← MarTech Podcast ™ // Marketing + Technology = Business Growth
MarTech Podcast ™ // Marketing + Technology = Business Growth
How can Marketing lead AI transformation
00:25:10
Loading summary
Transcript55 lines
- [00:00]
Podcast Announcer
The Martech Podcast is a proud member of the I Hear Everything Podcast Network. Looking to launch or scale your podcast, I Hear Everything delivers podcast production, growth and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit iheareverything.com.
- [00:26]
Benjamin Shapiro
From advertising to software.
- [00:28]
David Rabin
As a service to data across all of our programs and clients, we've seen.
- [00:33]
Benjamin Shapiro
A 55 to 65% open rate. Getting brands authentically integrated into content performs better than TV advertising. Typical life span of an article is about 24 to 36 hours. We're reaching out to the right person with the right message and a clear call to action.
- [00:51]
David Rabin
Then it's just a matter of timing.
- [00:54]
Podcast Announcer
Welcome to the Martech Podcast, a member of the I Hear Everything Podcast Network. In this podcast you'll hear the stories of world class marketers that use technology to drive business results and achieve career success. Here's the host of the Martech podcast, Benjamin Shapiro.
- [01:16]
Benjamin Shapiro
19% Only 19% of B2B marketing teams have integrated AI into their daily workflows. According to the Content Marketing Institute. The gap between AI hype and actual implementation is massive. Any idiot with a keyboard can use AI. So why are most marketers stuck in the AI experimentation mode, running isolated pilots and struggling to scale beyond basic use cases? The technology isn't the problem anymore. The real challenge is organizational readiness and strategy. So how do you move from AI experimentation to enterprise wide transformation? I'm Benjamin Shapiro and joining me today is David Rabin, the Chief Marketing Officer at Lenovo's Solutions and Services group which helps companies build right size AI solutions. And today David is going to share how you can overcome organizational barriers that are blocking your AI adoption. David, welcome to the Martech Podcast.
- [02:10]
David Rabin
Thank you for having me. That's a great intro. I hope I can follow through on what you just said. Let's see, let's give it a shot.
- [02:15]
Benjamin Shapiro
Hey, here we go. Let's start off with basic blocking and tackling here. Lenovo. I think of Lenovo as like the little black boxes, the computers with the little red mouse. And now you have a services and a solutions group. So how did Lenovo go from a hardware manufacturer to getting into services to being an AI led services organization?
- [02:36]
David Rabin
This has been a thoughtful journey for the company. As you said, we are known throughout the world for. You called it the little black box. We'll call it the ThinkPad, the world's best and best selling business laptop. But we've really gone beyond that and if we can be totally candid, hardware in our Industry is necessary, but in many cases, it's gotten commoditized and the company realized, great hardware. Awesome. How do we go deeper? How do we provide more value and solve more customer problems? So about a decade ago, we went out and bought the IBM server business and we bought the Motorola phone business from Google. Then about four years ago, we realized for us to truly solve customer problems delivering the outcomes they want, we got to get deeper into solutions and services. And so here we are. Today, we're a $69 billion company. Of course, we're number one in PCs around the world. One out of every four PC is a Lenovo product. But more and more customers know us for delivering real solutions. And of course, we're going to talk a lot about AI today. We're one of the pioneers in delivering AI solutions. So it's been a great journey.
- [03:35]
Benjamin Shapiro
A I knew that you guys had bought the hardware business from IBM. I didn't realize that Motorola was part of Lenovo as well. This great hardware infrastructure you're adding services on the deck gets reshuffled. Everybody's talking about AI. So tell me about what some of the enterprise struggles that you're seeing with AI adoption with some of the people that are working with Lenovo Services Group.
