WEBVTT - Smart Talks with IBM: Brewing Smarter: How HEINEKEN Is Using AI To Revolutionize Its Global Operations 0:00:00.160 --> 0:00:02.960 Hey everyone, it's Robert and Joe here. Today we've got 0:00:02.960 --> 0:00:05.240 something a little bit different to share with you. It's 0:00:05.280 --> 0:00:09.400 a new season of the Smart Talks with IBM podcast series. 0:00:09.720 --> 0:00:13.520 This season on Smart Talks with IBM, Malcolm Gladwell is back, 0:00:13.520 --> 0:00:15.760 and this time he's taking the show on the road. 0:00:15.880 --> 0:00:19.160 Malcolm is stepping outside the studio to explore how IBM 0:00:19.239 --> 0:00:22.880 clients are using artificial intelligence to solve real world challenges 0:00:23.079 --> 0:00:25.080 and transform the way they do business. 0:00:25.480 --> 0:00:29.560 From accelerating scientific breakthroughs to reimagining education. It's a fresh 0:00:29.600 --> 0:00:33.040 look at innovation in action, where big ideas meet cutting 0:00:33.120 --> 0:00:33.960 edge solutions. 0:00:34.320 --> 0:00:37.160 You'll hear from industry leaders, creative thinkers, and of course 0:00:37.280 --> 0:00:40.920 Malcolm Gladwell himself as he guides you through each story. 0:00:41.440 --> 0:00:44.680 New episodes of Smart Talks with IBM drop every month 0:00:44.680 --> 0:00:48.279 on the iHeartRadio app, Apple Podcasts, or wherever you get 0:00:48.320 --> 0:00:52.240 your podcasts. Learn more at IBM dot com slash smart Talks. 0:00:52.880 --> 0:00:55.360 This is a paid advertisement from IBM. 0:00:57.640 --> 0:01:01.000 H'm Malcolm Gladwell. Welcome to Season seven of Smart Talks 0:01:01.000 --> 0:01:04.680 with IBM. This year, we're exploring new stories about how 0:01:04.720 --> 0:01:08.480 companies are using the latest advancements in AI and quantum 0:01:08.520 --> 0:01:12.520 computing to create smarter business. For the first episode of 0:01:12.560 --> 0:01:16.160 the season, I flew to Austin, Texas to join Sergei 0:01:16.280 --> 0:01:20.800 Ghosh on stage at south By Southwest. Sergei is Chief 0:01:20.840 --> 0:01:25.600 AI Officer at Heineken, the world's pioneering beer company. Founded 0:01:25.640 --> 0:01:28.880 in eighteen sixty four. Heineken is deep roots, but it 0:01:28.920 --> 0:01:33.319 continues to push the boundaries of innovation today. In twenty twenty, 0:01:33.319 --> 0:01:35.760 the company came up with a goal to become the 0:01:35.800 --> 0:01:39.640 world's best connected brewer. Surge plays a key role in 0:01:39.720 --> 0:01:42.720 leading that transformation, and I sat down with him in 0:01:42.760 --> 0:01:45.640 front of a live audience to understand what that journey 0:01:45.640 --> 0:01:48.760 looks like and what it takes to reinvent a global 0:01:48.760 --> 0:01:55.160 company from the inside out. And before we get to 0:01:55.280 --> 0:01:57.960 the question of what you do in your job. So 0:01:58.120 --> 0:02:00.440 I'm really interested in people who have jobs it didn't 0:02:00.440 --> 0:02:03.160 exist for most of their life, and I'm curious how 0:02:03.200 --> 0:02:03.760 you got there. 0:02:03.960 --> 0:02:06.360 Yeah, first of all, thanks for having me here. Yeah, 0:02:06.640 --> 0:02:09.959 actually it did exist, and people sometimes don't realize AI 0:02:10.080 --> 0:02:13.640 is not new. It's been there for seventy five years, 0:02:13.680 --> 0:02:14.680 since nineteen fifty. 0:02:15.680 --> 0:02:16.919 It just changed over time. 0:02:16.960 --> 0:02:19.720 How the application is happening, right, So one thing to 0:02:19.800 --> 0:02:22.240 keep up with is as AI became more popular and 0:02:22.280 --> 0:02:25.560 more embedded in business, how do we upscal ourselves to 0:02:25.600 --> 0:02:28.960 stay at par with the technology trends. So the preparation 0:02:29.080 --> 0:02:31.280 for me personally started actually a long time ago, so 0:02:31.280 --> 0:02:33.359 when I was in grad school in US, and I 0:02:33.400 --> 0:02:34.800 also to live in US by the way, for a 0:02:34.800 --> 0:02:35.280 long time. 0:02:35.560 --> 0:02:36.400 You're Indian. 