Transcription
Rory, thank you so much for joining this Liberty Ventures podcast. Now, for anyone who's really not familiar with Oglev or your background, I'd love it if you could just start with a brief overview of what's led to who you are and what you're up to today.
So, I mean, I could go back uh to sort of childhood, but I mean, suffice to say that I left university having studied classics, not really knowing what to do. I spent a year training to be a teacher and then decided I couldn't quite face going school, university and then back to a school. So applied to various ad agencies and was I think very very lucky. I think it was just one of those perfect um a perfect combination of time and place that I started to work uh as a graduate trainee in what was then the direct marketing part of Ogulvie or Ogulvian Mather as it was then called. uh that was then called Oglevian May the direct now called Ogleby one and it was effectively I think the perfect place uh in terms of both culture and of course the discipline of direct marketing which then of course became turbocharged with the advent of digital media and so you know I was just privileged to be I think in a lucky place at a lucky time.
Ogulvie itself was founded in the United States. Well, it's complicated but effectively was founded in the United States by David Ogulvie who was an expat Brit who had spent his time I think in World War II working with people including Ian Fleming uh in effectively British the British Secret Service I think in particular trying to get the United States to enter World War II which obviously which plan ultimately succeeded. um and then spent a time variously as tobacco farmer and market research executive before founding what became I think you know one of the two or three most interesting advertising agencies of the second half of the 20th century. And so I was lucky enough I met him once actually and I was lucky enough to do that as well. But um it was just a very very um pertinent and and beautifully uh relevant time at which to start a career effectively six years before the web came along.
Um, I also had a second privilege uh which was that my brother was an well still is an astrophysicist. Um, but he was then in Burke in in the early 90s he was in Berkeley California where he effectively roommed with someone who'd previously been the roommate of Tim Berners Lee at CERN and Tim Berners Lee being effectively the father of the web. I got to hear about all that disproportionately early. I failed to turn it into a billion-dollar fortune, but it did at least mean that I was sort of, you know, six months ahead of the curve when that hit. Um, and so that was that was also a lucky accident of timing really.
So, there's a lot for us to go on there, but you also haven't really gone into some of the other incredible things that you've done in in the in the advertising space. You've written several books and you really have helped push the the the world of advertising forward. And so I'm actually really curious to hear how it's evolved over the course of your career. You talked about direct marketing, which for anyone who may not know what this originally was because they're so familiar with the digital world now, this this was and still is to a certain extent sending physical letters out in order for people to re in order to discover things. So the only individually addressable media were really the telephone and the post, the physical post. And so, you know, it was it was sometimes called database marketing because you had a database of customers and you wrote to them.
According to various stories, it was actually invented by Benjamin Franklin, funnily enough, when he ran a bookshop in Philadelphia. Uh he he I haven't heard this yet. I to be honest, I've never quite been able to verify it, although it seems to have appeared in the Financial Times. Uh you know, enough people believe it for it to be possibly true. Um, and he took to effectively when he ran a bookshop in in Philadelphia, obviously some of his customers lived some distance away and of course effectively Franklin had created the postal service. So he would then write to customers and suggest exactly as Amazon does today with people who liked this also like that. You know, people who bought this also bought that. Franklin would do effectively the 18th century equivalent of that by writing to people and suggesting other books by authors they'd bought before or books that he thought were complimentary to the kind of books they liked uh previously.
So um the other form of it was of course where you ran a press ad which solicited a direct response from the customer. In other words, when I say when we use the word direct it meant that the you sold directly to the customer. you ran an ad in the newspapers, uh people would reply with a coupon which was coded. There was generally a code in the coupon which which related to both in which publication it had appeared and which creative execution it represented. And that made possible something which was the testing of obviously different media on the one hand but also different creative approaches.
