Latest / Shiny New Object - a Marketing Podcast / Episode 311 / Hernando Ruiz-Jimenez / Geisinger Health / Chief Marketing and Communications Officer
Transcript
- 0:00Hernando Ruiz-Jimenez: You really need to understand the
- 0:01basics. It's more of a problem that we think that people are
- 0:05relying a lot on the technology and not really knowing what lies
- 0:08behind that technology.
- 0:16Speaker: Are you a fitness loving foodie? Then defy snacks
- 0:18is for you. Defy offers real high quality chocolate you love
- 0:21combined with functional ingredients, giving you as much
- 0:24as 26 grammes of protein per bag. So you can be both a little
- 0:27bit bad and a little bit good at the same time as a female owned
- 0:31brand, defy dedicates a portion of its profits to other women
- 0:34owned businesses currently sold in the US. You can find us on
- 0:38defy with an i snacks.com and follow us on social at
- 0:42defisnacks.
- 0:48Tom Ollerton: Hello and welcome to the shiny new object podcast.
- 0:51My name is Tom Ollerton. I'm the founder of automated creative,
- 0:54the creative effectiveness ad tech platform, and this is a
- 0:57podcast about the future of data driven marketing. Every week or
- 1:00so I have the pleasure and the privilege of interviewing one of
- 1:03our industry's leaders about their vision for the future of
- 1:06data driven marketing. And this week is no different. I'm on a
- 1:09call with Hernando Ruiz-Jimenez, who is Chief Marketing and
- 1:13Communications Officer at Geisinger. So Hernando, for
- 1:17anyone who doesn't know you, can you give us a bit of background
- 1:20and tell us what you got, what you do.
- 1:22Hernando Ruiz-Jimenez: Hi, Tom, thanks for having me on your
- 1:25podcast. I'm Hernando Ruiz-Jimenez, I'm the Chief
- 1:28Marketing and Communications Officer for Geisinger Health.
- 1:33Geisinger, it's one of the largest health systems in the
- 1:36state of Pennsylvania. Just to put it in context, if
- 1:40Pennsylvania where a country in terms of GDP, would be one of
- 1:44the top 20 countries in the world. So we are a pretty large
- 1:49health system with over 10 hospitals, over 7 million
- 1:52patient visits a year. And also, what makes us also interesting
- 1:57is that we are in the more rural and industrial areas of the of
- 2:01the state. So we're kind of solving health care. We're
- 2:05looking at, how do we provide the best health care to those
- 2:09populations? You know, when most of the focus is most in urban
- 2:13and suburban areas? So, you know, great organisation, doing
- 2:17great work. My background for context, I started in CP
- 2:21actually, I started in advertising. So I guess we all
- 2:25have a dark past. They moved to CPG with Pepsi and Diageo for
- 2:29many years. That's where I got my traditional academic
- 2:33background. Became an entrepreneur on the digital
- 2:36front. After that, a publisher for a while, and then I joined
- 2:41healthcare about 10 years ago on the insurance side of the
- 2:45business, then moved to to the provider side with New York
- 2:48Hernando Ruiz-Jimenez: speak.
- 2:49Presbyterian, and then now with Geisinger Health with, you know,
- 2:49Tom Ollerton: Does she understand what you do for a
- 2:53in the role of Chief Marketing and Communications Officer.
- 3:41living?
- 3:42Hernando Ruiz-Jimenez: I don't think so, to be honest, but
- 3:45let's not tell her that.
- 3:47Tom Ollerton: I have this theory, though, that if you can
- 3:49explain to your parents what you do for a living, then you're
- 3:51probably not making lots of money. Maybe that's I haven't
- 3:56stress tested that anyway, I hope you have a fantastic
- 4:00birthday with your mom, or birthday. Sorry, so are you a
- 4:05marketing book worm, or do you sort of learn on the job?
- 4:08Hernando Ruiz-Jimenez: I do both. I like to, you know, I'm
- 4:12actually a professor of marketing, you know, on at City
- 4:17College of New York, once a quarter, once a year, over a
- 4:20semester, I teach a class on on marketing, basically, on what
- 4:23doesn't marketing department do for the for one of the Master's
- 4:27programmes. So I like the academic side of marketing a
- 4:31lot, but at the same time, I think that you learn by doing,
- 4:34and that's, that's, that's a lot of my focus.