0:02:37.040 --> 0:02:39.200 I'm an Indian originally, but it's news. I did my 0:02:39.200 --> 0:02:42.800 grad school here and there. Actually I started taking courses 0:02:42.800 --> 0:02:46.160 in newer networks and at fission diagents back in two 0:02:46.200 --> 0:02:48.880 thousand and two, and it wasn't popular back then. 0:02:49.200 --> 0:02:51.040 I was just curious, what is it? Maybe it's the 0:02:51.080 --> 0:02:51.720 next big thing. 0:02:52.200 --> 0:02:54.560 And I'm so glad I did that because that sort 0:02:54.600 --> 0:02:55.960 of helped me build that foundation. 0:02:56.400 --> 0:02:59.519 What was it you said you were? You were curious 0:02:59.520 --> 0:03:02.160 about it? You're curious about it? Why what caught your 0:03:02.200 --> 0:03:03.200 eye about it? 0:03:03.240 --> 0:03:06.760 Was very different because the main difference was before that 0:03:06.840 --> 0:03:08.280 I was an engineer by professions. 0:03:08.480 --> 0:03:09.760 I went to engineering college. 0:03:10.200 --> 0:03:12.919 Everything is rule based, Everything is based on a formula, 0:03:13.000 --> 0:03:16.880 a physical equation. AI is something different because based on 0:03:17.000 --> 0:03:19.920 data and statistics, it never gives you a clear answer. 0:03:20.000 --> 0:03:22.440 It gives you a probability, and I just thought this 0:03:22.480 --> 0:03:25.639 is very interesting because if you're trying to solve a problem, 0:03:25.919 --> 0:03:27.920 you don't know exactly how to solve it. There is 0:03:27.960 --> 0:03:30.720 no equation. How do you get around that? I think 0:03:30.720 --> 0:03:33.760 that's where AI comes in. It finds those patterns within 0:03:33.880 --> 0:03:37.360 data and comes up with some prediction that intrigued me. 0:03:37.800 --> 0:03:41.800 So this is what year that you start I started dabbling, 0:03:42.040 --> 0:03:44.080 let's call it dabbling in AI was two thousand and two. 0:03:44.400 --> 0:03:45.960 It was almost twenty four years ago. 0:03:45.800 --> 0:03:49.120 Four years so put what you were playing with in 0:03:49.160 --> 0:03:52.960 two thousand and two was an extremely primitive version of what. 0:03:52.920 --> 0:03:53.560 We have now. 0:03:53.920 --> 0:03:56.360 I think it was very relevant because the way I 0:03:56.440 --> 0:04:00.080 see it, should I have skipped all the foundations that 0:04:00.120 --> 0:04:02.040 I learned over the years and just gone to the 0:04:02.040 --> 0:04:05.880 current state. Maybe, But when I look back, I think 0:04:05.880 --> 0:04:09.440 that foundation really helped me because back then, and surprisingly, 0:04:09.480 --> 0:04:11.560 by the way, new on networks. When I talk about that, 0:04:12.040 --> 0:04:15.360 it's still very ralliant and relevant within AI. The entire 0:04:15.400 --> 0:04:18.719 foundation is new on networks. So I think that foundation 0:04:18.839 --> 0:04:21.080 really helped. Yeah, and I still find it very relevant 0:04:21.120 --> 0:04:22.120 and I apply it day to day. 0:04:22.400 --> 0:04:26.120 Yeah, imagine in having a conversation with you twenty years 0:04:26.120 --> 0:04:29.280 ago and I say what are you up to? And 0:04:29.320 --> 0:04:33.599 you say, I'm playing with this thing neural networks. Early version, 0:04:33.640 --> 0:04:35.680 would you have used the term artificial intelligence? 0:04:37.080 --> 0:04:37.599 Probably not. 0:04:37.800 --> 0:04:40.520 I probably would have used something is called statistics, which 0:04:40.560 --> 0:04:43.600 is everyone is aware of. Back then, it was more statistical, 0:04:43.680 --> 0:04:46.520 so you don't have these big algorithms at that point. 