- [03:57]
David Rabin
When I started at the company, then about 19 years ago, we were about a $12 billion company. Today, almost 70 billion. And I take credit for about three quarters of. So I really feel like I've been instrumental to the company. But in all seriousness, we spent a lot of time thinking about the IT decision maker, what's happening in the walls of the enterprise. And of course, we also do a lot of consumer research as well. But itdms told us in a study earlier in this year, three big blockers. First, roi. How do you calculate ROI on a process or a tool or a program that you've never done before? So we got to help customers figure out how to get a fast path to roi. The second thing is the data. You cannot run a without organized data. And again, we deal with a lot of companies that are struggling on that. The third is the people. None of us were trained at university in being AI solution deployment experts. So we got to make sure that IT departments have the capabilities and the resources and the knowledge to deploy AI. Now put your marketing hat on. We're struggling with the same things, right? How do I deploy marketing solutions when I don't know what the ROI is going to be? When I don't have organized data and I don't have the people that have the skills to do it. So a lot of parallels in IT and marketing.
- [05:09]
Benjamin Shapiro
In my intro I mentioned any idiot with a keyboard can use artificial intelligence. The playing field has been leveled, so why is there such an implementation problem? Seems like the tech adoption should be easier. The barrier to entry for this advanced technology has been leveled, but yet people are really struggling to do anything other than like, Well, I had ChatGPT write a poorly worded email for me.
- [05:34]
David Rabin
Think about the requirements to build an AI solution. Okay, the easy one is you got to have the hardware. So maybe you're doing it on a laptop, maybe you're storing data on a server. That's maybe the easiest one. You're going to have to have the right model, you're going to have to have organized data, you're going to have to have software to power it all. And then in some cases you're going to need a solutions company to tie it all together. So the point being, yes, any idiot with a keyboard can go onto ChatGPT for sure. Any idiot with a keyboard can use an AI photo generation tool. But is it going to be secure? Is it going to tie into the rest of your martech stack? Is the company going to have the right governance to even let you do it? There's a lot of barriers and you know, again, we can talk throughout the course of this conversation. A lot of it is tops down. If you don't have a leadership team that's saying go, don't wait, go, you're going to be stuck in stall mode for a long time.
- [06:25]
Benjamin Shapiro
Yeah, it's something that is hard for me to wrap my head around. The enterprise grade barriers. Yeah, for a long time I worked at ebay, so I'm not a total novice when it comes to some of the enterprise restrictions, but I'm a guy sitting at home using cloud based, browser based solopreneur tools and I'm able to build relatively articulate, artificially intelligent LED workflows that are enabling me to scale, to be at least as powerful just myself as the marketing team that I ran when I ran a startup 10 years ago. So walk me through. You mentioned the governance, the pain points. It's like, why aren't larger organizations just saying, look, you guys have a million tools that you can use, go get them and let's figure it out.
- [07:10]
David Rabin
Yeah, but let's turn back the clock a decade. Why did Uber succeed? Why did Airbnb succeed? Why did Netflix succeed against giants? Right, David and Goliath? Why did Amazon succeed against the Barnes and Nobles of the world? Because they were nimble, they were flexible, they were able to cut corners. So, I mean, history has shown, unfortunately, that large enterprises generally act slower than more nimble startups. It's also one of the reasons why so many of the software companies that are delivering AI based solutions are companies that 12 months ago you never heard of. And it's one of the reasons why you're seeing companies like Microsoft and Amazon and Meta buying up these organizations because they're not as nimble as these smaller organizations. So again, your point, a solopreneur can do things faster than a large enterprise company. Now, you can't do it at scale necessarily. You're not going to have the backing of the large company. This is a pervasive issue in the corporate world today. And it's one of the reasons why you're seeing more organizations quickly putting governance programs in place. Not to slow things down, but to recognize we got to move fast. So get the governance in place, get your security checkpoints in place and just start moving. And it's taken us a while to get there.
- [08:15]
Benjamin Shapiro
Walk me through that, because I think of my wife's company. She works at a big box retailer in strategy. She was just in a meeting with the CEO and they asked her to both lead the meeting and take the notes at the same time. And I was like, why don't you just have your note taker on? Like, just get the transcript. She's like, how do you do that? I'm like, all right, if you guys aren't able to figure out how to use a note taker, I'm sure their company will be fine. But this seems like an enterprise scale problem where it's like, well, the privacy restrictions and the, oh my God, you can't use a note taker. It's like, it's such commonplace problems. How do enterprises defend some sense of turf if they can't use the basic technologies? And sorry to my wife's company.
