading: happiness gets shares, sadness gets clicks, fear gets loyalty, anger goes viral. It carries no source, and one row of it is the opposite of what the research found.
The paper worth knowing is Jonah Berger and Katherine Milkman’s What Makes Online Content Viral, published in the Journal of Marketing Research in 2012, which analysed every article the New York Times published over a three-month period. Their finding is that sharing is driven by physiological arousal rather than by whether a feeling is pleasant. High-arousal emotions, positive or negative, get content shared: awe, anger, anxiety. Low-arousal emotions do not. Positive content outperforms negative content overall, but valence alone predicts very little.
Berger and Milkman name sadness specifically as a deactivating, low-arousal emotion, and content that evokes it is less viral, not more. The chart has it driving clicks. That is not a rounding error or a difference of interpretation; it is the finding inverted.
The practical read is not “make people sad” or “make people angry”. It is that a calm, pleasant, agreeable piece of work is the one least likely to move, and that is a harder brief than the chart made it look.
[IMG: High-arousal feelings get content shared whether they are negative, like anger and anxiety, or positive, like awe. Low-arousal feelings do not: sadness is deactivating and makes content less viral, not more, and calm agreeable work is the least likely thing to move.]
Four things carry most of the weight, and one popular fifth does not.
An emotional campaign needs a specific person on the other end, and most briefs do not have one. What actually worries them, what they are proud of, what they would not admit to wanting. This is unglamorous research and it is the part that most often gets skipped in favour of choosing a feeling first and finding an audience for it afterwards.
Colour does real work, and almost everything written about which colour produces which feeling is invented. The percentages you will find, most of them some version of “colour improves brand recognition by up to 80 per cent”, trace back through a chain of marketing blogs to a law-review footnote that cites a colour consultancy and adds the words “citation omitted”. Every version of it I have chased ends at that footnote.
What colour reliably does is make a brand recognisable at a distance and hard to confuse with a competitor. We design for Coca-Cola. That red is doing an enormous amount of work, and none of the work is making anyone feel joy on contact; it is making the can identifiable in a fraction of a second from across a shop. Own a colour, use it consistently, and stop asking it to carry a mood.
A story lets an audience arrive at a feeling instead of being told to have one, which is the whole difference. The ones that travel are built on shared experience: a difficulty overcome, a goal reached, an ordinary moment noticed. This is also the cheapest of the five for a small brand, because it needs a good idea rather than a budget.
Belonging is the most durable of these effects, and the hardest to fake. It is also the one with the clearest cautionary tale. TOMS built a decade of brand affection on One for One, where each pair bought sent a pair to a child who needed one. That model has been retired. TOMS now describes One for One in the past tense on its own impact page and directs its giving through grants to partners including Save the Children, Didi Hirsch Mental Health Services and GirlForward. The lesson is not that TOMS was wrong to change. It is that when the mechanism a movement was built on goes away, the emotional equity built on it does not automatically survive.
Red Bull is the reference case: the product is a can of caffeine, and the marketing is people doing things that look impossible. It works because the brand actually pays for the events rather than borrowing footage of them. Aspiration is expensive to fake and cheap to spot.
Five campaigns, with what is actually documented about each and what was invented about it.
Dove launched “Real Beauty” in 2004 with women who did not look like models, against an industry that used nothing else. The campaign argued with its own category, which is the rarest and most defensible form of this work.
It said something specific and contestable at a time when nothing else in beauty advertising did, and it kept saying it for twenty years, which matters more than the launch did.
The figure that travels with Real Beauty is a 700 per cent increase in social media engagement. It does not exist. The documented commercial outcome, reported by Ad Age, is that Dove soap sales went from $2 billion to $4 billion in the three years after launch. A doubling, not a sevenfold rise, and not a social-media measurement at all for a campaign that started before social platforms had scale.
The campaign also has a criticism worth carrying, since a post about emotional marketing that only tells the flattering half is doing the thing it is warning about. The retoucher Pascal Dangin later described digitally manipulating the “real” images, and Unilever was marketing skin-lightening products and Axe advertising over the same period.
Apple’s World Gallery went up in March 2015: photographs by 77 photographers from 70 cities in 24 countries, on more than 14,000 sites worldwide including over 770 large-format billboards, across 79 cities in 26 countries. Customers made the advertising. It later won the Cannes Lions Grand Prix for Creative Effectiveness.
It is a product demonstration that does not feel like one, and it makes the buyer the author. The proof and the flattery are the same object, which is very hard to do and very cheap to run once it works.
The figure attached to this campaign is a 30 per cent increase in iPhone sales, credited to Adweek. No such Adweek article exists. Apple’s own fiscal 2015 results report company-wide revenue of nearly $234 billion, up 28 per cent, across every product it sells, in the year the larger-screen iPhone 6 generation was on sale. Someone took a whole-company figure, cut it to a round number and pinned it on a poster campaign.
Dan Wieden wrote “Just Do It” in 1988 at Wieden+Kennedy. It has outlived every product it was written for, which is the actual achievement, and it is the clearest example on this list of a line that works because it is about the audience rather than the shoe.
Three words with no product in them, addressed to the person who has not started yet. It fits a marathon runner and someone deciding whether to go for a walk, and it has needed no revision in thirty-eight years.
