Sir Martin Sorrell on the 5 biggest ways AI will change marketing

0:06Ladies and gentlemen, welcome back to Uncensored CMO. Now in this episode I am talking to one of the titans of our industry, someone who I think has had a bigger impact on our industry than perhaps anybody else. He is Sir Martin Sorrell, founder and owner of WPP, the biggest holding company in the world. He has since gone on to set up S4 Capital, so now finds himself in the challenger position rather than the dominant player. I wanted to talk to Sir Martin about what he sees as the biggest challenges facing our industry today. What are the disruptions coming down the line that are going to affect and shape our industry in the future.

0:42What's his advice to CMOs? What do CMOs need to care about and what should they be doing? What skills they need to deal with the challenges coming at them in the world today? And because it's can, I wanted to ask him about AI, of course. And in fact, just to do that, I asked ChatGPT to give me some questions about AI to Sir Martin. I also want to find out a little bit more about him as a person, his background and what motivates him and why is he still going having had the success he has here's my conversation with sir martin sorrell listen i thought oh it is okay obviously you know you're reasonably well known i think it's fair to say a little bit but i'd love to start at the beginning in my own mind in your own mind yes indeed take us back to uh when you were growing up what did you want to be i wanted to open the batting for england okay fair enough nobody understands what that means i was very um leonard hutton was my sort of um idol i think the york england captain sir leonard hutton actually so i probably want to open the batting for england was probably the answer to that um i got into to advertising by accident really because my dad said uh you know what you should do is you should find an industry you enjoy find a company in it within it that you enjoy and have fun with and then build a reputation.

2:02Not a reputation to do podcasts with you, it can, but just in the area in which you operate. And then if you fancied doing something on your own, and the magical age was 40, actually, because 40 was the age when you'd done your first 20 years because people used to retire at 60. They didn't carry on like I've done. but so you know build a reputation if you fancy doing something on your own when you're around 40 go ahead and do it so that was it and I sort of fell into it because I was working for James Gulliver who was a very successful food retailing entrepreneur he sold fine fare to RCA corporation don't ask me how a retail food operations ends up inside RCA corporation but it did and James had a fair bit of money and he hired me as his personal financial counsellor, which actually wasn't an accurate description.

3:00I was really his gopher. He used to carry his bag. And we invested in a number of things. We invested in Tavener Rutledge, which was a suite manufacturer up in Manchester, Sayers Confectioners, which was a bakery, also up in the north, in Alpine Double Glazing. these were all sort of small public companies which had done pretty well or were stagnating and then the other thing was Garland Compton and Garland Compton was run by a man called Ken Gill and it had gone public so it was the Garland agency in London plus Compton in America had a stake in it and Ken Gill was worried about the creative reputation of Garland Compton, you know it had Procter as its biggest client jane jay i think eli lily and some others and he said to james a good friend of james gulliver uh said you know who i'm thinking about goosing up the uh creative reputation and i've got you know these saatchi brothers and and what morris and charles did was effectively reverse they did a reverse takeover of garland compton which had gone public and james was consulting for them and i ended up being the sort of consultants.

4:15I used to go into Charlotte Street one day a week. I had an office in Charlotte Street, not on the top floor where Morris and Charles were, but I think it was on the second or third floor. And I used to do consulting for them for about half a day a week, a day a week. And then Morris wanted to hire a CFO. And the consultants were with a firm called Goddard, K and Rogers. And the consultant was called Brian Burwash. I mean, it's all very funny stuff, actually. And Maurice said to Brian one day, because they had difficulty finding a CFO for Garland Compton as a listed company. The guy who was running, it was past 11 o 'clock in the morning, was somewhat non-compost mentis.

