ATRIUMsearch → argument graph
Video · 2026-07-27 · 1h 15m · 6 moments

Killer marketing secrets that always work (ft. Ogilvy Vice Chairman)

✦ AI generated

timeline · colored by role

01
Claim

An invention isn't an innovation until it changes behavior.

Rory Sutherland argues that the true measure of innovation is behavioral change, not invention. He illustrates this with the story of Watt and Bolton inventing the 'horsepower' as a marketing unit to help mine owners calculate savings, which was the real innovation that drove adoption of the steam engine.

transcript

Rory Sutherland: An invention isn't an innovation until it changes behavior. I think it's Stuart Butterfield of Slack who says something very similar. No, the only real measure of the effect you're having on innovation is the extent to which you change behavior. And you can invent anything you like, but if you can't get anybody to adopt it, it's an invention, but it's not an innovation. And what was selling steam engines to mine owners? And the purpose of the steam engine was to replace the horses they used to walk round and round in circles draining the mines so that miners could go in and effectively mine coal slate coal whatever without drowning. What he realized is that you could you as an engineer you could talk around you know the calorific capacity of the boiler or the length of the piston stroke or whatever and these people didn't want to know. So Watson and Bolton amongst themselves said, 'Well, what do these people really want to know before they're prepared to buy a steam engine?' They said, 'They want to know how many horses they no longer need to feed if they buy a steam engine. How many horses can I get rid of if I buy the steam engine?' So what went out and invented a unit we still use today, which is called the horsepower. And the reason it's not named after a famous scientist like the M or the Newton or the Koulom or the Watt for that matter or Celsius or anything of that kind is because it's a marketing unit. It was invented for marketing purposes because he could then go and say if you buy 25 horsepower steam engine you can actually get rid of 75 horses cuz I think the horses worked in shift. So it does the work of 25 horses but it does it 24 hours a day. So you can now get rid of of 75 horses. And these people would go scribble, scribble, scribble, scribble, scribble. Cost of horses, cost of feeding horses, cost of looking after horses. And you know, on the back of an envelope, they could then go, we'll have two of those.

gives example · 1supports · 2

02
Mechanism

Reducing anxiety is often cheaper and more effective than solving the physical problem that causes it.

Sutherland argues that businesses pour trillions into solving physical problems (like battery range) when psychological reframing would be far cheaper. He uses his own irrational panic at 16% battery in his Lotus (58 miles range) versus calm at 56% in his wife's Mini (56 miles) to show range anxiety is psychological, not physical.

transcript

Rory Sutherland: I make the same point about electric cars, which is you have this thing called range anxiety, and we're spending billions and billions of dollars a year trying to produce batteries with a higher energy density. It's a really good thing. I'm not I'm not saying, 'Hey, engineers, you're wasting your time.' But isn't it cheaper just to reduce anxiety rather than to increase range? Because if the way to reduce range anxiety is always to increase range, we'll end up with electric cars being heavier than they need to be, more expensive than they need to be, and you know, with 50 kwatt hours sitting outside your house 95% of the time effectively doing nothing. So reducing anxiety and I realized by the way how irrational this is and I explain the story which is it it fascinates me because one of the things I think that helps if you want to do this is to have some degree of metacognition. which is you don't just think, you think about your own thinking. It's apparently true of fighter pilots. The really good fighter pilots have good cognitive skills, but they also have good metacognitive skills. They don't just go, I'm going to do that instinctively, but they also ask, 'Is there a reason why I shouldn't be doing that in this instance?' Okay? Or this time it's different, as it were. And so my wife's car is a Mini Kooper electric and it's got about I I guess it's about uh 28 30 kWh battery and a range of about 100 miles. And I've got the Lotus Electray which is 112 kWh battery, a range of about 300 miles. And I got back from quite a lot of driving. I've been down to Wales and back and the car's down to 16%. Okay. And I'm going and all the lights have gone amber because it's at 16%. And I'm get I'm I'm going white knuckle on this. I'm going god I'm down to I'm 16%. Oh my goodness. I was down to 16%. Then I look at the actual range and it's about 58 miles. Now my wife's car that's 56%. Okay. In my wife's car we drive around at 56% i.e. with a range of 56 miles all the time without the slightest smidgen of anxiety. But when that's expressed as 16% not 56 miles, okay, I'm suddenly having conions and really panicking. Now, what that suggests is that range anxiety is much more a factor of psychology than it is of physics. And the problem with trying to increase battery range is that laws of physics are actually kind of set in stone to a large part. Okay? There's not much you can do about the laws of physics. Whereas the laws of psychology are magnificently malleable.

explains mechanism · 1gives example · 1supports · 1

03
Mechanism

The best way to differentiate is to find the metric your competitors have neglected and double down on it — reverse benchmarking.