- [08:54]
David Rabin
No, no, no, we don't need to apologize to them. I'm sure it's a great company. But think about the first thing I said. I mean, one of the things we talk about is roi. AI costs money. It is not free. So even though you think that when you're doing a ChatGPT query it's free, it is not. We use Copilot. We've got a great relationship with Microsoft. Copilot is not free. There is a cost incurred per seat, per month to use Copilot. So now organizations have to say, okay, let's take your wife's company. I've got 5,000 employees for every employee that I give a copilot license so that it enables me to take notes and get transcripts. Great. It's going to cost them money. So they got to figure out, is the ROI going to be worth it. Now, most companies will tell you they're not increasing their IT budgets. Why not? Because they don't know they need to. So what they're doing is saying, if you think this AI thing is so valuable, cut something else. Well, now as a decision maker, I have to cut something that's working. I have to cut something that's giving me return. So it's more than just governance, it's more than just security. There's a cost component of this as well. And we're facing it, right, because we're trying to sell AI solutions. We're trying to sell AI PCs with copilot licenses and we get conversation with customers who say, is it going to be worth me spending an extra $300 per user to do that?
- [10:06]
Benjamin Shapiro
Why does it cost so much more for an enterprise than it would for a smaller company to use the advanced technology?
- [10:12]
David Rabin
I don't know if it costs so much more, but again, we've got to put a lot of governance in place. We have to put additional security in place. You don't want your wife's company allowing a leak somewhere where it's going to affect you as a consumer. So we do have a bit more of a layer cake than you would as a consumer. Plus, as consumers, we're pre wired not to care. We're pre wired to see that little screen that says read this whole thing and check this box if you're willing to sell your entire life story. Companies don't want to do that.
- [10:36]
Benjamin Shapiro
All right? So there's these big companies that have to pay extra to be able to use the tools that for down market solutions. We don't have the same governance and administration and user management and security problems. And so they cost more. I understand, but it seems like everybody that I see that is an exec on LinkedIn. Every CMO is like, here's how we mastered AI. Everybody is waving the flag and claiming victory. How many of their teams are actually implementing AI and is that all just BS to get personal brands out there?
- [11:06]
David Rabin
Well, I don't want to give trade secrets of the CMO community. You know, we meet as a community every week over coffee and we talk about this stuff. No, I mean, look, the reality is we all know there's a lot more bluster than action happening right now. But I'd say the good news is that what we're starting to see is pockets of AI activity. So let me give you an example. We have been doing our best to partner much closer with our sales organization. And one of the tools that is not new to the industry, but that we've been deploying is more account based marketing. Gives us an opportunity to sit down with our sales leadership and talk through how do we do sort of individual account by account marketing with them. Now, historically, that's a very manual process. We've got to sit and identify a customer list, we've got to then kind of research out the buying groups, we've got to do firmographics and technographics and intense signals. It's a very, very manual process. So I can now automate that. We use Demandbase, it's one of our strategic partners. We use LinkedIn matched audiences. We're able to sort of automate a lot of that stuff, but there's still a lot of manual process involved. So it's way too soon to declare victory. We know what the goalpost looks like. We know that a truly automated process which would go from identifying the customer all the way to building a marketing program around them, delivering personalized marketing to them, measuring it, and then delivering an intense signal to a seller saying it's time to call this customer. That is the holy Grail. We're probably 28 to 32% of the way there right now. So yeah, on LinkedIn for sure we declare victory. Behind the scenes.
- [12:33]
Benjamin Shapiro
We're at the 30 yard line, 28 to 32%. Very scientific.
- [12:37]
David Rabin
Yeah, something like that.
- [12:38]
Benjamin Shapiro
I'm going to basically say that 72% of LinkedIn posts are pure bullshit.
- [12:43]
David Rabin
I would agree with that.
- [12:44]
Benjamin Shapiro
Talk to me a little bit about organizational behaviors for the 28 to 32% of the CMOs who are claiming AI victory that have actually had their teams implement some useful AI solutions. What are the behaviors that those teams are exhibiting that are allowing them to actually adopt AI and get an ROI from it?