0:04:47.400 --> 0:04:49.200 But then something happened. I don't know if you heard 0:04:49.200 --> 0:04:52.000 of this company called Cago. They used to host these 0:04:52.000 --> 0:04:55.320 sort of data science competitions and anyone can participate and 0:04:55.360 --> 0:04:58.039 if you do really well, you get a price. That 0:04:58.200 --> 0:05:00.599 was a good motivation, just to see, okay, learned something, 0:05:01.080 --> 0:05:02.680 let me apply it and see how good I am 0:05:02.880 --> 0:05:04.880 I'm getting at it. So I think that was my 0:05:04.960 --> 0:05:07.440 first entry point where I really got hands on into AI, 0:05:08.120 --> 0:05:10.240 and that probably stayed with me for a while. I 0:05:10.240 --> 0:05:12.800 think that was back in two thousand and six. That's 0:05:12.800 --> 0:05:15.120 what I started getting hands on. And the funny thing 0:05:15.200 --> 0:05:17.600 is when you look at these Caggle competitions, the use 0:05:17.640 --> 0:05:21.240 cases they used to give actual industry applications, so you 0:05:21.240 --> 0:05:24.400 were really dealing with business problems applying AI to solve it, 0:05:25.000 --> 0:05:27.280 and then you know, wait a minute, a medical company 0:05:27.320 --> 0:05:29.719 is using it. A manufacturing company is using it, a 0:05:29.800 --> 0:05:32.480 banking is using it. And this is two thousand and six, 0:05:33.040 --> 0:05:36.080 so it already started and then it just yeah today, 0:05:36.120 --> 0:05:38.040 it's a different ballgame now. 0:05:38.080 --> 0:05:40.840 So you you came to Heineken. 0:05:40.920 --> 0:05:45.560 When twenty twenty, twenty twenty right middle of COVID, right. 0:05:46.120 --> 0:05:48.880 And where you brought in to be the chief AI officer? 0:05:49.040 --> 0:05:51.600 Was you explore it and the title changed? But yes 0:05:51.640 --> 0:05:52.520 I was the global leader? 0:05:52.600 --> 0:05:55.520 Yeah yeah, And what made you want to take the job? 0:05:56.360 --> 0:05:59.159 I was actually working for Amazon at that point, but 0:05:59.200 --> 0:06:01.200 when I looked at High and I thought, okay, this 0:06:01.360 --> 0:06:05.479 is a legacy, traditional company, right, and AI was not 0:06:05.720 --> 0:06:07.520 a capability embedded. 0:06:07.040 --> 0:06:07.880 At that point of time. 0:06:08.360 --> 0:06:10.719 So it's a great opportunity if I can start something 0:06:10.720 --> 0:06:15.240 from scratch, really build it across the entire valuation of Heineken. 0:06:15.760 --> 0:06:17.960 I mean, that's probably the best job anyone can even 0:06:18.000 --> 0:06:20.760 ask for. Yes, it's of course a lot of responsibility 0:06:21.320 --> 0:06:23.040 that you have to make sure that you really build 0:06:23.080 --> 0:06:26.000 the right products and right capability, But that also happened, 0:06:26.480 --> 0:06:28.599 so I look back, it's quite fulfilling. 0:06:29.200 --> 0:06:32.040 But it's also if I might playable was advocate for 0:06:32.120 --> 0:06:34.960 a moment, you're also taking a risk going into an 0:06:35.080 --> 0:06:38.800 established how long has Heineken been around one hundred and 0:06:38.800 --> 0:06:41.480 sixty two years to be specific, eighteen sixty four. It's 0:06:41.520 --> 0:06:43.960 a very different proposition walking into one hundred and sixty 0:06:44.040 --> 0:06:46.240 year old company and saying I want to bring the 0:06:46.279 --> 0:06:48.400 future to the way you operate. Then it is with 0:06:48.480 --> 0:06:49.440 the startup. 0:06:49.960 --> 0:06:51.800 That is true, but it's also a challenge. 0:06:51.839 --> 0:06:54.599 It's a good challenge, and also that Heineken is also 0:06:54.680 --> 0:06:57.080 looking externally. There are companies that are picking up speed 0:06:57.120 --> 0:07:00.600 and embedding and adopting AI, so should be all behind 0:07:00.760 --> 0:07:02.800 not really, so we also need to pick it up. 0:07:03.120 --> 0:07:05.680 So I thought it was a good challenge because the 0:07:05.800 --> 0:07:08.560 use cases were there, the opportunity I put sense the 0:07:08.600 --> 0:07:11.040 business really was having the appetite. 