Sutherland introduces 'reverse benchmarking' — rather than copying category leaders, find what everyone in the category neglects and do it spectacularly well. He illustrates with Will Guidara's restaurant turning mediocre coffee into a signature, Apple asking emotional questions in a tech-obsessed industry, and Buc-ee's transforming women's restrooms into a competitive advantage.

transcript

Rory Sutherland: I call this reverse benchmarking. In other words, you look at all the metrics that everybody cares about in the category. You find a metric that's been completely and ridiculously neglected and you double down on that thing. Now I first came across this in Will Gdara's fantastic book, Unreasonable Hospitality. He goes to the number one restaurant in the world. All of his team say, 'We ought to copy this. We ought to do that thing with the napkins. I really like what they do in the bathrooms with the scented jawsticks. Let's copy that.' And Gdara goes, 'Not going to copy any of that.' Because two, one, we can't afford to. And two, they're already doing it. What I want to know from you is what out of this evening at the world's best restaurant, a Michelangar restaurant somewhere. What was a bit disappointing? What was a bit me? And they said the coffee was a bit average, you know, it was nothing special. and the beer drinkers, probably the chefs who'd gone along, got treated really crappily, shabily compared to the wine drinkers. So he goes back to his own restaurant and he appoints a coffee sleier and a beer sumelier. And he says, 'Your job is not just to benchmark against these people, it's to hit it out of the park.' Now, if you think about it, taking something that's bad about the category, not not saying we need to raise our level to the category average, but instead doing it spectacularly well, something that nobody's expecting. That was what Apple did, I would argue. Okay. That's what Bies did. I don't know if you're Bies. Where where are you in the US? Yeah, I lived in Texas. We we we're big Bucky guys. Yeah. Okay. It basically started with an insight around women's restrooms. Now, you could have just had averely clean women's restrooms and you would have benchmarked. No, no, no. They're like the bloody hall of mirrors at Versailles, right? I mean, I haven't been in them, but the men's restrooms are pretty good, but the women's ones are apparently sensational.

gives example · 1supports · 4

04
Data

Most consumer behavior is irrational, and asking people what they want has poor predictive value — you need to understand hidden psychological bottlenecks.

Sutherland argues that economic logic and conventional market research have poor predictive value for consumer behavior because most of what we feel isn't even thought, let alone spoken. He proves this with a direct mail test for phone company services where offering both phone and postal response produced a 7% response rate — more than the sum of phone-only (2%) and post-only (5%) — showing that the ordering mechanism itself was a psychological bottleneck more important than the product or price.

transcript

Rory Sutherland: What you have to acknowledge is two things I think. One, you got to let go of two handholds at once, which is what makes it difficult. One is economic logic. You know, if we reduce the price, more people will buy it. Economic logic is has very bad predictive value. I think in the you know in consumer behavior and the other thing you've got to you got to partly abandon is conventional approaches to market research where you you think that we've asked the customer what it is they want and they will tell us because most of what we feel isn't even thought let alone spoken and also there are very very large areas of consumer behavior which are they may be metarrational at some level but nobody will ever tell you, I would buy that product if it were more expensive. But there are plenty of documented cases of exactly that happening. We tested three kinds of letters, 50,000 people each, randomized, completely randomized selection as with the newspapers. And one lot got postal only, one lot got phone only, and one lot as before got the choice of phone or post. Post only 5% response rate. Phone only 2% response rate. When you offer people a choice, it was just it was 7%. not quite but almost nearly I think the sum total of the of the two independent response rates that that may not surprise strike you as all that weird but to an economist this would be completely baffling because it would suggest that the more important factor affecting whether you bought the product was not what the product was or how much it cost but how you were able to actually order it. And from that moment on I remember thinking okay this is extraordinary because every now and then there are almost certainly brilliant businesses which are failing because they've failed to unblock some sort of psychological bottleneck.

explains mechanism · 2gives example · 1provides context · 2

05
Mechanism

Public companies are structurally risk-averse because accountability pushes downside avoidance ahead of upside opportunity, killing innovation and marketing.