- [13:03]
David Rabin
I'm going to get in trouble here, probably. So I'm kind of hopeful some of my team doesn't watch this, but here's sort of how we would look at this. And actually let me give you a real world example of something we built. About 18 months ago, we built a tool called Studio AI. It is a content development tool that's generative, AI based.
- [13:19]
Benjamin Shapiro
Internal tool.
- [13:20]
David Rabin
Internal tool, yep. And I can tell you all about the backend technology, about the models we use and about the software and about the data. But the change management piece has actually been the most difficult part of it. Sure, I'm making the IT part sound easy. I didn't have to do that. I handed that off to an IT team. They went and built it. But here's how change management goes. Let's say I need to make a print ad or I want to make a brochure. It doesn't matter what it is for a new product. The model today for a lot of companies is write a brief. So marketing manager writes a brief and then they send it somewhere. They send it to an internal agency, to a copywriter, to a designer, or in a lot of cases, you send it to an external agency partner. And then what do you do? Take a vacation for a week and you go out to the beach and coffee. Coffee, Read a book, you do your thing for a week, and then maybe a week later, the agency comes back, they give you something to look at. I kind of like it. But you need to fix a few things, go on another vacation for a week, you come back, and maybe it's 90% of the way there.
- [14:11]
Benjamin Shapiro
God, work was wonderful before. Go on.
- [14:13]
David Rabin
Yeah, exactly. Here's the deal with Studio AI. Grab that coffee cup again, Ben, because with Studio AI, you upload your brief to Studio AI, you push the button and say, make me a brochure, grab a cup of coffee, come back, it's done. It's done. It's sitting right there in the tool. Now, as the marketing manager, I'm the one responsible for. I don't really love the headline or why'd they choose this photo? Let me kind of make this a little bit more bold. Hey, let's bring in a customer case study. I'm going to reprompt the tool. It's going to regenerate it. Now. I'm going to do that all in 15 minutes or 20 minutes instead of the two weeks. But I've lost the safety net. It's all on me now. So my safety net of the agency screwed it up. The agency took too long. The agency went over budget or the designer or the technical writer. It's gone now. It's all on me, the marketer. So the change management is really kind of proving to be the tougher challenge than actually building the tool and seeing results. And that's the thing. I've got real time data showing it's working. It's still a change process to get people to kind of stop doing things the old way and the new way. And it actually is One of the reasons why for our IT customers, we actually sell an advisory and change management service to them because we know we're going to have to come in and help them, not just build the tool and deploy it. We're going to have to educate the organization on how does it work and why you should use it.
- [15:22]
Benjamin Shapiro
When you say change management, is it getting people used to the new workflows and adopting the tools and getting them to apply them, or is it more of a mind shift change where they have to understand the power of the tools and the capabilities that they have?
- [15:41]
David Rabin
Absolutely both. It's truly both. I would say the second one is probably the bigger piece, but here we go. This is exactly the world we're living in right now. Five years ago, you never wanted to talk to tools, and now it's all we want to do. You've got to convince people of the power of tool. Actually, we talk a little bit here about chatbots. Sometimes at some point when you started calling to return an item you bought at a clothing store, they offered you the opportunity to do it online with a chatbot. Walmart does a great job of this. And the first time that service is probably teed up, you're probably thinking, come on, I want to talk to somebody, right? Let's get this thing resolved. And then you realize, I can do this from the couch or from the toilet or from my bed. I can do it without talking to a human being. Let me use the technology. And guess what? It works pretty damn well. So a lot of this is just get somebody over the hurdle, force them to do it once they realize it's going to work, and then you kind of get the change management ball rolling pretty fast.
- [16:33]
Benjamin Shapiro
Yeah, I'll fight you on that one. I think most of the AI LED customer service is a terrible experience, mostly in E commerce, where I find myself the equivalent of when you used to have the landline and you're just yelling, talk to a representative like smash zero as much as you can. Whenever it starts with AI, I'm like, okay, I'm giving all my information. It always ends up being, hang on, let me go get a person to solve this problem. And I was like, ah, yeah.
- [16:59]
David Rabin
But here's what I'll tell you. And this is one of the reasons we focus on diversity of marketing and research so much. I'm going to guess, Ben, you're over 22 years old.