5:00And they went looking in Fleet Street for a CFO. Most of the CFOs were non-compost mentis after 11 o 'clock, if you get my drift. Yes. And Brian said to Maurice, what sort of person do you want? Somebody like Martin. So have you asked Martin? No. So Brian asked me and that was it. Wow. So that was how I got it. I sort of fell into it. And Maurice I liked. And Charles I liked. And the atmosphere was great. Tim Bell was there. Jeremy Sinclair. Bill Muirhead. I mean, all of whom have survived to this very day. And it was an exciting place to be. Nothing was impossible. Yeah. except if you've got public recognition for it.

5:38So it's very funny, actually, because Morris would say to me, or say to everybody, Procter and Gamble never have an individual. It's always Procter and Gamble, both of whom are dead. And, you know, Sartre and Sartre would be the same. And I'd say, well, Morris, but the slight difference, you know, you're alive, Charlie's alive, and Sartre. So you could do anything you liked, as long as you didn't get any public recognition for it. And Tim was, often people referred to me as a third brother or Tim as a third brother. Tim was really the third brother. Sadly, died a few years ago. But Tim fell into that trap with the Conservative Party and Margaret Thatcher.

6:23He became too famous as an individual. We'll come back to this because I think it's an important point in our industry. it's what bedevils our industry to some extent anyway so Tim carved out a really with Gordon Reese Gordon Reese was Maggie Thatcher's sort of political guru the one who lowered her voice and looked after you know how she looked and what she said and Tim got very involved as a personal advisor and that's but Sarcher's was a great business show Manhattan Landing the work for Israel the pregnant The Man ad, I mean, it was all part of the folklore. Not that you see anything like that at Cannes anymore.

7:04Well, we're going to come on and talk about that. I'd love to get your views on it. I don't think the ad that BA did matches Manhattan Landing. Well, let's jump to that. They won an award here this week. How do you assess the state of creativity now as you look? Well, firstly, how many times have you been to Cannes? I would say, I don't know, it's about 15 or 20 times. It never used to come. We basically set ourselves the objective at WPP to win at Cannes. Yeah. And we had a run of, I think it was about five years. I think it started about 2014. And when I left in 18, they flubbed it in 2019, and then they seemed to reassess the importance of it.

7:48It's very different to what it was. I mean, it's certainly nowhere near what Roger Hatchwell envisioned. I think it was a bunch of French creatives having a glass of rosé in the south of France in June. And I don't think if you and I were trying to redesign or set up the equivalent of Cannes, that we would do it in June in the south of France. I think we would probably pick an off-peak period like February or March or something, or maybe January rather like Davos and find a place where the rates were slightly lower and maybe a bit more isolated so that people couldn't do anything else. I mean, the trouble here is that there's lots of other things to do in it.

8:36And in fact, I saw it. It's really interesting you asked me the question because I saw somebody today who was her first time here who worked for a UK company, a retailer. And I said, you know, what do you think of it? and she said, well, nobody looks at the work. And that is true. I mean, most people bugger off on Thursday now. They don't stay for the awards on Friday except the winners. Whereas before, that was very much the reverse. We all stuck around. We wanted to see who won, et cetera. So I think it's very different. To me, Cannes has become like CES, and CES has become like Cannes. And maybe the common coefficient of that is MediaLink.

9:18Because CES was too technological. Yes. So they wanted to be a bit sexier, so they called on MediaLink. And then, you know, CAN with Essential or MediaLink is not part of Essential anymore, but it's the same thing. But I think it's, I'm going to say it's past its sell-by date. I think it's got to change. In what ways do you think it's changed? Well, I think a number of people said to me today and yesterday and the day before that it's too much. And it's become very superficial. So, you know, it's speed dating, which is good because it brings everybody together at the same point in time. So it's a networking event.