- 4:36Tom Ollerton: Do you have a particular marketing book that
- 4:38you find yourself going back to or recommending?
- 4:40Hernando Ruiz-Jimenez: When I go back to I like the classics. One
- 4:44of the first books that really got me into marketing and
- 4:47actually into advertising was Confessions of an advertising
- 4:51man by Ogilvy. It's basically his autobiography. It's a
- 4:55fascinating book, and it actually makes you think about
- 4:59how. How all this started to get qualified, and how this people
- 5:05were actually looking at data and understanding what was the
- 5:08things that worked and didn't work in advertising. So that's,
- 5:13for me, that's a fascinating book.
- 5:14Tom Ollerton: What's the main takeout like? If you're gonna
- 5:17put that in someone's hand and say, Look, you have to read this
- 5:18book. What's the kind of main lesson that you stayed with you.
- 5:22Hernando Ruiz-Jimenez: I haven't read it in a while, but I think
- 5:24that, back to your point is, what stayed with me was this
- 5:27idea that there's a science behind everything we do in
- 5:30marketing, that even in the 40s and 50s, they were looking at,
- 5:37how do we create more precise advertising that has better
- 5:39response rates. And they were looking at type, you know, type
- 5:45phones. They were looking at amount of words in a page. They
- 5:49were looking at what type of photographs to use, which is not
- 5:51different from what we've been doing when we do AB testing. So
- 5:55it's, it's, it's, for me, the the big takeaway is that you
- 5:58really need to go back and understand where things are
- 6:01coming from, to really know where you're going.
- 6:09Tom Ollerton: So do you have a practical piece of advice for
- 6:13data driven marketers on how to be better? Is there something
- 6:16you find sharing in an academic scenario or a live working
- 6:20scenario most often?
- 6:22Hernando Ruiz-Jimenez: This is one that I am truly passionate
- 6:24about it, and I try to instil it in my teams and in my students,
- 6:27is that you really need to understand the basics. And we
- 6:31think we do, and sometimes we don't. Let me give you an
- 6:33example when, when I talk about the funnel, the marketing
- 6:37funnel, in class five years ago, probably do you understand what
- 6:41it is? Probably half the class would raise the hand. Half the
- 6:44class wouldn't today, it's more like 90. But then I do an
- 6:48exercise where I go through the funnel, and then I give them a
- 6:53test, and I take them, you know, I give them, you know, it's like
- 6:56you have so much budget you're going to buy at this CPM. This
- 7:00is your margin. What is the ROI? Take me through the math. Take
- 7:04me through the logic. Take me what have you done? And then I
- 7:08asked them, you know, so you need to improve that ROI. What
- 7:11would you do? Because the ROI first is negative. So what would
- 7:13you do in terms of of, you know, through the funnel, what are the
- 7:17type of things you can do on how would that affect your ROI? And
- 7:20you'll be surprised that a lot of people don't know how to do
- 7:25that. They actually, they understand the funnel because
- 7:29they see it at work, and they work with it, and they come out,
- 7:31it's automated, and they they can see it, but when they really
- 7:35have to understand that, you know, a click is somewhere
- 7:38actually going and clicking, and then that actually opens up a
- 7:41landing page. And then landing page, you have all these
- 7:44elements, and you need to work with them, going through that
- 7:48logic and really, knowing that really helps you be a much
- 7:51better you know. And it gets more complicated as you do other
- 7:54things, you know, but a much better marketer. And just to
- 7:59finish on the point I've been interviewing, not now, but in
- 8:03the past. Job interviewing marketing directors, digital
- 8:06marketing directors, that when I went through that level of
- 8:09detail, they couldn't give me answers. So it's, it's, I think
- 8:13it's a, it's a more of a problem that we think that people are
- 8:17relying a lot on the technology and not really knowing what lies
- 8:21behind that technology.
- 8:22Tom Ollerton: So we're gonna move on now to your shiny new
- 8:25object, which is the responsible use of AI in healthcare
- 8:30marketing. So that's a fairly self explanatory one. Sometimes
- 8:33they're a bit vague for that one's right on the money. So why
- 8:36have you chosen that and help the audience understand exactly
- 8:39what you mean?