- [17:09]
Benjamin Shapiro
Yep. Just barely.
- [17:10]
David Rabin
Okay, talk to one generation below. Both of us, they don't want to talk to human anymore. So we're Also preparing kind of coaching people for what's coming next. But I really believe, and I will fight you back on the use of AI tools. We're coming to a point where it's going to be very difficult to get to a human because almost everything's going to go through an AI filter. And candidly, they're getting better on a daily basis. We use a lot of AI for service desk capabilities and trust me, the amount of calls we can reduce, friction we reduce in the system and cost is pretty staggering.
- [17:37]
Benjamin Shapiro
Yeah, I just, my personal experience is the reason why the costs are going down is because the user's getting frustrated and they're hanging up the phone, not actually completing the journey. So it actually gets into the next thing is how do you measure the ROI of all of this? Whatever your AI implementation is, sure. Hey, I'm using my chatbot. Going through customer service and we have less calls, people are probably getting frustrated and quitting. How do you measure what the business impact is of the AI implementation from.
- [18:05]
David Rabin
A studio AI standpoint? This tool that we're using for almost all of our marketing enablement in our solutions and services group, this was easy to measure. I'm measuring time savings and cost savings. My cost per asset go down about 90%. My time savings goes down about 70%. Because the example I told you earlier, it's not this two weeks of back and forth. It's about 15 minutes, 20 minutes. Okay, maybe it takes a day because you got to route it to somebody else to look at. So that's material savings. Go back to the demand based ABM program we talked about earlier. I know that my engagement with customers is higher. Why? Because I'm targeting them better and I'm giving them content which actually makes sense for what they want to see as opposed to we're using a lot more human touch here. I think the customer might want this and I can't give you a number yet, but we are seeing shorter sales cycles. Right. Because we're speeding up the train. The customer is getting information faster. That makes more sense. I'm handing over a lead to a sales rep a lot faster. So the cycle's getting shorter, so we're seeing a lot of that. The other way to look at this. And I've actually read a little bit about kind of the AI view of the old marketing adage about faster, better, cheaper. You can pick a couple of those.
- [19:10]
Benjamin Shapiro
You took the words out of my mouth.
- [19:11]
David Rabin
Oh, did I really?
- [19:12]
Benjamin Shapiro
I was just going to say everything I know about the way the world works. Fast, cheaper, good. Pick two. And it seems like we're like, hey, we could do a lot and we could do it fast and it doesn't cost very much. It's like, all right, well, are we sacrificing quality?
- [19:24]
David Rabin
Exactly. Okay, so let me talk out of both sides of my mouth. The first is again, go back to the example I gave you. If you're going to see significant savings in time, time to produce, and in cost, what about the better? Is it going to be better today? And I'm going to be totally blunt with you. I don't care if it's better today because the cost savings and the time savings are so material. Even if it's 15 or 20% less effective, and I'm not saying it is, but even if it is 20% less effective, I'm okay with that because I'm going to move the train so much faster at a lot lower cost down the road. If we take things like, why do we let airplanes get flown by computers? Because guess what, they make fewer mistakes. Now, you need the human touch there. But I do believe we will get to a point using the tools that we're using in marketing, that the quality and the output actually will become better.
- [20:11]
Benjamin Shapiro
I've had different experiences and I do think that there's a scenario where you can do fast, cheap and good, but it requires more implementation up front. And I'll give you the example of sort of our pre production process when we write our interview scripts. It used to be judgment me looking at the company that you work for and looking at your LinkedIn page and looking at maybe some of your previous speaking experience and seeing if you're somebody that I want to talk to. And I'm just using intuition. We move towards an AI first model where we are doing more deep research on who you are, your background, your company, your previous speaking experience, your alignment with our goals. Are you talking about topics that are interesting to our audience and we're able to do deeper, better fact checking to understand who you are as a speaker? That's how you got on the podcast. I might have looked before and been like, Lenovo, that's a hardware company. And now I'm like, oh no, they have a services business and here's why. They're the perfect fit for coming on the Martech podcast. I might have just completely steamrolled your application and it's the AI that helped me do more thorough research. And I actually do think it's faster, cheaper and better.