10:01But, you know, you lose sight of the event. It's rather like Davos in the sense that it used to be that we used to listen to the content, to the panels and the speeches and the initiatives. I sort of feel that this is more you know there's a lot of people here don't have badges the price of badges has gone up so significantly price of a hotel room is ridiculous I heard last night that the pre-COVID agreements that the hotels entered into they're not prepared to stick by them so you've got pricing and dare I say gouging on pricing and I think you know the funny way can and this is the trouble with all conferences they're not in real time so they're planned a year before so this planning might have been right for a year ago but it's not right for this point in time and some of the things going on here are probably downright embarrassing from an external point of view in terms of cost and entertainment and it almost comes down to who's my party is bigger than yours or whose party is the biggest And so I think it's sort of lost a bit of its focus.

11:13Now, having said that, we'll come on to this. Technology is, there are two drivers of, I think, of our industry. One is geographical, a geographical driver, and the other is a technology driver. And obviously, technology is paramount. And you see the six big platforms represented here, the three western and the three eastern. And you see Amazon and Apple and Microsoft with big presences as well. And, of course, Salesforce and Adobe and LG and Samsung and Shopify and Spotify and Oracle and scads of others. But Pinterest and Twitter and Snap and et cetera. It's very technologically driven. And I think it's lost the focus around creativity.

12:00so we can get into that a little bit more well let's obviously uncensored cmo podcast with you know talking to an audience of cmos what should cmos be concerned about now as you know with your experience looking around the world well i think there were there were three things that we bang on about and i had a meeting with a major global or two major global customers this afternoon before coming here and it permeates those conversations i mean obviously ai is a focus but just we'll come onto that as well but we're abstracting that from that i mean i think agility people talk about agility every ceo says their organization is agile including agencies and if you have to deal with them you find out that they are so agility is key taking back control is key and as well you know after the great financial crisis the the vogue thing was zero-based budgeting which was nothing new but you know 3G Jorge Lehmann and his colleagues really sort of got people to focus on it and that meant you know defenestrating everything starting from the base of zero and building up it wasn't a new concept I mean it was messaged as being new but it wasn't I mean when I was at business school 55 years ago we had it Hal Janine at IT &T used to to practice it was nothing new but what it did was it pushed a lot of stuff out of the client.

13:24So it was outsourced. So we have three models at MediaMonks, S4's operating brand. We have the classic outsource model where agencies operate. We have an embedded model where we put people into the client. You know, they might have the client's T-shirt on, but there are people on our payroll. And then you have the in-house model, which we're not frightened to do ourselves out of business. Most of the holding companies are terrified of losing. We think in-housing works in various situations, both media and creative. We were the subject of a Harvard Business School case study on the in-housing of Sprint.

14:03We went on to do T-Mobile as well after they took over Sprint. So I think taking back control is really important. Interesting at the moment, because there's this focus on short-term results and short-term performance, which I think will continue. You know, the idea of in-housing probably CFOs and chief procurement officers are saying, no, we don't want more headcount, so outsource more. But in time, particularly driven by AI and AGI and blockchain and metaverse, which I think are the three technological entities and crypto, sort of fading a little bit into more than a little bit into the background.

14:43But those three things, blockchain, meta and AI, AGI are fundamental. And I think that will drive things more in-house. And I think that will become it. So take back control. So agility, take back control. And then the last point is first-party data. So when Google announced the deprecation of third-party cookies, although we still haven't got the final timetable, and when Apple did their change in IDFA, that meant that clients who had been building third-party data platforms had a real problem. I mean, it was first announced by Google, I remember. but the last thing before COVID in January of 2020 that I went to was a conference in Scandinavia, and Google had just had the second blog on deprecation of third-party cookies.

15:30A lot of the clients there, although at that conference, a Google-organized conference, although they were great fans of Google, otherwise they wouldn't have been there, they were somewhat irritated by the fact that there was a deprecation of third-party cookies because they were building third-party databases. Yeah, exactly, yeah. And so first-party data becomes really important. So you see the growth of the retailers' walled gardens as well, which is their answer to the Amazonian surge. So I think those three things. So, you know, agility, taking back control, and first-party data. Now, almost reluctant to mention the word, because there always seems to be a word, isn't there, or a theme at every can, but we must talk about AI, of course.