Um, and so interestingly if we go back to the early history of this the advertising industry effectively invented the randomized control trial uh probably about 60 or 70 years before it really became popular in for example medicine or other forms of the human sciences. uh the the way it worked and this is an interesting bit of history and I I should research this a bit more but American and British newspapers that were produced in volume were produced on two presses in parallel and I don't know whether this was at the behest of the advertising industry but somebody ingenious said if you place on one press one creative uh um execution and a different advertisement in another creative execution and then you interleave the newspapers as they come off the two presses or in some cases four presses. So it was either AAB or ABCD ABCD ABCD. That was important because obviously if you I if if newspapers from the different presses went to different locations, it would invalidate the test. But because you interleaved the newspapers ABA, you could assume that pretty much the people who got um execution A and the people who got execution B were were representative samples. In other words, there was no geographical distribution of the print output to skew your results, then you could confidently say that um uh you know, you could confidently then say that if if if advertisement A drew 30% more coupons than advertisement B, then in future you should run a hell of a lot more of A and a hell of a lot less of B. So it was effectively this by the way I think was being pioneered certainly in the Edwardian era possibly in the late Victorian era. And of course what what these were thinking back on it nobody nobody thought of them as such at the time but they were social science experiments performed uh you know at enormous scale.
And very quickly in this form of advertising you learned that now in just as an interesting detail you weren't allowed to test different prices for example or different fundamental offers because the newspapers argued I think with some justification it isn't fair for some of our readers to be given a better deal than other readers. Um, and so you the price had to be the same of what you offered. you couldn't offer credit terms on one and none on the other, but in all other ways you could you could test totally different um uh press executions. There was also DRTV of course which you'll know and QVC being the most sort of uh probably the most obvious, but there were also things like late night infomercials that came along later. But essentially the origins of this form of advertising were actually quite scientific.
Um, and what you learned very quickly was that the kind of things that economics would tell you was rational and b the kind of things that people told you in market research didn't have nearly as much predictive value uh as a really solid well-grounded test. And it also meant I mean there are wonderful books since there's one by I think Timothy Wilson called Strangers to Ourselves. It was also highly, I think, highly revealing of human psychology to the extent that people would always tell you that they'd be much more influenced by, you know, an offer of 10% off than an offer of a free cuddly bear. But when it came to actually testing the reality of the thing, you very often found the reverse. Um, and so it was highly revealing of the fact that uh you know large parts of the unconscious um could be reached by advertising in a way that no at least in a way that market research simply couldn't reveal uh because nobody's going to tell you that kind of thing. So I mean inherently inherently a wonderful place. David Ogdelby always said that where possible you should start your career particularly as a copywriter working in direct marketing so that you learn what works. He was you know a highly creative figure but also absolutely on the money in terms of effectiveness of advertising. And so as as a consequence, you know, in the period before the internet became a thing, I the internet existed for quite a few years beforehand, but when the web effectively made the internet a consumer proposition rather than an academic and military one, um this was actually a great place to be. I would argue now that things have gone far too far in the opposite direction. And this often happens. I would argue absolutely down a hill for the fact that it is good to be able to measure the effect of your advertising. We've now created a world where unless you can measure it, you can't do it. And I see increasingly, by the way, that's a trend in large parts of business where the need to quantify can actually override the the uh freedom to make a good decision. In in other words, I think business has become obsessed with this false god of perfect quantification. And I think in marketing, there will always be things you should do and should do extensively uh which you can't perfectly measure, but that shouldn't be an obstacle to doing them.
So it sounds like what you're saying is that marketing and advertising's principles have remained constant throughout history even with new technology and the new types of advertising available today. Is that right?
Uh there are I I would say there are certain okay this this will vary because obviously there are cultural and contextual differences when when any human being makes a decision. um they and this might be an area where actually eastern thinking is rather better than western in that there's a generalization but that in the west when somebody does something we tend to attribute it to their personality i.e something innate in them or something about their character in the east they're more likely to look at the circumstances and context within which they made the decision so the the standard way in which this is often voiced is that in if you have a serial killer who suddenly appears in the United States people you know or a school shooter people will immediately leap to a description of his personality i.e bit of a loner, kept himself to himself, etc., etc. Whereas in the east, they will say he was under considerable stress at work. They will look for external events that that drove the behavior. I'm not saying neither culture is is obviously um 100% one way or the other, but it's it's a tendency in terms of looking for explanations. And obviously the same person in different circumstances will make very different decisions in a different context, in a different mood even um will make very different decisions. But at the same time there are um this is not the laws of physics but there are general recurring patterns which you would be well advised not to ignore. That's that's the way I put it.