0:07:11.120 --> 0:07:12.160 Let's do something different. 0:07:12.760 --> 0:07:15.040 When we apply AI in a corporate setting like this, 0:07:15.480 --> 0:07:18.080 it's super important to understand how the business actually works. 0:07:18.400 --> 0:07:20.680 What's the value chain looking like, what are the nuances? 0:07:20.680 --> 0:07:22.760 Where can I And once you get an understanding. It 0:07:22.800 --> 0:07:24.720 took me some time, by the way, to understand the 0:07:24.760 --> 0:07:27.760 full business and the complexities, but once you cross that 0:07:27.800 --> 0:07:31.040 threshold you figure out what's feasible what's. 0:07:30.840 --> 0:07:32.960 Not, then it opens up. 0:07:33.080 --> 0:07:35.440 Wait a minute, within the valuetion, I see ten areas 0:07:35.600 --> 0:07:36.520 I can optimize. 0:07:36.600 --> 0:07:38.400 You say, once you want to see in the business, 0:07:38.400 --> 0:07:40.920 describe the business. What the Heineken puzzle? 0:07:41.160 --> 0:07:43.920 So well, puzzle, Let's see if it's puzzle. 0:07:43.960 --> 0:07:46.720 After I explained it's actually you start with the procurement 0:07:46.760 --> 0:07:49.520 where you get the glasses, cans and all the raw materials. 0:07:49.880 --> 0:07:52.000 Then it comes to the brewery where the magic happens. 0:07:52.000 --> 0:07:54.920 That's what behinneqn beer is produced. Then it goes to 0:07:55.000 --> 0:07:58.560 the distributors. Basically a supply chain takes over. Then it 0:07:58.600 --> 0:08:01.280 goes to the customers. What we refer to as customers 0:08:01.320 --> 0:08:06.000 are the bars, restaurants, retail stores, moment pop stores, convenience stores, 0:08:06.560 --> 0:08:10.080 and that's where actually consumers then come and actually consume 0:08:10.120 --> 0:08:10.560 the product. 0:08:11.000 --> 0:08:12.320 So that's actually the value chain. 0:08:12.400 --> 0:08:14.720 It's actually pretty linear when you think of it, but 0:08:14.800 --> 0:08:17.760 there are nuances depending on the country and the market. 0:08:18.160 --> 0:08:20.120 There are some specific rules and guard rails that you 0:08:20.160 --> 0:08:20.760 have to be aware of. 0:08:21.080 --> 0:08:24.400 So you have that process going on all around the 0:08:24.400 --> 0:08:27.600 world and across multiple brands. 0:08:27.480 --> 0:08:30.560 Multiple brands, multiple countries, multiple operating companies. 0:08:30.680 --> 0:08:34.440 Yeah, from your perspective as someone who is the chief 0:08:34.480 --> 0:08:37.440 AI officer, what are the tasks in front of you? 0:08:37.880 --> 0:08:38.840 What's your opportunity? 0:08:38.880 --> 0:08:42.760 There any process that you think that is maybe not 0:08:43.040 --> 0:08:46.400 digitized or maybe not data driven, you can optimize. I 0:08:46.440 --> 0:08:49.480 look at it like a pendulum. So one side of 0:08:49.520 --> 0:08:52.880 the pendulum you have complete gut based decision making. The 0:08:52.920 --> 0:08:55.920 other extreme is completely data driven. So the idea is, 0:08:56.040 --> 0:08:58.280 can we swing this pendulum it will be towards data 0:08:58.360 --> 0:08:59.319 driven from where we are. 0:09:00.200 --> 0:09:03.360 Give me a specific example of a problem you set 0:09:03.400 --> 0:09:05.520 out solve or address. 0:09:06.559 --> 0:09:08.080 There are quite a few, but if I want to 0:09:08.559 --> 0:09:11.560 pick one of the most fun one fun one may be. 0:09:11.440 --> 0:09:14.240 The most most complex one. Let's bring that one. I think. 0:09:14.240 --> 0:09:17.320 So we spend quite a bit on advertising, and Heineken 0:09:17.559 --> 0:09:20.560 is a largely a lot of it marketing company, and 0:09:20.640 --> 0:09:22.600 we really care about our brands. 0:09:22.280 --> 0:09:25.400 And products, you know, we're almost obsessed with it. Let's 0:09:25.440 --> 0:09:26.000 take an example. 