Sutherland recounts Richard Thaler asking eight division heads if they'd take a bet with a 50% chance of 50% profit increase and 20% chance of 30% loss — six said no because they'd lose their jobs. The CEO wanted them all to take it for aggregate gain. Sutherland argues this structural risk aversion makes large companies conservative, and since innovation and marketing are 'fat-tailed' activities where 10% of efforts produce most value, eliminating the discovery layer in pursuit of efficiency proves fatal.

transcript

Rory Sutherland: Richard Thaylor the Nobel prizewinning economist and behavioral economist the author of Nudge. He once spoke to a board of about 10 people very large company. And he goes to the eight heads of the largest divisions of this company and he asks them all simultaneously a question. Would you take a decision if it had a 50% chance of increasing your profits next year by 50% and a 20% chance of reducing your profits by 30%. And six out of the eight of them said no. And Thaylor goes back and says well you know you're all good enough mathematicians I assume to realize these are highly favorable odds. to a gambling man, you know, this is a very very good bet and yet you declined to take part. Why is that? And they reply six out of the eight of them uh because uh 20% of the time or 30% of time I'd lose my job. And then the interesting thing happens which is the chief executive is sitting at the end of the table and looks a gasast at the eight people and goes but I want all of you to take those odds because net net in aggregate we'd almost certainly end up massively better off. Yes, two divisions, one division might have a slightly disappointing year but four of them would perform spectacularly. And you realize that the way that businesses are structured, as you push responsibility and accountability further and further down the organization, they become more and more risk averse and they become more and more uncertainty adverse. So they would prefer a definite 5 to 10% to a probabilistic a 50% chance of 50%. And what happens then is that you fundamentally you become highly conservative. You're more worried about downside avoidance than you are upside opportunity. And as a result, obviously both innovation and marketing, I would argue, are fat tailed activities where 10% of what you do is probably more valuable than everything else. You know, you can't tell in advance which 10% it's going to be. They are processes of exploration and discovery. And what you do is you get rid of the discovery layer in the pursuit of efficiency. And in the short term it looks like a great idea but in the longer term it it proves fatal I think because you've lost the capacity to adapt to reinvent to reposition in pursuit of the occasional um the occasional breakthrough.

explains mechanism · 1provides context · 1supports · 5

06
Mechanism

Explicit trade-offs work — if you tell customers what they won't get, they accept less and love the focused experience.

Sutherland explains that customers are happy making trade-offs so long as they are explicit. He uses Moxy hotels (tiny rooms, no room service, but great ground floor and bar check-in — 90% love it, 10% expecting Marriott are pissed off), the Slate Truck (a $25K bare-bones electric pickup), and his own coffee concept 'Flat White or Off' (two options only, name makes the trade-off explicit) to show that expectation-minus-reality is what matters, not absolute quality.

transcript

Rory Sutherland: Expectation minus reality, you know, is the formula that matters. You can provide less of something so long as you don't actually disappoint. Have you seen the slate truck? Yes. What is it? So, it's an absolutely minimalist electric vehicle. It's about $25,000 and you can effectively customize it with all manner of sort of automotive bling. But but but the the core trade-off is up front. They're like there's no heat seeders, there's no screens, there's just knobs. You got to crank the window yourself. Um, it's basic. We're not filling it up with stuff, but it's if you want simplicity and you want the price to be, it's a $25,000 electric pickup and you're just not going to get all this random junk that every car company puts in. We're not doing that. So, it was absolutely explicit and in a sense that's what, you know, Avis did with we're number two, so we try harder. This is, you know, don't okay, we don't have the scale of the other guys, but we make up for it somewhere else. And the consumer is very happy making trade-offs so long as they're explicit. The problem with a lot of kind of cost reduction and um you know and uh efficiency pursuit in business is it isn't chosen by the consumer. It's imposed on them which is a very very different thing. But no, the slate truck is a brilliant example of what you might call explicit minimalism. But but it's brilliant. They they underdid the competition where everybody else tries to overdo the competition and they were just very explicit about it and it resonated like crazy. I mean, I put I don't I've ne I would have never bought a pickup truck and I put a deposit in. I couldn't help it. I I love what they I love what they're all about.

gives example · 3supports · 2

Highlight slides
Related episodes