16:09Yes. How real a game changer do you think AI is? Oh, I think it's real. And where do you think the impact is going to be? I think real. When people talk about AI or AGI, there are two sort of emotions that it engenders. One is wonder. I mean, isn't it incredible and it's going to be doing incredible things? The other is fear. So I did a conversation with Auntie Scaramucci in London recently. and a woman came up to me with her daughter afterwards and said, my daughter works at one of the holding companies. I won't say which one. And is she going to lose her job? And, you know, she might. She might.

16:54Could be the answer. But I said, well, not if she learns to use it as a tool. So with that as background, you know, fear and wonder, or wonder and fear, because, I mean, it is as big as the web. It is as big as smartphones. I mean, it's another industrial revolution, and it's fundamental. We see already five areas where AI, AGI is developing. The first is on productivity, improvements on copywriting, and visualization. So ads that took two weeks to do take two hours. So huge savings, huge improvements in productivity. so much so that we've altered our go-to-market from faster, better, cheaper to faster, better, cheaper and now, sorry, and more and now being instantaneous implementation but more means we can produce more assets and be more effective.

17:50So I would say that's the first thing so visualizing copyright and I should say one interesting thing is the Unreal Engine which is Epic Games' Fortnite technology which we had utilized extensively. We built a studio in New Delhi, which means that we could shoot any commercial anywhere in the world from New Delhi. Wow. You know, if you want to do Rio de Janeiro, we might have to have Brazilian or Portuguese-speaking actors in India. But essentially, what's really interesting about the AI surge is the Unreal Engine also almost did the same thing. You know, you don't have to go to six locations and spend three months producing the most beautiful film, you can do it.

18:33But anyway, so that's number one. I think the obvious question in people's minds with that is what happens to originality and creativity and how do you avoid basically everyone creating the same... Well, no, you can do it to such a heightened degree. So for Netflix, which I think is the model, it's effectively a data-driven model where the data determines the creative and you pump it out through digital media, you look at the results, and then you refine the product. If it doesn't work, you refine it or you alter it. So for Netflix, historically, we might produce 1.6 million. In the Narcos example, 1.6 million.

19:13Theoretically, different creative executions, you could now do 5 million, right? So what was interesting about the conversation we had this afternoon was, you know, there were 12 product managers, global sitting around, And what it meant was that the production period gets concertinaed down to literally hours, which gives them more time. They produce more assets, which means that they have to be more selective in what they use and it has to be more iterative. So I think it's all to the good. I mean, it may be that clients, you know, we've had one or two clients who really said to us they want productivity savings or to share the productivity savings with us.

19:54But what it does is it enables, it's really interesting, it enables people to focus on the content and how effective the content is rather than the production of the content. So you've shut down that period and it's expanded the amount. So it's a bit like when you say about AI, I will destroy the mundane jobs and give people more time to do the brain jobs or the whatever, however you describe it. So that's the first area. The second area is hyper-personalization at scale. so we can do on a super scale, hence the more. The third is media planning and buying. And I think that, you know, the best way of putting it is, you know, you wouldn't – what would you prefer to do?

20:34Use a 25-year-old media planner or buyer or an algorithm? And certainly in the digital space, which is now two-thirds of client budgets, and the media market is$900 billion, probably digital media is about$600,$650, and it's going to be 75 % of client budgets by 2025, according to the forecasters. So I think the holding companies are in for a very rude shock on this. I mean, there is no way that if they employ 350 ,000, 400 ,000 people currently to distribute media planning and buying, it is not going to be the case in two years that that is the situation. Is that because they're overhead they're carrying?