0:09:26.080 --> 0:09:28.600 Let's say you have X million dollars as your budget 0:09:29.200 --> 0:09:32.280 and you have two brands, let's say Heineken and Tosa Kiss. 0:09:32.520 --> 0:09:34.760 I think the crowd audienceale will be familiar with that. 0:09:35.600 --> 0:09:38.959 And then you have three touch points TV, YouTube, Instagram, 0:09:39.360 --> 0:09:43.840 and I want to optimize my advertising budget between different 0:09:43.880 --> 0:09:46.960 brand and touch points, so heiniken on Instagram. 0:09:46.960 --> 0:09:49.640 How much should I spend? It's a very easy question to. 0:09:49.600 --> 0:09:52.320 Ask, but to actually solve this you have to study 0:09:52.440 --> 0:09:55.600 historically how these performed and then create a model and 0:09:55.640 --> 0:09:59.080 then predict if I allocate my budget in this way 0:09:59.559 --> 0:10:02.440 that's probab more optimal. Before it was more like somebody 0:10:02.440 --> 0:10:05.319 to con gun based decisions saying okay, here goes xpillion, 0:10:05.360 --> 0:10:07.560 here goes five million, and we say no, no, no, that's 0:10:07.559 --> 0:10:08.520 not the right proportion. 0:10:08.800 --> 0:10:11.959 What did the AI tell you about the accuracy of 0:10:12.000 --> 0:10:14.240 those spending decisions? 0:10:14.679 --> 0:10:17.319 In the past, we looked at the return on investment 0:10:17.360 --> 0:10:20.200 from this advertising, so how much incremental volume or volume 0:10:20.240 --> 0:10:23.520 of brre reselling or revenue are recreating and we find 0:10:23.520 --> 0:10:26.640 out can we improve that. It's a moment to apply AI. 0:10:27.200 --> 0:10:29.440 And when we look at that this significant improvement. In 0:10:29.440 --> 0:10:33.439 some cases we have thirty percent uplift, thirty percent, thirty percent, 0:10:33.480 --> 0:10:37.000 three zero not everywhere. In some places we got but 0:10:37.040 --> 0:10:39.439 it ranges between ten to thirty percent uplift depending on 0:10:39.480 --> 0:10:41.520 the type of AI product you're building. 0:10:41.559 --> 0:10:43.079 And that impacts the top line. 0:10:43.440 --> 0:10:45.960 So it's very easy to also realize that value. 0:10:46.000 --> 0:10:47.000 People get to see. 0:10:46.760 --> 0:10:49.480 It, so you say, oh, we we can do a 0:10:49.480 --> 0:10:51.640 way better job if we spend x more or x 0:10:51.760 --> 0:10:54.440 less in this particular area. Give me another example of 0:10:54.480 --> 0:10:54.920 a if. 0:10:54.920 --> 0:10:57.880 Another one would be we have a very big large 0:10:57.920 --> 0:11:01.679 salesforce within Ainiken. So this sales reap. What they do 0:11:01.720 --> 0:11:04.480 they go to the outlets, the bars and restaurants, and 0:11:04.559 --> 0:11:08.040 they maintain that human to human relationships with these our 0:11:08.080 --> 0:11:11.240 customers is super important to maintain that, and they go 0:11:11.360 --> 0:11:13.959 solve the customer problems. Let's say someone is out of stock, 0:11:14.040 --> 0:11:17.320 someone is about to churn, or their surprise mismatch, something 0:11:17.400 --> 0:11:19.720 like this, and before they us to go like this, 0:11:19.880 --> 0:11:22.120 let's say a sales rep on a day to day 0:11:22.200 --> 0:11:25.640 job has to visit five places ab CD E, five 0:11:25.640 --> 0:11:29.760 different outlets, and he used to go ABCDE. Turns out, 0:11:29.920 --> 0:11:32.679 the model tells you on any given day, if you 0:11:32.760 --> 0:11:37.120 optimize taking into account the traffic conditions, instead of going 0:11:37.160 --> 0:11:40.120 from that linear route, you go to D first and 0:11:40.160 --> 0:11:42.000 then to B, then to C, then to E and 0:11:42.040 --> 0:11:44.240 then to A. And the reason for doing that is 0:11:44.600 --> 0:11:47.439 the model tells you, if you visit customer D first, 0:11:47.440 --> 0:11:49.760 he has the biggest problem that needs the most amount 0:11:49.800 --> 0:11:52.600 of time to be solved, and that's how it optimize. 