- 8:40Hernando Ruiz-Jimenez: When you work... again, let me go back
- 8:43when I talk to my CPG friends and we talk about the
- 8:47sophistication of marketing in CPG versus healthcare. People
- 8:53always think of CPG as the, you know, the gold standard, and
- 8:57healthcare has always been behind. But when you go deep
- 9:01into it, the amount of data that we have on our patients and on
- 9:05our clients and our customers, it's huge. And the problem, not
- 9:10the problem is. The issue is that this is very personal data,
- 9:14and this is something that requires a lot of privacy, and
- 9:19it's not hard to do things that can go wrong. An example, I mean
- 9:26sending a list of celebrating, you know, births to a couple
- 9:36that cannot have kids, and you know, they're going through an
- 9:40infertility programme or sending a congratulatory message to
- 9:46someone who has just died. Those are kind of like simple ones,
- 9:50but things can actually go wrong if you don't use data the right
- 9:55way. So the challenge that we have is tonnes amounts of data.
- 9:59How do we actually use it in a way that can really help our
- 10:05customers? How can we use them to use predictive analytic tools
- 10:10to try to get you into the right journeys towards better health,
- 10:17and at the same time, do it in a way that it doesn't affect your
- 10:21privacy. You don't feel that you're being invaded and it
- 10:27we're not giving data to other people that you don't want
- 10:30people to see or to know about.
- 10:37Tom Ollerton: This episode of the shiny new object podcast is
- 10:41brought to you in partnership with madfest, whether it's live
- 10:44in London or streamed online to the global marketing community,
- 10:47you can always expect a distinctive and daring blend of
- 10:50fast paced content, startup innovation, pictures and
- 10:53unconventional entertainment from madfest events. You'll find
- 10:56me causing trouble on stage, recording live versions of this
- 10:59podcast and sharing a beer with the nicest and most influential
- 11:03people in marketing. Check it out www.madfestlondon.com.
- 11:14So I love that intention. So you're sort of pulled in two
- 11:17directions there, aren't you? You, you want. You've got a load
- 11:20of data and owned data, and also good data, right? It's there.
- 11:25You're not trying to sell someone a product they don't
- 11:27want or need. You're trying to improve their health, right? So
- 11:30you've got a tonne of data, tonne of levers to pull in order
- 11:33to deliver about a health outcome, which is fantastic.
- 11:36Then on the flip side of that, you've got the excitement and
- 11:39the innovative possibilities with artificial intelligence,
- 11:43but in a world of hallucinating, of artificial intelligence, that
- 11:47must be terrifying, and the two examples you gave are awful, and
- 11:50I'm sure they've happened and feels bad for the people that
- 11:54are on the receiving end of it. I'm sure there was no ill
- 11:57intent. So how do you manage that huge chasm between your
- 12:01want and need to do the right thing, and the ability to do it
- 12:04with the reality that AIs make stuff up.
- 12:09Hernando Ruiz-Jimenez: That's why we are very, very careful on
- 12:12how we move. And we need to put guardrails on AI. We all want to
- 12:21run with it. We know what it can do. Part of the problem, to your
- 12:26point, is, not only that it hallucinates is yes, that it has
- 12:29to do with the data that you have. If you have data that is
- 12:31biassed, AI is going to take that bias data and, you know,
- 12:35algorithmically, take it to another level. An example could
- 12:40be, you know, who do you give a new medicine? And you know you
- 12:45have data that this medicine works better on white male kids.
- 12:52But it may be, have to do, the fact that your population that
- 12:56has been taking this before is, yes, that type of population. So
- 13:01it's, it's hard to really, really make the systems and the
- 13:07algorithms work in a way that really takes into account all
- 13:13the measures and all the all the parameters that we needed to
- 13:16take. And it's right now, it's working progress. So that's why
- 13:21you see that in healthcare, there's certain parts where
- 13:24we're moving very fast, fast with it, which is more on the
- 13:26medical side of it, on the marketing side, we need to be
- 13:30careful.