So what I always say when we run a behavioral science practice is we cannot tell you perfectly what to do. What we can do is surprise you with what might be worth testing. So we might go to somebody and say this little detail here in your website for example you know in in the design of your website seems relatively trivial. However, if you test different approaches, you may find that one or two new approaches will actually achieve a kind of butterfly effect. In other words, it will pay back, you know, many fold. And we one of the things I would most encourage I was talking to a wonderful she's actually an American resident in the UK called Margaret Hefernon who's just written a book effectively on embracing uncertainty and she argues that we are overwhelmingly and I think business culture makes you know values decisiveness to an extent where people are now dangerously unhappy to say I'm not sure I need to find out let's investigate this and they also have this slightly the naive belief that the data will tell us what to do, which is a convenient belief because you can never be blamed for doing what the data tells you to do. Okay. Um, but actually uh one of the things Margaret was saying is she finds it absolutely baffling how little businesses test an experiment. And she came up with a wonderful phrase this morning which is she said actually all people say well what if the test doesn't work? Okay. And she said in a funny kind of way, all tests work because even if they fail, you still learn. Okay. you know, if you do something which seems wonderfully sensible and the opposite happens, which by the way is it's you know, it's by no I mean I think I think there are tests where you can say it's highly unlikely this will have a downside and and the upside may be considerable and you should always engage in those tests because it's a very asymmetric outcome. For example, you know, we made British Airways at one point, you know, you know, a seven figure sum by changing the button from find flights to find my flights. Okay. Uh, you know, now the point about that thing is, you know, those are the kind of tests which which Jeff Bezos would call them a two-way door. you know, there's not much actual irreversible commitment to doing them and the potential gains can be huge. And it's a mistake to argue those things to death because in the event that that doesn't work, which is I mean it's not impossible, but in the event that it doesn't work, you just reverse it very quickly. It and what we tend to do in business is we we make two mistakes. We test too little, but we also sometimes just apply too much rigor. So that Jeff one of Bezos I think is a very interesting um uh decision scientist or at least his approach to decision making is unusually nuanced and intelligent. And one of the things he noticed as a flaw in business is obviously there are irreversible decisions like you build a 10 million square foot distribution hub you know north of Memphis. That's a pretty hefty commitment. It's irreversible. you need to make that a pretty rigorous decision because if you mess it up, uh, you really are trapped. Whereas he argued there are an awful lot of things where to be absolutely honest, you know, he called them two-way doors. If you don't like what you see through the door, you just walk back. And I I think I think that's a fair observation that we tend to we tend to see the quality of rigor as emblematic as of the quality of the decision. And actually if you're if you bake probability into the decision and you say well you know what are what are the possible scenarios that emerge after this and what happens if for example our test doesn't work very well and the answer to that is will you stop doing it very quickly in which case you know the downside is measurable and the upside is potentially disproportionately high you you know any good poker player would always take those odds okay you know it's kind of tales I don't lose very much, heads I win. Um, and I don't think, you know, and I think one of the interesting things that happened at Amazon was that apparently the person who initially made the case for launching AWS or Amazon Web Services didn't do a sterling job of making the case and people were pulling the whole thing to pieces. At which point Jeff apparently said, "Look, this is a two-way door. We've got to invest a lot in bandwidth and processing power and everything else. uh we got to do that anyway because we're Amazon. If we can sell that on to other people, whoopde-doo, fine and dandy. And if we can't, well, we haven't lost very much anyway. You know, there's not much to be lost in trying. And those are the kind of things where I think um we do need to There's a great book, by the way, Annie Duke's book, Thinking in Bets. She's a former professional poker player. Great book. And so yeah, I mean one of the strangest things is how little attention we pay as humans to thinking about decisions and thinking about you know different ways we could approach decisions, different questions we could ask, different ways we could define the problem. And I think the reason is that we do have a an inherent uh psychological tendency that broadly speaking once we've rationalized something we consider the job done. I've come up with a rational explanation for why we should do this. Consequently because it's rational all other considerations might are somehow wrong or sub optimal and move along. Now nothing to see here. And actually, it's very very possible to be highly rational and completely wrong. Okay. And yet we don't really we don't really consider that possibility. It's also by the way you know it's also possible to be right and fail. That's something poker players would know that you can make absolutely the right bet under the circumstances given the information available and you can fail. We're also prone to that thing that Daniel Carnean called whizattati which is what you see is all there is. So people inherently in are reluctant to say I need more information. I don't know enough to make a decision here. Uh I'm not comfortable with the information I have. Let's explore further. What we tend to do is we'll take whatever information lies to hand and we'll then construct a sort of rapid on-the-fly model using that limited information and we'll then proceed from there with the kind of confidence we'd show if we'd had all the information we needed.