- [21:20]
David Rabin
Yeah, let's go back to that help desk. Example, at an E comm establishment, they're likely willing to sacrifice a little quality because they know it's going to be infinitely faster and substantially cheaper. They're probably going to be kind of closing down some of their service centers. They're willing to accept a little bit of a degradation in quality and customer satisfaction today.
- [21:40]
Benjamin Shapiro
But doesn't that have a long term impact on the brand?
- [21:43]
David Rabin
Yep, probably does. But tomorrow it's going to be better and the day after tomorrow it's going to be even better. I mean, look, this is when, you know, companies, including us, you start looking at less expensive ways to service customers. Some companies will offshore, right? You know that there's going to be some sort of a trade off. You just hope the trade off to the good outweighs the trade off to the bad.
- [22:02]
Benjamin Shapiro
All right, the last question I have for you. What should marketing leaders expect to break in their current stack when they're moving towards an AI first approach?
- [22:12]
David Rabin
I mean, something will break, right? Because the perfect world for an AI solution is that tool A works with tool B, which works with tool C. When it's all manual, it doesn't really matter. When I've got a content development tool that doesn't really need to talk to the data, it doesn't need to talk to how I publish because we're all manually. It's a relay race. I'm going to hand from point A to point B to point C. It doesn't really matter in the future. I mean, let's use web as an example. If I want to get something up on the glass in 15 minutes, something's got to write it. I got to know who I'm talking to, what I want to talk about. It's got to feed in the product imagery. I've got to have a content library to store it, all this stuff and I got to be able to publish it. And qa, it's all of it has to work together. If one of those tools doesn't sync up. As a marketing leader, I'm going to find another tool. So I'm going to put that first tool out of business, go find one that works. So something along your marketing stack will likely need to be replaced because it's not all going to string together like you want and like we do today, because it's manual.
- [23:05]
Benjamin Shapiro
The biggest thing that I've learned about doing my implementation, albeit at a much smaller scale than some of the clients that you're advising is the context filtering. We think about prompt engineering. I just interviewed the head of AI at HubSpot. And he said it's not prompting it's context. The models now are so good that you can type in basically garbage and it'll figure out what you mean. But if you're not restricting what information it's looking at IT trials to boil the ocean and you get these wide variances in responses. So when you're thinking about what's going to break, to me, it's that context layer and whatever you're using to filter what information is relevant. That's the thing you have to figure out first whether you're pulling it from an ABM content market. Like all the different tools, there's the plumbing, the sort of IT and infrastructure. Yes, you have to build that. But what actually makes it good is figuring out the right context filtering. And that's something that really requires human thought to be able to figure out.
- [24:01]
David Rabin
Absolutely. And it also requires upscaling. Again, instead of me writing a brief saying I need a brochure on this new product and we're trying to talk to an IT leader who buys infrastructure solutions, I've got to put a little bit more thought into it now to make sure that the output is kind of what I'm thinking about. So, absolutely, the people that get ahead of the race and start getting a little bit smarter than they are today because they lose the crutch of more teams and resources, they're the people that will succeed over the next year. Three years, five years. But I totally agree with the HubSpot commentary. Prompt engineering is sort of being phased out into something much bigger.
- [24:34]
Benjamin Shapiro
All right, and that wraps up this episode of the Martech podcast. Thanks to David Rabin, the Chief Marketing Officer at Lenovo Solutions and Services Group, for joining us. If you'd like to contact David, you could find a link to his LinkedIn profile in our show notes or on martechpod.com or you can visit visit his company's website, which is lenovo.com and if you haven't subscribed yet and you want a daily stream of marketing and technology knowledge in your podcast feed, hit the subscribe button in your podcast app or on YouTube and we'll be back in your feed every week. All right, that's it for today. But until next time, my advice is to just focus on keeping your customers happy.
- [25:15]
David Rabin
Foreign.
- [25:32]
Podcast Announcer
Thanks for listening to the Martech podcast and I hear everything. Production Looking to launch or scale a podcast like this one for your brand? Then visit iheareverything.com.