- 13:30Tom Ollerton: And what are those steps that you are putting in
- 13:33place to be careful. So automated creative has always
- 13:37worked with healthcare clients to different degrees in
- 13:40different markets for different parts of healthcare. And
- 13:44obviously it's an incredibly regulated space, and we produce
- 13:49hundreds and 1000s of ads in different campaigns in different
- 13:52languages all over the world. And so we have to put safeguards
- 13:55in place to make sure that all of the right messaging that
- 13:59supports the ad goes in the right ad, and all the different
- 14:01sizes and all the different places. And it's been and it's
- 14:04been work, and we've got that fine art down, but that is when
- 14:07it's it's specified, right? But in an AI world, when you've got
- 14:11AI making decisions, what kind of safeguards do you put in
- 14:14place? Is it, do you use different AIs to police other
- 14:16AIs? Or is it, do you like a human layer? How's that working?
- 14:19Hernando Ruiz-Jimenez: I think what what we're looking at is,
- 14:21how do we limit the role of the AI to various specific things?
- 14:27It's more going on the route of agents than full fledged AI. I
- 14:32think that's where the solution might lender might end up. We
- 14:35are. We're doing pilots. One of the key things is start small,
- 14:39and so if you make mistakes, the mistakes are small, and you can
- 14:44control them. So we are working on on pilots. The other key
- 14:48thing, I believe, is it all starts with your data. The
- 14:52quality of the data needs to be excellent, because that's where,
- 14:56also where things can go can go wrong. You know, I think one of
- 15:00the key things is it's start small and pilot things. Control
- 15:04the errors. As I was saying, we're probably going to go more
- 15:09the route of agents. So it's very controlled. AIs, it's not
- 15:13full fledged AI, trying to do everything with a black box in
- 15:17the back end. So the more we can control it and say, Okay, we
- 15:21want it just to do this type of thing. That's where we believe
- 15:25that the use of AI can be very, very effective. An example, for
- 15:29example, something we're going to be testing is pulling lists
- 15:34for for campaigns, email campaigns. Right now, it takes
- 15:39someone hours to go in and look at different databases, clean
- 15:43the list. I mean, it's very manual stuff. If we can get AI
- 15:47to take care of all of that and actually do some of the data,
- 15:52the quality assurance of it, that'll be fantastic. So that's,
- 15:57I think, where we may end up going at the beginning, as we
- 16:01get a better handle of what AI can do for us and where things
- 16:05could go wrong, and we can control that.
- 16:06Tom Ollerton: So this is a point that I'm really struggling with
- 16:15when it comes to AI in marketing. So the example you
- 16:18gave, and I'm sure there's lots of others, and I appreciate you
- 16:20sharing a live example with me, which is, instead of someone
- 16:23pulling an Excel spreadsheet and finding all the people in New
- 16:26York state who have this and then pulling together a list and
- 16:29then dumping that into an email service provider, the AI, can do
- 16:33that in what seconds, right? Or even less. So what's happened
- 16:37there is that you've saved time, right? And so therefore, if it's
- 16:41a freelance cost, then you say, also save money, right? So my
- 16:45argument is, is that you can do that, and that's brilliant, but
- 16:49then so can all of your competitors, right? So, and
- 16:51everyone should be this stuff's not expensive. So what, what
- 16:54everyone has the ability is to save that, that hour long task.
- 16:58But as in, as an industry, as a vertical, everyone's still in
- 17:02the same place. Everyone's just saved an hour. So assuming
- 17:06everyone's going to do all of the obvious things, what's the
- 17:10extra stuff that brands can do when it comes to healthcare
- 17:13marketing? Where's the new frontier that AI is opening up?
- 17:16Hernando Ruiz-Jimenez: It's going to be on engagement.
- 17:18That's where I think, because, to your point is, one thing I
- 17:21would like to point out healthcare is that we are we're
- 17:24kind of frenemies. It's great if the health system next door does
- 17:28well with its patients, we're all better off. So we are very
- 17:33competitive, but at the same time, we collaborate a lot. So
- 17:38I'm not that worried about, you know, the competitive advantage
- 17:41this could give us. For me, it's more about how do we better
- 17:44engage with our patients to have the behaviours that we need,
- 17:48healthy behaviours, getting them to actually come and do the
- 17:54screenings they need to do. An example I can give you
- 17:58something. We're working on lung cancer, there are other cancers.