The claim in circulation is that Nike’s financial reports credit the slogan with driving revenue to $46.7 billion in 2022. Nike’s reports say no such thing. The $46.7 billion is real, from the year ended 31 May 2022, and Nike’s own release attributes the 5 per cent growth to double-digit growth in NIKE Direct, with net income down 5 per cent in the same year. Four years on, in the year ended 31 May 2026, revenue was $46.4 billion, flat on a reported basis and down 2 per cent currency-neutral. Same slogan, slightly less revenue. A tagline that cannot explain the flat years cannot be credited with the good ones.
Ogilvy Sydney put the 150 most common Australian first names on the bottles in 2011 and replaced the most valuable logo in the world with them. Coca-Cola’s own account is that it sold more than 250 million named bottles and cans that summer in a country of just under 23 million people. It has since run in more than 70 countries.
It converted a mass-produced object into a personal one at no real cost, and it made buying the product an act of giving it to someone else. Coca-Cola is a client of ours, and this is still the campaign of theirs I hold up as the standard, for an unromantic reason: it is a packaging idea rather than an advertising idea, which is why it kept working after the ads stopped running.
Two figures circulate for Share a Coke, both credited to Marketing Week: a 2 per cent sales rise in the campaign’s first year and a 6 per cent rise in social media interactions. Marketing Week published neither. The 2 per cent figure comes from the Wall Street Journal in September 2014 and describes United States soft-drink sales over that summer, which was the American launch and the campaign’s fourth year, not its first. Coca-Cola declined to release official figures at the time, so even that number rests on unnamed sources. The social-media figure has no traceable origin at all.
Old Spice’s “The Man Your Man Could Smell Like” broke in February 2010 from Wieden+Kennedy, and addressed the women buying the body wash rather than the men using it. The follow-up, in which the same character answered people on video in near real time, is the part that actually changed what agencies thought was possible.
It solved a targeting problem with a joke. It also correctly identified who was standing in the aisle, which is a research finding disguised as a punchline.
The number that follows this campaign around is a 125 per cent sales increase, credited to The Guardian. The Guardian never published it, and 125 per cent matches nothing in the primary reporting. The real figures are Nielsen’s, reported by Adweek in July 2010: Old Spice Body Wash sales up 11 per cent over twelve months, 55 per cent over three months and 107 per cent over the previous month. And the 107 per cent was disputed at the time, because Old Spice was running a two-for-one coupon promotion concurrently. The honest sentence is that sales rose sharply during a campaign that coincided with a discount, and that nobody separated the two.
Neither of these is attached to a campaign, but both turn up wherever emotional marketing is argued for, and I have gone looking for the study behind each of them. There is no study behind either.
“90 to 95 per cent of purchasing decisions are subconscious.” This is a misquote of Gerald Zaltman, who wrote it about “thought, emotion and learning”, not about purchasing decisions, and whose own source calls it a rule of thumb among cognitive scientists rather than a study. The number is real; the sentence it was taken from is about something else entirely.
“You have seven seconds to make a first impression.” This one has a single named source, the marketing professor Michael Solomon, and he has said on the record that it is an urban myth and that no such study exists. It is quoted as though it came from a laboratory, and the man it is credited to has publicly disowned it.
Probably, and none of these five campaigns proves it. Every one of them ran alongside product changes, price changes, distribution changes and in Old Spice’s case a coupon. The measurable claim is narrower: high-arousal emotional content gets shared more, which Berger and Milkman demonstrated. Everything past that is inference.
Because a number with an institution’s name attached reads as evidence, and almost nobody clicks through. Every failed chain I have traced ends the same way: a real, narrow finding gets rounded up, widened, and then given authority by a masthead rather than by a citation.
Go to the named source and find the sentence. If the trail is a blog citing a blog citing an infographic, there is nothing at the end of it. If the figure is a round number, treat that as a warning: real measurements are 11 per cent and 55 per cent, not 700 per cent. My rule is that a statistic I cannot read in the original does not go in a deck.
None of these five campaigns is a template you can copy on a small budget, and Share a Coke is the only one that would survive being done cheaply, because it is a packaging idea rather than a media buy. Start where the customer touches the product. That is the part you own.
These five campaigns are worth studying and the arithmetic attached to them is not. Dove argued with its own industry, Apple made the customer the author, Nike wrote three words about the audience instead of the shoe, Coca-Cola gave up its logo for a summer, and Old Spice worked out who was actually holding the bottle. Not one of those insights needed a percentage, and four of the five percentages usually attached to them turn out to be fiction.
If you want the ground underneath this, how we build a brand strategy before any campaign gets written is the prior step, the branding process we actually run is what it looks like as a project, and what still works in copywriting now that machines can draft covers the execution.
If you would rather we did it, we are a design and branding studio in Kolkata, and we will tell you when your campaign does not need a statistic.
Every figure in this article traces to one of the following. Publisher and publication date are recorded so a claim can be checked against its origin, and re-checked when the origin changes.
![[IMG: Emotional Marketing]](https://pixelstreet.in/blog/media/2024/09/emotional-marketing-1-1.webp){kind=link}