0:11:52.640 --> 0:11:54.280 And also the sales reps now. 0:11:54.120 --> 0:11:57.600 They are becoming so educated with some of these AI models, 0:11:57.840 --> 0:12:00.800 they are now becoming a business advisors. So they are 0:12:01.080 --> 0:12:04.280 no longer just solving little problems. They are having the 0:12:04.360 --> 0:12:06.400 time to say what else can I do for you 0:12:06.960 --> 0:12:09.400 as the customer? So that I think it was a 0:12:09.400 --> 0:12:12.000 big change within Heineken because it impacted a lot of 0:12:12.000 --> 0:12:13.679 people that were using that. 0:12:13.880 --> 0:12:16.560 In an instance, it requires not just building a model 0:12:17.040 --> 0:12:19.600 that can be smarter about how people should spend their 0:12:19.600 --> 0:12:21.720 time and what they should say, but you have to 0:12:21.880 --> 0:12:27.280 obviously educate your salesforce to believe in what the selling 0:12:27.840 --> 0:12:29.840 tell me about that piece. Is that a piece that 0:12:29.520 --> 0:12:31.360 you that you're a part of or is someone else? 0:12:31.480 --> 0:12:33.959 Yeah, that's also part of because that is super important. 0:12:33.960 --> 0:12:36.680 I think we can build the best models. Best algorithm's 0:12:36.760 --> 0:12:39.800 highest accuracy doesn't mean anything if it's not used the 0:12:39.840 --> 0:12:42.199 right way. So what we do we have within our 0:12:42.240 --> 0:12:46.440 company a pretty big upskilling program. So bring everyone along 0:12:47.080 --> 0:12:50.640 in common understanding, basic understanding of what AI does. Not 0:12:50.720 --> 0:12:54.640 everyone needs to understand new networks or algorithms right, But 0:12:54.760 --> 0:12:58.320 what we do is give them a handheld device and 0:12:58.360 --> 0:13:00.839 an app which is driven by AI. Play with it, 0:13:01.200 --> 0:13:03.760 have fun, see how it changes your life. And once 0:13:03.800 --> 0:13:07.160 you start liking the product, liking the UIUX, then you 0:13:07.200 --> 0:13:09.680 start getting more. And the insights also tell you the 0:13:09.679 --> 0:13:12.319 story because once you start getting the value, I am 0:13:12.360 --> 0:13:16.840 not having to pitch my models anymore. The sales reps 0:13:17.080 --> 0:13:19.439 and the markets they are pitching on behalf of us. 0:13:19.960 --> 0:13:21.760 And that's such a good place to. 0:13:21.720 --> 0:13:26.040 Be, is it. It's interesting? So in that instance where 0:13:26.040 --> 0:13:29.760 you're designing a more efficient form of interaction and fruitful 0:13:29.840 --> 0:13:33.000 for of interaction between sales reps and customers, I could 0:13:33.040 --> 0:13:35.920 see a version of that where it is really clear 0:13:36.559 --> 0:13:40.480 looking up from a high level that things are working better, 0:13:40.720 --> 0:13:43.680 but it might not be clear to the salesperson. Is 0:13:43.720 --> 0:13:46.680 the salesperson who's now following the direction of the AI 0:13:47.200 --> 0:13:48.800 aware that they are more efficient? 0:13:48.920 --> 0:13:50.359 They are? 0:13:50.559 --> 0:13:51.600 How are they aware the market? 0:13:51.600 --> 0:13:54.360 They realize few things that they were visiting customers just 0:13:54.400 --> 0:13:57.040 because they had to visit because it was in the schedule. 0:13:57.520 --> 0:13:59.959 Now they go there and they find out, wait a minute, 0:14:00.200 --> 0:14:03.160 I never tackled this big problem that was not being addressed, 0:14:03.600 --> 0:14:07.120 and they solved it. And the customer feedback also comes 0:14:07.160 --> 0:14:09.920 back saying we are really happy. So for all these products, 0:14:10.000 --> 0:14:12.000 we get the feedback not just from the sales reps, 0:14:12.160 --> 0:14:14.760 but also for the customers. Do you really like the 0:14:14.840 --> 0:14:17.920 recommendations we are giving you? And that's the best validation 0:14:17.960 --> 0:14:20.440 you can think of, because it's four stand or feedback. 