Back to Basics: Essentials for Product Development in the Age of AI

In this podcast episode, Dr. Beverly Wright, Vice President of Data Science & AI at Wavicle Data Solutions, and Marcy Cunningham, Head of Information Governance at IHG Hotels and Resorts dive into a timely conversation on the importance of returning to foundational data practices, especially in an AI-driven era. This episode explores why understanding and reinforcing the fundamentals can make or break product development efforts powered by AI.

Speaker details:

• Dr. Beverly Wright, Vice President of Data Science & AI at Wavicle Data Solutions
• Marcy Cunningham, Head of Information Governance at IHG Hotels and Resorts

Watch the full podcast here or keep scrolling to read a transcript of the discussion between Beverly and Marcy:

Beverly: Hello, I’m Doctor Beverly Wright and welcome to Tag Data Talk. With us today I am thrilled to have finally, Marcy, you’re so overdue on being on this podcast. Marcy Cunningham, head of Information Governance at IHG Hotels and Resorts. Thanks for being here, Marcy.

Marcy: Thank you for having me. I’m really excited about it.

Beverly: Very nice. So, tell us a little about yourself. Why are you so cool, besides your neat boots? We’ve got to say something about the boots.

Marcy: Well, thank you. I probably wouldn’t normally put that label on myself, but given our audience, I will say I’m cool because I loved data before data was cool. I’ve been in this field a long time, started technologist in general, but quickly found a love for data. I’ve had an opportunity to work for some great companies, Home Depot, Terradata, and now IHG hotels and resorts, and I did my first job out of school was for the Florida Lottery, which was a really fun experience.

Beverly: I know some people are into scientific games, and they say it’s a fun experience.

Marcy: Yes, they were a great partner for us then. But it’s really been an exciting journey to kind of work as part of our technology and data teams at IHG, but then also kind of with this recent role of information governance, I actually work in our business reputation and responsibility organization, which looks out for the reputation of our brand and making sure we’re doing the right thing for our customers. And so, it’s been a great opportunity to kind of learn the business from a different perspective.

Beverly: Brand equity is everything. I mean, it’s so crazy how much it matters, yeah.

Marcy: Oh, totally.

Beverly: Totally. Yeah, interesting. What is it you love so much about data? I have to ask you that follow-up question.

Marcy: I just get so excited about the opportunities, right? I mean, I guess, back to the cool thing. I’ve just always known it had such potential and such power. So, whether it’s making a guest experience better or you’re helping a child who has a condition that’s not been discovered before. Any of the things that can be very personal, or it can be kind of very broad for a company. I just love it. It’s a puzzle. You’re constantly trying to figure out the puzzles.

Beverly: Yes, I love that. Thank you so much. I’d forgotten you were at Teradata too. Very interesting. So, we’re talking about a topic that I absolutely love right now because AI is all the rage. And everybody’s talking about AI. And when we had a prep call, you said, hey, we need to talk about back to basics. So, we’re talking about back to basics: essentials for product development in the age of AI. And we’re trying to get data and we’re using AI to do so. We’re trying to get it into the hands and really into the eyes of these consumers to make more data-inspired decision making. But some of the issues are that we’re not doing the basics. What do we mean by that? What does that mean? The basics.

Marcy: I think anyway, there will be some components we very much touch on the AI piece of it. But when we talk about just being more data-driven as organizations, it really gets to a point where we’re trying to, there are various stakeholders across the company who now want to leverage data. They also understand, they get excited about the opportunity it can have to help make them more efficient or better or achieve their goals. But yet they don’t have all these years of kind of the discipline behind it. And I feel some of those basics are just having business context for data, understanding what something is. There’s a difference between a loyalty member or someone who stays with us? Or is there a difference between a reservation?

Beverly: Stay or something?

Marcy: Stay exactly. There’s actually quite a bit of difference between the two, so it’s.

Beverly: Go ahead, go ahead.

Marcy: I was going to say, so it’s having that, it’s having that context for people who aren’t used to having to think in that way. it becomes super important and that kind of verges into quality and things like that which become much more important on the AI side as well.

Beverly: Well, what I was going to hint about is you’re talking about a different audience. Really. It sounds like with AI we’re experiencing a completely different group of people, whereas before, actually, Shannon Harlow and I call it convergent and divergent.

Marcy: Oh yes, I’ve talked to Shannon about this before. I love that.

Beverly: Convergent is data and sequel and BI and analytics and modelling and now AI. And so, if you’ve been through that journey and those steps in that process and you’ve advanced, you’ve got enough context that when you’re leveraging AI, if something’s not right, you kind of know it. You have a sixth sense. But you’re saying nowadays with AI making things more user friendly, which is a good thing.