So I I mean one one interesting thing I've been talking about quite a lot is I think um you know really good decision making is a bit like police work and what I mean by that is police work actually accepts the fact that there's an investigative phase and then there's an evidential phase ultimately yes you've got to prove it okay I get that you can't you know you you can't imprison someone for 20 years on the basis of you know a suspicion or whatever. Ultimately, yes, you should proceed towards ev, you know, evidence. However, what I think we've lost um is the acknowledgement that we don't have enough information that the and by the way, we we also probably don't acknowledge the fact that context may have changed, that what worked in 2017 no longer works in 2025. Um, and I think as a consequence of that, we've telescoped One one question that always fascinates me that the police ask when you think about it, okay? Is they will go doortodoor if someone's murdered and they'll ask a question like, "Did anyone notice anything unusual last night?" Now, when you think about that, that is about the most open-ended question you can possibly ask. You know, the answer could be literally anything. You know, the dog didn't bark in the night if you're Sherlock Holmes. It could be I heard a scream. It could be nothing. Okay. Uh it could be a strange van was parked outside. None of that stuff has any evidential value. Uh you can't make a decision. You can make a decision on the on the basis of it, but the only decision you can make is I need to investigate this further. What you can't convict on the basis of it. Um, but that's that's for later. And I think Margaret Hefernan's point that we're so deeply uncomfortable with ambiguity that we seek to remove it from our lives as hastily as we possibly can. And she makes the point that creative people have a distinctive characteristic or people who people who are prone to coming up with creative solutions or ideas have a distinctive uh characteristic which is I mean for whatever reason and one of them could be laziness by the way sheer idleness. Okay. But they are actually happy existing in a state of not doing uh you know, not knowing. And you know what one of the benefits of of waiting before you make a decision is simply that new information may emerge or a new idea or insight may emerge or in inside your own brain you may look at a problem in a completely different way and realize that what seemed to be problematic no longer is.
So I mean obviously in your business in in investment it's very very interesting in the sense that um people are desperate to try and construct kind of rules and by the way I think there are checklists. I'm not such an anarchist that I'd say there's no value to checklists. I think it always pays to go is there something the cop thing will be have we forgotten to collect DNA? Have we looked here? Did we do a search of that? I think checklists are immensely valuable um because they prevent you from just sort of either willfully or accidentally overlooking things but I don't think there's actually a process in terms of where you handle information because what information emerges then determines what further information you look for. I by the way I mean an interesting aside on this is I thought that in the COVID outbreak some of the failures of this need for complete certainty I thought were made manifest where people were unw first of all the need for experimentation was f was much greater than people acknowledged. People went on information that was relevant to previous corona viruses even though this one was novel. And then I noticed people using phrases like the science tells us when there's no such thing as the science. Okay, science is not a singular. And and and it strikes me as really interesting unless there was an ulterior motive. It was interesting to me that they were so slow to clock that it was basically transmitted. It was it was airborne transmission in confined spaces over a period of time. And I've got a friend who's a doctor who investigated this who said you could have avoided nearly you you would probably have had to close down pubs, restaurants, churches, confined spaces, but in warmer climates we possibly could have led our lives perfectly comfortably wearing masks out of you know wearing masks and uh not much more. But the but again that you know I I support the precautionary principle there. it's not worth playing fast and loose. But I think it was fairly obvious there wasn't actually that the transmission conditions were quite it quite clearly delineated the conditions under which it would have been transmitted and you probably could have made things quite a lot easier.
