Unilever's '69% Faster' Was a Headline, Not a Finding. Then Its Own CEO Quietly Conceded the Point.

There is a number that did more to sell the idea of corporate purpose than any business school case ever has. In June 2019, on a stage at the Deutsche Bank Global Consumer Conference in Paris, Unilever's CEO Alan Jope told investors that the company's 28 'Sustainable Living Brands' had grown 69% faster than the rest of the business, and delivered three-quarters of its total growth.¹ It was tidy, it was enormous, and it traveled. For years afterward you could find that figure in keynotes, pitch decks, and op-eds, offered as proof that doing good and doing well were the same thing. There was only one problem with it as evidence. It wasn't one.

The official story is that Unilever ran a grand experiment and discovered that purpose-led brands grow faster. The real story is that Unilever published a scoreboard and called it a study - and four years later, its own incoming CEO quietly conceded the difference by abandoning the strategy the number was meant to justify.

A scoreboard with no opposing team

Look closely at how the 69% was made, and the spine of it disappears. The comparison wasn't purpose brands versus identical non-purpose brands; it was a hand-picked set of 28 designated 'Sustainable Living Brands' measured against the leftovers of Unilever's own portfolio. There was no control group - no version of those same brands run without purpose, in the same markets, over the same period, to measure against. So the result can't separate the effect of purpose from the effect of being the brands Unilever chose to call its purpose brands. Strong brands tend to be the ones a company highlights, fund, and dresses in a mission. The label may have followed the growth, not caused it. The number was real; what it proved was not.¹

Terry Smith was not a sustainability skeptic looking for a fight - he was the founder of Fundsmith, which held roughly £888m of Unilever stock, and he wanted the thing to work. His objection was methodological, and it was lethal: the company had confused correlation with cause and effect.³ A year earlier he had put it more bluntly, writing that a company that feels it has to 'define the purpose of Hellmann's mayonnaise has in our view clearly lost the plot.'² The line landed because it exposed the strategy's weakest seam - not purpose for Dove, which sells confidence, but purpose for condiments, applied as a universal rule whether it fit or not.

The tell was in the incentive

Here is the part that turns a soft statistic into a hard one. The 69% figure was not announced by a researcher; it was announced by a CEO, to investors, about a strategy on which his own standing depended. That is not proof of bad faith - it is the textbook condition under which selection effects masquerade as findings. A self-reported metric, with no control and a curated sample, presented by the person who most needs it to be true, is exactly the shape a number takes when it is built to persuade rather than to test. None of which makes purpose worthless. It makes the evidence for it thin where the company claimed it was overwhelming.

The CEO who conceded the case without saying so

If the thesis had been sound, you'd expect it to survive a change of management. It didn't. On 26 October 2023, new CEO Hein Schumacher told investors that Unilever would stop 'force fitting' purpose to its brands, conceding that for some it could be an 'unwelcome distraction' or simply 'irrelevant.'⁴ In the same Growth Action Plan, filed with the SEC, he was unsparing about the record the strategy left behind: 'our performance in recent years has not matched our potential. The quality of our growth, productivity and returns have all under-delivered.'⁵ He did not abandon purpose - he was explicit that for Dove and Lifebuoy, done credibly, it remains effective.⁴ What he abandoned was the mandate that every brand carry one regardless of relevance. He replaced it with something duller and more defensible: roughly 30 'Power Brands' representing over 70% of turnover, funded by what they sell.⁵

The filings show the retreat in slow motion. Comparing consecutive SEC documents, the 2022 language about 'brand teams driving social purpose into their brand's proposition' softens, by the 2023 annual report, to 'social purpose where appropriate' - and the company flags 'consumer preference' risk as having increased.⁷ Two words, 'where appropriate,' carry the whole reversal. And then the verdict arrived in the only currency investors trust. In full-year 2024, the Power Brands grew 5.3% with volume up 3.8%, underlying operating margin rose 170 basis points to 18.4%, and brand and marketing investment hit 15.5% of turnover - its highest level in over a decade.⁶ The company spent more on brands, not less. It just stopped requiring each one to mean something it didn't.

But didn't Hellmann's grow 20% with purpose, not despite it?

The honest counter is the strongest one, and it points the other way. In the very window after Smith mocked Hellmann's 'make taste, not waste' campaign, the brand grew 20% in the first half of 2022 - and Unilever attributed part of that to the waste-reduction advertising itself.⁸ So purpose was not commercially counterproductive there; it coincided with a strong run. That's a fair point, and it's the right correction to the lazy story that purpose always sinks sales. But notice what it concedes: it's a single brand, in a single window, with the same attribution problem in reverse - the company crediting purpose for growth it can't cleanly isolate. The whole argument has always run on anecdotes dressed as evidence. The case against the strategy was never 'purpose fails.' It was 'Unilever claimed proof it never had, and made every brand pay the relevance tax to maintain the fiction.' Hellmann's 20% doesn't refute that. It just shows the molecule sometimes works - which was never in dispute.

It would be too neat to say Unilever proved purpose doesn't pay. It didn't - and the better reading is that it never proved the opposite either. For more than a decade, a company ran a strategy on the strength of a statistic that was built to be repeated rather than tested, and the moment a new CEO had nothing to defend, he let the mandate go and the numbers got better. The lesson isn't that brands shouldn't stand for things. Dove still should. The lesson is older and harder: a figure you can't argue against isn't a finding, it's an advertisement - and Unilever spent four years discovering that the most expensive thing a company can believe is a number that was never trying to be true.