We’re not saying it’s not a good thing, it’s a great thing. But now we’re dealing with this divergent nature of it where we’re pulling in stakeholders that are leveraging AI, but they don’t have those steps. They haven’t experienced a model that failed because you put the Y on the X side or you know, they don’t have some knowledge in the subtext of having gone through that. Yeah. And you think that’s the kind of maybe the root of where this challenge is coming from?

Marcy: Yeah, I mean, I think that that’s a big part of it. And that’s actually one of the things I’ve learned about working in different parts of our company is I’m oh, I didn’t realize I was using terms others didn’t understand or I didn’t realize that I had all this context and or experience that would lead me to making choices about whether or not something was good or bad or usable or not usable, or the fact that I might need to drill down or look at it from a different angle. It’s just not something that is part of the DNA of a lot of the stakeholders of the data now.

Beverly: So, me and you talking would be different, but you’re saying that now because of AI we have natives and immigrants. Is that accurate? We have natives and immigrants of the data.

Marcy: Yeah, yeah. I mean I find myself doing a translation on a regular basis.

Beverly: I see. So that’s made, you’ve had to sort of go back. Now I see why you’re saying I have to go back to basics because the things that are presumptuous for me and you may not be so for these new group of users, these new immigrants, so how what kind of basics are. They may not understand the statistics that are going behind the scenes, so they may not understand certain factors. Tell us what those basics are?

Marcy: So, I think there are some statistics and those aspects behind it, but I think there’s even a deeper level than that. I think it does come back to business context for the data and key characteristics about the data. I’ve thought a lot about this because I do think it’s important, especially for those of us who have been in the field for a while to not just say, well, we’re doing this because we’ve always done it, and this is what you do. It’s 2 + 2 for us to some degree. But I don’t love questioning things 2 + 2. But then I also think we should, we should test ourselves to make sure these things are still valuable.

And I again, just have discovered in conversations and working with different teams that having that base understanding of the data and what that means and the business context for it. It’s just something you have to have. And I also kind of equated it just over time and thinking about when you, it comes back to our title a little bit, products, right? And you would never think about creating a physical product that you were not capturing certain characteristics and attributes about that. You have to know how you’re going to manufacture it, how you’re going to create it, where, what type of place does it need to be stored, how?

Beverly: What kind of material?

Marcy: Is it sensitive? Is it diamonds? You need to, you know, you need to protect it in a certain way, and you need to make sure that then when a consumer goes to find that they can find what they need. It’s such a basic concept when we think about physical products, it really shouldn’t be any different than for data. We’re just trying to make it that easy for the people who need to leverage it to understand what it is they’re getting. Is it going to meet their needs? Is it fit for purpose from a quality perspective?

And that’s what I really mean when I’m thinking about those basics is in those characteristics, it’s just making sure that we are giving all the right context and characteristics of that data. So, when you as a data consumer, not a guest, but a data consumer needs to use it, that you know what you’re getting, and you can be as independent as possible when you get to that point.

Beverly: Love it, love it. Very interesting. So, I like your analogy of physical tangible products, and we would never do that. We would never say I don’t know where the supply chain is going to come. We would obviously know exactly where the supply is going to come from and the quality of that. And we talked about data quality, hinted about that and I’ll come back to that in a second. But as an example, a couple years ago I had, this is a big company, they were a great client in the end, but they wanted us to better understand customer experience and loyalty. And so, we came up with a business problem.

We came up with a hypothesized model. We said, ok, this is what you’re going to need, where’s the data? And they gave us sales data and somehow, they knew they in their head, they said this is the data. So go tell us about customer experience and customer loyalty. And so, there’s that sort of mismatch with not understanding what kind of data is even needed? And that’s a part of it. It’s the business context, understanding the actual data. And then you talked about quality of data. So, tell us more about that.

Marcy: Yeah, I think quality of data is oh, there’s a love hate relationship with that because everybody can intuitively say, oh, of course you need good quality data. In our world people say things like garbage in, garbage out.

Beverly: Right, but what does that mean?

Marcy: Right. Yeah. And frankly, when it comes down to quality, we’re thinking about fit for purpose. Again, kind of just revisiting that reservation and staying a little bit, right? You know, if you, if you made a reservation, but you don’t actually stay, you can’t use the reservation to sum up what all your revenue is, because you may or may not have gotten that people may not have gone through with a, with a stay. And those are just things that I think are part of making sure you give some characteristics around what that data, you know, represents, such that then the consumer of it can do their best to understand that.