Rory, I can see why the Guardian called you one of the most unlikely Tik Tok sensations of the day and how you've been so successful on Tik Tok and social media. And I want to say thank you for sharing all of this. I wish we had more time, but is there anything else that you want to share that we that we haven't got to? I know there's so much more that we could get to.
Simply that uh very simple thing I would argue that uh this is just a useful insight. If if you're talking to people who have a great invention, the natural assumption is which I made myself that the bigger your idea, the less marketing you need. And we traditionally looked at things and thought well obviously if people are deciding between two near identical laggers or they're looking uh to decide between two, you know, fairly well functioning washing detergents, then obviously marketing is going to have an effect and people will choose the one they you know that where the message is better conveyed to them. But they always thought that we don't need marketing if we've got a really big idea. And my contention would be that um uh uh actually my contention would be that um uh when you have a big idea, it requires a lot of behavioral change. Humans find changing their behavior fundamentally difficult. It requires them to abandon an old habit. Sometimes in the early days of a new behavior, it requires them to do things that nobody they know does. You know, I had an early mobile phone and you you would get abuse from using it on the street. Okay, we forget all this stuff. And actually, humans find behavioral change difficult. So, a big idea actually requires a lot of marketing thought, not a little. And I think we've often made this mistake that people in the tech world often assume that my idea is so great it will win uh on its you know superior product attributes alone and there's no need to actually uh practice marketing or translate what I can do into human language or human benefits and I think that's you know I think that's fundamentally wrong I I and my argument would be that to to get people to do anything new or innovative Any to give you something contemporary, the mobile phone, that's done and dusted. Okay, we've created a social norm there. You're considered now an oddball if you don't have a mobile phone, etc., etc., etc. That's no longer, but I spent 10 years of my working life um not all of it, but I mean long periods of my working life for 10 years were persuading people to get proper broadband at home rather than dialup or in the early days persuading them to get dialup internet. Okay. Would you believe it? Okay, that that sounds ridiculous now. You wouldn't move to a house with no internet access, but at the time that was a 10-year process. The iPhone did not take off that quickly. I mean, that was that was, you know, one of the most successful product launches in the last 30 years, but if you actually look at the data, it's still that sigmoid curve of early adopters. And at the moment, we have things I think like solar panels. the um the but but the way I describe it is that in many cases actually I think very good products and ideas have failed because no one's actually worked out how to market them. Um, uh I don't think you can market a crap product successfully except in very weird areas like sort of fashion. I mean there are you know there are weird sort of outliers where uh you know products which purely have a display value for example and have no utility you can you know the French fashion industry to an extent is you know 90% marketing um uh but um uh but but at the same time I think one of the things we don't understand with things like solar panels heat pumps electric cars etc that uh you know you will always encounter at the very least friction and sometimes push back. People don't like being told that what they've always done is wrong. Okay, we probably forget that because we're naturally kind of exploratory, innovative people, but there's a very large sway of the population who kind of resents the fact that they're being told that something that they've made work for 30 years is now wrong. But all of those things, I think, are going through that same period of sort of painful adoption. Um, you know, you could extend that to air fryers, you know, fantastic category, which I was a very, very early advocate of. And I could see it was one of those technologies which would ultimately succeed significantly because nobody I knew who bought an air fryer ever went back. You know, nobody ever said, "Okay, I can do without house. I'll get rid of that and I'll do without it." So, it clearly had a really strong value. But again, you still encounter all of those things. It's not just I mean in many cases the financial element of this may be secondary to the emotional component and yet what we tend to do is we hand problems to economists who assume that everything's just a matter of cost and benefit and consequently I think there's a huge amount of misdirection of effort.
That is such an important lesson to leave the Liberty Ventures Network with. Rory, we need to get more time for another podcast and I'm so and get you out to one of our events soon. We're going because there's so much more I want to get into with you. I'd be absolutely delighted. Let me know about all your events. I'd be so happy. Wonderful idea. Wonderful. Thank you so much, Rory.