0:14:20.880 --> 0:14:23.800 When the AI is doing this ranking, it wants you 0:14:23.840 --> 0:14:27.920 to focus on the customer with the biggest problem first 0:14:28.000 --> 0:14:29.240 or is it much more complex than that. 0:14:29.320 --> 0:14:30.680 It's a little bit more complex than that. 0:14:30.760 --> 0:14:33.400 Yeah, but usually it's a rank pordert in terms of 0:14:33.400 --> 0:14:35.960 which one is the biggest problem that needs the most 0:14:35.960 --> 0:14:36.640 amount of time. 0:14:37.200 --> 0:14:38.160 That's how it's ranked order. 0:14:38.240 --> 0:14:40.440 But sometimes you can also override the model, right, We 0:14:40.520 --> 0:14:43.720 also give options to people. You don't have to all 0:14:43.800 --> 0:14:46.080 the time one hundred percent follow the recommendation. If you 0:14:46.080 --> 0:14:48.920 have some urgent priority, you can overwrite that. 0:14:48.920 --> 0:14:51.800 That's also possible with something like that. Is there a 0:14:51.920 --> 0:14:55.200 next level you can go to? So you design this 0:14:55.280 --> 0:14:56.920 system and you say, oh, I can make our sales 0:14:56.920 --> 0:14:59.480 staff a lot more effective and the way they operate 0:14:59.560 --> 0:15:03.960 with their customers, and then you see that it works, 0:15:04.320 --> 0:15:06.000 and then it comes back and then you say, okay, 0:15:06.000 --> 0:15:09.240 what's two point oh? Is there a two point oh. 0:15:09.040 --> 0:15:09.480 It could be. 0:15:09.600 --> 0:15:13.120 So it's always about innovation. Then you think, okay, today 0:15:13.160 --> 0:15:15.360 we go and solve the problems that have already happened. 0:15:16.400 --> 0:15:19.080 What if we solve the problems that are likely to happen, 0:15:19.680 --> 0:15:22.440 that will be the next step. So this customer hasn't 0:15:22.480 --> 0:15:25.160 been very active for a while, there's a high chance 0:15:25.200 --> 0:15:28.240 that that customer might churn out of Heineken. So what 0:15:28.320 --> 0:15:31.640 actions can I actually recommend to make sure And we 0:15:31.720 --> 0:15:33.640 do this, by the way, we also gather a lot 0:15:33.680 --> 0:15:37.040 of customer feedback and complaints and feedback, and we use 0:15:37.160 --> 0:15:39.600 LLLMS to extract and glean information. Okay, what are the 0:15:39.600 --> 0:15:42.080 real pain points? What's the theme and the topic that 0:15:42.200 --> 0:15:44.760 needs to be addressed. And once you do that, then 0:15:44.800 --> 0:15:47.520 also you can prepare ahead of time. We're already there, 0:15:47.560 --> 0:15:49.120 by the way, when I say two point zero, we're 0:15:49.160 --> 0:15:52.600 already testing it. You solve problems or you'll try to 0:15:52.600 --> 0:15:55.440 solve problems before daven occur. So I think that's a 0:15:55.480 --> 0:15:57.280 little bit of a two point zero. Then we have 0:15:57.320 --> 0:15:58.720 to see what else we can do with it. 0:15:59.240 --> 0:16:03.600 Tell me you've had this partnership at Heineken with IBM 0:16:04.240 --> 0:16:08.480 for since twenty thirteen, twenty thirteen, So you came in 0:16:08.600 --> 0:16:12.200 and there was already a strong working relationship. Tell me 0:16:12.240 --> 0:16:14.760 about how that relationship started and what does it mean 0:16:14.800 --> 0:16:18.920 on a practical basis. You're building all these tools, how 0:16:18.920 --> 0:16:20.400 does the interaction with IBM work. 0:16:20.920 --> 0:16:22.640 Yeah, so I think good to give a little bit 0:16:22.640 --> 0:16:26.760 of context. Heineken started this digital transformation journey in twenty 0:16:26.840 --> 0:16:30.040 twenty formally, but the tech was already there. We had 0:16:30.080 --> 0:16:34.120 our systems, platforms, data, everything was there. So all the 0:16:34.200 --> 0:16:37.600 IT four ID systems is where IBM was partnering