21:17Well, there's no need for it. Again, I say the algorithm will prove it. You'll need agencies to validate. You wouldn't give your media budget to Rupert Murdoch and say, invest it without validation. No insult intended to Rupert Murdoch, but there would be bias there. So you need somebody as a countervailing party or as a reseller to validate what's going on. I also think it means that the platforms are going to get closer to the clients. So the six, I would come on to, I think, I also think that it means that the six big platforms are going to become even more significant. You know, they're being checked by the regulator from acquisitions, doing acquisitions, you know, Giphy and Activision with Microsoft.

22:01So six platforms, three West, three East, plus Microsoft and Apple, Salesforce, et cetera, they're going to become, NVIDIA obviously have a monopoly position in the picks and shovels for AI. I think they are going to become even more important. So if you thought they were big, now stand by. And the regulator is forcing them. Their research budgets are huge. Their research budget is a percentage of their revenues are huge, and in an absolute sense are huge. So they will be investing more in organic development, AI, AGI gives them. So media planning and buying. The fourth area is use of AI as a super tool, i.e.

22:39improving our processes, improving clients' processes. And the fifth area, which I think is the most interesting, is knowledge. So it comes back to what I said before about egos and industry. You know, good people are difficult to manage, which is a dangerous thing to say because average people then think to be good they have to be difficult. So you defeat your purpose. But the real jewels in our industry are those people who are good and good at building their brand if it's a multi-branded model. Ours is a unitary model that are good at participating in that, but engage other people who are good individually and make great decisions and make great progress, but they bring in the team.

23:23Those people are very rare. You walk down the Quasette. I'm still a shareholder in WPP. And what do I see? I see WPP, Group M, and Ogilvy. Makes no sense. What's the point? I mean, if you need more space, you're trying to create one firm. Even publicists, talk about the power of one. You walk into their reception and you see 47 brands. So I think our industry is bedeviled by that problem, that the individual almost becomes more powerful than the entity. and there's a great parallel with the fashion industry. There was a tremendous documentary recently, I think it was on BBC iPlayer, about Arnaud and Gucci and Pino and about Tom Ford and Alexander McQueen, etc.

24:19And the lesson from the film was, it's a great film, everybody should watch it, particularly in our industry, was that the creative directors weren't the most important people. It was the brand. The brand that always won in the end And Arnaud understands that big time. He's built the biggest company in Europe over half a trillion dollars of market value. Was the world's richest man. And I think Elon Musk has edged ahead again. But it's built by building, by understanding the brands are really important. I mean, the people are really important, but the brands is where the strength. And there's a parallel to our industry.

24:57So getting people and going back to the knowledge factor around AI, AI will enable us to enfranchise everybody inside the organization with the power of knowledge. So what was happening now is that people protect their areas because they control the knowledge. At least in theory, AI will enable everybody. We have 8 ,700 people. And I often say I wish everyone knew what the other 8 ,699 knew. And the guy who resigned from Google said he was resigning, he said it was an existential threat because the bots, he used 10 ,000 bots. Every bot would know what every other bot knew instantly. And in a way, it will get us to a situation not quite as perfect as that, but where if we give people open access and if the data that is ingested is good data and gets more and better and better, that will give knowledge to everybody in the organization so the thing that drives me nuts which is a question i always like to ask people what drives you nuts and about the business or the brand you're running or whatever it happens to be is when somebody in our organization says i didn't know that we had done that or so and so have been involved in that or we were in this geography or we had this capability or whatever so with all this in mind uh what's your advice to marketing teams what kind of skill set do they need to you know be able to handle this well you know if you think about ai the the the oil of ai is data and you know going back to the agility and take back control first party data mantra clearly control of data first party data for deprecation reasons idea for reasons as well becomes critically important and you know i would advise every client to look at their agreement with their media agency as to who controls the data.

26:59Because many media agencies control the data, and that's the way they exert control over the client, and they prevent change. In an AI world, at the very least, everybody should be experimenting now. You should break off a brand, preferably give it to us to work on. You break off a brand and start to experiment with the productivity around copywriting and visualization around hyper-personalization. You've got nothing to lose and you should do it quickly. You don't have to turn the whole organization upside down by doing it, but break off a bit and do it. So I think those are the skills. So the data skill is going to be absolutely critical because the real, the engine oil, the oil that lubricates all this is going to be the personalized data.