I also think there’s a little helping to qualify what you can do and can’t do or what maybe has not been explored, and you just want to give enough ticklers that you can then still have conversation. We’re not saying everything won’t require some subject matter experts to kind of embed in it, but this is kind of, I think where we start to hit on the AI components a little more when you’re making decisions at that volume. You can get way out of whack quickly.

Beverly: This is out at scale. Everything is at scale at this point.

Marcy: You know, and you can read, you can read the news on a daily basis to find those examples of where some pretty severe things for companies have happened because of that. So, I think just that is another piece where if you’re not kind of thinking about quality seriously, you can end up just not where you intended.

Beverly: Yeah, no kidding. And just a little tiny bit off over and over and over and over. And you do big numbers. It’s problematic. And I think about too, as a data scientist, in the years that I’ve done modeling, you have to really interrogate the data. I mean, interrogate it, you almost assume evil from the data, and you have to prove that it’s good and it’s right. And maybe it sounds like you’re saying these immigrants don’t know to do that. And so, we’re having to sort of, that’s why we’re having to go back to basics. Does that seem?

Marcy: Yeah, I think that’s right. And I mean you should be doing it any time you have kind of a new use of data evaluating to make sure it really is. But I think there’s a point in there too that I kind of wanted to make around. There are producers of data within the company and one of the really exciting things about all of this is that their products that they are creating are now being used by so many different people and teams across the company.

And this is an opportunity to say where maybe we haven’t always been able to kind of really have intense kind of quality rule sets and monitoring in place. This is a real opportunity to put some of that in place such that you can operate at, at scale. There are some efficiency components in there as well, which is another big just kind of win.

Beverly: Right. Yeah, Yeah, it makes sense. That makes sense. When you talked about terminology too. And tell us more about that. What are your thoughts on, when we go back to basics and we have to think about the terminology, reservation versus a stay or those kinds of terms. Tell us more about it. You think there’s a little bit of a LAX with people not thinking about exactly what that variable is?

Marcy: Of course.

Beverly: Yeah, there is.

Marcy: And I do think, I mean, we do things where, and it’s not just here, I’ve done this at other places as well to at least categorize things at a high level, right? You know, a customer and then you, you have breakdowns of that where you have, loyalty members, non-loyalty members, frankly, our owners are also customers. And so really trying to give them tools at a high-level tag again to help prompt some of that thinking characteristics, right. I would not have known that I wanted a pre lit Christmas tree, except that all of a sudden, I’m Oh, this one says pre lit, this one doesn’t say pre lit.

That’s probably something I should, think about. And so, I think some of that very basic terminology so that you’re also not over complicating the process is important to prompt the thinking. And then just mature that over time, as new use cases come into play, as the business starts to leverage in different way, just make sure that that’s a constant curation process.

Beverly: You talked a little about if you’re not careful and you’re not thinking about the basics, sometimes you can be off a little and then a little, a little, a little. And it almost just keeps exacerbating itself because of the volume of some of these big companies. But are there other challenges or big problems or issues that happen as a result of not really thinking through just the basics?

Marcy: I always say and then in my current role again, the information governance, what we do is think about both the value of data and then certain risk associated with it and how to balance those decisions within the company. But I always have in my little risk list the 1st, the biggest risk for the company from a data perspective is not leveraging our data to help us achieve the objectives of the company.

I mean we have to do that in a thoughtful way. It is making sure we’re thinking about the risk etcetera, as we make those decisions super important given our industry. But if we are not making that data findable, is that a real word?

Beverly: Or the center of the decision, you know, a part of it.

Marcy: I guess if I’m going back to the big risk of not using the data, if you can’t find the data, you’re not going to be able to use it. And so, I think again back to making sure you’ve got those key characteristics about the data such that those who need to find it know how to find it or even know it exists in some cases.

Beverly: So, there’s all kinds of little, not little, but there’s all kinds of problems and issues that can happen. I’m going to be, straightforward. It’s problematic. Give them incorrect answers, answering the wrong problem or, or the wrong questions or, not addressing the right opportunities. But the biggest thing is to just not even use data and just follow your gut. Can you imagine that today?

Marcy: A really bad place to be in.

Beverly: Yeah, yeah, it doesn’t seem a good idea.

Marcy: I just, I do think, I think it’s an undervalued practice to do that. And it is just we’re not quite yet making the full connection between how we think about physical products and how we think about data and how important those characteristics are, you just wouldn’t even question.