with 0:16:37.680 --> 0:16:40.280 us from the get go, from twenty thirteen, and it's 0:16:40.320 --> 0:16:43.080 a very long standing partnership because as we found the 0:16:43.160 --> 0:16:46.560 tech is evolving, our partnership also kept evolving because we 0:16:46.600 --> 0:16:49.600 need to keep up up to the speed. So it 0:16:49.720 --> 0:16:53.520 was more about IT for IT systems, cybersecurity platform, data, 0:16:54.280 --> 0:16:56.600 incident management, service level, you name it. 0:16:57.000 --> 0:16:57.520 All of that. 0:16:57.680 --> 0:17:00.600 IBM was supporting us both in terms of hands and 0:17:00.680 --> 0:17:01.560 also in terms. 0:17:01.360 --> 0:17:02.760 Of strategy to create together. 0:17:03.360 --> 0:17:05.800 But that also evolved, like I said, when we went 0:17:05.840 --> 0:17:09.600 into this digital transformation journey in twenty twenty, then we 0:17:09.680 --> 0:17:13.960 started building this digital core, which is the central nervous 0:17:13.960 --> 0:17:17.680 system software system of finikin that's where IBM is really 0:17:17.720 --> 0:17:20.600 partnering with us and helping us not just shape the 0:17:20.640 --> 0:17:22.560 whole thing in terms of building it hands on, but 0:17:22.600 --> 0:17:25.080 how do we strategize that so that it lands well. 0:17:25.160 --> 0:17:28.720 So that's yeah, it's a long trusted partnership. I think 0:17:28.720 --> 0:17:30.200 we are going to go a long way together. 0:17:30.560 --> 0:17:33.520 Yeah, Heineken Space just sells out of Amsterdam. 0:17:34.160 --> 0:17:36.480 The head office is absolutely so the IBN. 0:17:36.160 --> 0:17:38.359 People who work with you are they on site? 0:17:38.640 --> 0:17:40.720 There are some on site and there are some teams 0:17:40.720 --> 0:17:43.680 in India some times spread across the globe. But for 0:17:43.680 --> 0:17:46.000 for tech, I think the location doesn't matter. But you 0:17:46.040 --> 0:17:48.080 still need people on site to actually talk with the 0:17:48.119 --> 0:17:50.960 business and really understand what the problem for. 0:17:51.080 --> 0:17:52.720 Those interactions are also very important. 0:17:53.040 --> 0:17:54.639 When you said you wanted to when you got there, 0:17:54.640 --> 0:17:57.560 you wanted to build a digital nervous system, what does 0:17:57.560 --> 0:17:58.000 that mean? 0:17:58.440 --> 0:17:59.720 Maybe good to give you an example. 0:18:00.200 --> 0:18:04.919 Let's say iPhone, right, it's a central platform, but you 0:18:04.960 --> 0:18:08.440 can download thousands of apps there and all of them 0:18:08.800 --> 0:18:11.600 once you download seamlessly integrates with the system and you 0:18:11.600 --> 0:18:14.439 don't see any difference. This is the same thing. So 0:18:14.480 --> 0:18:16.639 what we want to build within Heineken is a central 0:18:16.680 --> 0:18:19.159 software system which the old school way of saying it 0:18:19.200 --> 0:18:22.879 is the ERP enterprise resource planning. It removes the fragmentation 0:18:22.920 --> 0:18:25.760 of different platforms, it brings it all together. It makes 0:18:25.760 --> 0:18:29.240 sure all the business applications within supply chain, commerce, finance 0:18:29.440 --> 0:18:33.600 HR all in one place and coordinates them everything orchestrates them. 0:18:34.080 --> 0:18:37.040 The benefit of doing that is two one across the 0:18:37.119 --> 0:18:39.919 value channing of one way of doing business because everything 0:18:39.960 --> 0:18:44.560 is standardized, but we also have multiple markets globally. Across 0:18:44.600 --> 0:18:47.440 the multiple markets also it becomes one way of doing business, 0:18:47.640 --> 0:18:50.840 so it's both ways. And once you standardize it, we 0:18:50.920 --> 0:18:55.360 can embed new apps which will seamlessly integrate, and then 0:18:55.400 --> 0:18:58.200 it just keeps scaling further. Can we scaled very quick 0:18:58.520 --> 0:19:01.359