- 18:02There are very specific screenings. Colon cancer. You do
- 18:05colonoscopies, mammographies, lung cancer is much harder
- 18:10because there's not any specific screening that you do. So using
- 18:14AI, we could actually you know how to use those predictive
- 18:17analytics to find those that have the higher propensity, and
- 18:22how do we take them through the right journey and automate those
- 18:24things in order to get them quicker in front of the right
- 18:27doctor and the right treatment to prevent them to, you know,
- 18:32actually having the cancer, if possible? So for me, that's the
- 18:35way that AI it's going to help us really having that engagement
- 18:39and that trust coming from our patients, and that's why the
- 18:42responsible use of it is critical, because if we lose
- 18:46that trust, we would not be able to do it.
- 18:48Tom Ollerton: And so what's marketing's role in that? That's
- 18:51an amazing thought and a lovely ambition, that you would be able
- 18:55to crunch data quickly and encourage people to go into
- 18:59testing, maybe before they were thinking about it, and increase
- 19:02the quality and length of people's lives, but how did and
- 19:05understand how your own data might deliver that? But that
- 19:09feels like a an operations function, as opposed to a
- 19:12marketing one or am I missing the point?
- 19:14Hernando Ruiz-Jimenez: It's a marketing function in the sense
- 19:15of what we do in marketing is finding who is your who is the
- 19:19right customer for the right service, and that's exactly what
- 19:23we can accelerate. I can actually now use data to get in
- 19:27front of the right cardiologist, the right patient. I can
- 19:32actually and that, at the same time, results into revenue for
- 19:36the system and growth. So the way we see a role, it's it's,
- 19:40it's as a business, it's as a growth engine. You know, we're,
- 19:43we're business partners that are trying to do exactly that is,
- 19:47how do we, you know, create that relationship with our customers,
- 19:51you know, our patients. I like to call them customers, because
- 19:54I think they're, they're more than just a patient. And how do
- 19:59we build. That trust. So we are, you know, we get them on
- 20:02journeys that actually improve their health.
- 20:04Tom Ollerton: So unfortunately, we've come to the end of the
- 20:06podcast, and there's just so much more knowledge in your head
- 20:09that I would have to get out, but we've run out of time. So if
- 20:12someone wants to carry on this conversation with you, where is
- 20:16a good place to do that and what makes a message that you'll
- 20:19actually respond to?
- 20:20Hernando Ruiz-Jimenez: Probably the easiest way is through
- 20:22LinkedIn, Hernando Ruiz-Jimenez Geisinger Health, and you'll
- 20:25find me. The best way to do it is make it personal in the sense
- 20:30of, hey, I listened to the podcast. I have a question about
- 20:34x. That's easier than when I get, you know, you probably get
- 20:39them too. Is we share tonnes of people in common and interest. I
- 20:43would love to connect to you. So So, yes, tell me, you know, give
- 20:48us a little bit of context, and I'll be more than happy to
- 20:51connect.
- 20:51Tom Ollerton: That's a lovely bit of advice. Thank you so much
- 20:54for your time.
- 20:54Hernando Ruiz-Jimenez: No, thank you Tom. It was fun.
- 21:02Tom Ollerton: You Hi, just before you go, I'd really
- 21:06appreciate it if you could take the time to write a review of
- 21:09the shiny new object podcast on Apple podcasts, or iTunes,
- 21:13whatever it's called these days, or whichever podcast provider
- 21:16you use. We're at any podcast, so it would go a long way for
- 21:21us. If you could just share the Word and give us a bit of a
- 21:23support on those channels, that'd just be fantastic. If you
- 21:26haven't got time, that's also cool. And yeah, if you could
- 21:28tell your colleagues about the podcast, and also, if possible,
- 21:32don't forget to subscribe. And I'd love to hear your feedback.
- 21:35If you'd like to speak on the podcast or be a guest, or you
- 21:39think I'm asking the wrong questions, anything. I'd be
- 21:41super interested to hear what you think. So please email me at
- 21:45Tom at, automated creative.net that's Tom at, I'm not gonna
- 21:51bother spelling it anyway. You'll work it out. Thanks so
- 21:54much.