27:50The first-person data approved by consumers, no doubt. We'll get into that. In our pre-prep meeting, we talked a little bit about that issue. But I think that's the core skill. And then it's going to be about understanding. Well, agility is really important. So managing an organization extremely flexibly and responsibly. I mean, every conversation I have with somebody in the marketing fashion doesn't have to be the CMO or the CSO or the CIO complains about the fact that they don't move fast enough, they don't have enough control, and that the first-party data assets are very difficult to bring together.

28:31So that latter point is going to become absolutely critical. Oh, makes so much sense. Hugely, hugely important. Well, anyway, I thought I'd experiment with AI, obviously, with this conversation, so I asked it a few questions. Which one did you ask? ChatGPT, actually. Yeah, ChatGPT. So I said, give me some questions for Sir Martin Sorrell. The first one I thought is... They said who? Who? Exactly. Remind me. Anyway, first one. What's the best question for Sir Martin Sorrell? Were you to interview you, right? Now, I think this is a bit obvious, and you may have already answered this, so they get more interesting as they go along.

29:05The biggest disruption to our industry in the next 10 years... So I'll tell you what, I'll put a twist on it. You can't say AI. So beyond AI, biggest disruption to our business. There are two. I wouldn't say just one. One is geographical, so fragmentation. Different power bases, you know, the rise of Latin America, the rise of the Middle East, the rise of India, and in Asian context, not just India, Vietnam and Indonesia and Thailand and Philippines and Malaysia and Singapore. That doesn't mean old Asia, meaning Japan and Australia, New Zealand are not going to continue to be important, but the big question in Asia is China.

29:44Do you extend your business in China? If you have a big share, and I define big as, you know, China is 18 trillion out of 100 trillion GDP, so logically it should be around 20%. If you have 20%, maybe you stick because you're worried about Taiwan, supply chains, you know, Apple and Tesla going to Mexico, et cetera, for supply chains. If you're, you know, like, I think Unilever has 7 % of its revenue in China. You know, Procter went into China and Unilever went to India, so you want to up the ante in China. So we have a small business, a very good business. As my mother said, good things come in little packages.

30:24Our Chinese business is very strong, but we have to be bigger. And Chinese brands have been very expansive at the minute. And then outbound Chinese, what people don't understand is the platforms, in many cases, so for Alphabet, for Amazon, for Meta, their second big profit unit is outbound Chinese. So these are Chinese firms that are building their businesses internationally. So geographical fragmentation and the outlook for Europe. I hate to be depressing in France, but here goes. I think France, Germany, Italy, Spain, and the UK are going to have... Talking to clients, it tends to be... It's difficult to get growth in those markets, so we're very focused on efficient cost.

31:05So one thing is changing... So it used to be globalization. You can say it's deglobalization if you want, but it's certainly fragmentation. So picking your battlegrounds, before it didn't matter where you were, as long as the demographics were fine, free trade, always operated to help you. Now there are barriers. Make America great again, nearshoring, onshoring, whatever, makes it much more difficult. That's one thing. The second thing, and AI comes into this, is because GDP growth is going to be lower than it was and interest rates and inflation higher than they were and interest rates will be higher for longer, I think clients are going to find growth very difficult and margins will come under pressure.

31:53And as a result, they're going to be looking for efficiencies. And those efficiencies will be driven by technology, not just AI, but blockchain and metaverse too. next question i asked chat gbt give me the most surprising question for sir martin now the most surprising question according to chat gbt was if you could go back in time what advice would you give yourself you know when you look back on things you if you spotted something you you never do you never this is not a new thought because bill gates has said this i think on many occasions You never jump in. I mean, you may spot a change, but you never execute as quickly or as expansively as you should do.