Beverly: No, you wouldn’t. Well, and that’s a great segue because I was going to ask how do we get back to the basics? And it sounds like one of the ways is to think about data, not just as data, but you said data assets, to really think about it as a product and an asset. What are some other key tips for getting back to basics?

Marcy: I think I touched on this earlier when I talked about the producers of data, right? Yes, this is what we’ve been waiting for, right? For a long time, is for it to be able to be leveraged. And so I think to some degree, leveraging that excitement and recognizing, I don’t know another, if I stick with the product analogy, this is from the local fair where you’ve got your new mug or jewellery or whatever it is out there to now, you’re selling on Amazon.

Beverly: You’re big time now.

Marcy: And that may not sound appealing to everyone, but the reality is it means your data has value. And so if you’re thinking about the fact that, Oh my gosh, I get very pumped about the fact that, my data is now going to be leveraged by all these different parts of the company, I think it’s easier to get behind the fact that you then want people to be able to find it. So, I think there’s just a shift in mindset there. And if we start to think about it from that standpoint, you get really excited about, helping others find it. I think the other thing too.

Beverly: Just to make sure I play that back. It sounds like you’re saying that it’s not enough just to start leveraging it and making better decisions, but you also have to show your excitement. And so maybe there’s an advocacy play here too. Do we have a responsibility to kind of advocate and champion for our data?

Marcy: Yes, yes. And I think, at least when you’re thinking about it in that asset model, you’ve got people who kind of own it and have some responsibility for making sure that others in the company can leverage it. And so there actually is a promotion aspect to it as well. So, I think the other thing that I was thinking about is there is something about making that process not too cumbersome.

There are some, I may have talked to this earlier but just making sure that you’re kind of getting those most important characteristics and then just having it become part of your DNA that you are embedding this into the way that you work. It’s just a common principle and you just build on that over time, those characteristics just start to, develop and you can even make that a community effort as well that there are others who want to kind of add tags that have helped them find that data etcetera. So, I think it becomes very powerful when you start to see that.

Beverly: I see. So, some of the ways that you were saying we can overcome and kind of go back to basics are to think about your data as an asset, as a product, think about how you would handle it if it were an actual tangible thing. That’s definitely a great one. Second is we’ve been waiting for this day, and we’re excited. I remember when I graduated Marcy, there were six people getting decision sciences degrees. That’s it, and now there are hundreds.

Marcy: I think we were right about at the same time, yeah.

Beverly: There are hundreds in one program at one school and there were 6. They’re probably about 15 in the whole country. So now our day has come, and people are valuing data and so we need to show some excitement. It’s not enough just to leverage it and produce good models. We have to actually advocate for it and become a champion. And thirdly, I’m not going to use the term friction less or, completely frictionless, but at least try to make the processes, more palatable and try to make it what Monica Kudersky, she was on Tag Data Talk. She talked about it being business as usual. Yeah. How to make leveraging data BAU, just a regular thing. It’s not some tada, here I go. I’m going to use data. It should be a part of what you’re doing.

Marcy: Then that’s where I was going with that DNA comment, right? It just becomes part of the culture and part of the way that you work, it makes such a difference.

Beverly: Yeah, what? Go ahead. I was going to say what final piece of advice would you give Marcy for helping people really get back to basics, thinking about the essentials of product development in the age of AI?

Marcy: I’m going to stick with this is a passionate group of people who have been in this, in this field for some a long time, some not a long time, right? I even actually at the, at the tag event that was last night, I got something from someone who was fairly new in the field and just talked about how they’ve navigated because they knew that there were such great outcomes, in front of us. And I would just say, if you’re, if you’re a business partner who’s trying to leverage data, stick with those data people because they want so badly to help you use it. And I think the data people just know that this is your moment.

We’re super passionate about making sure that it can be used. I find that I’m going to say more so in this field than I have with others. And so, the final one was maybe that collaboration, I just think is still going to continue to be important. That’s maybe not the back to the basics, but in order to get back to the basics and kind of really get the value from it. I also think that the people who are using it should demand that we do these things for them.

Beverly: I love it. So, collaborate, bring people along in the journey, kind of have some empathy for others that don’t quite know as much, and maybe help them along and get immersed with each other’s worlds. And we can all sort of bond as a community. I love that. I think it’s going to take that.

That’s what it’s going to take and more and more. So, we got to kind of get it right now before it starts becoming even bigger, right? Thank you so much to Marcy Cunningham, head of Information Governance at IHC. Thank you for being with us on Tag Data Talk.

Marcy: Thank you for having me.

Mastering the fundamentals is non-negotiable in the age of AI. From treating data as a product to embedding business context and ensuring clarity in terminology, organizations must refocus on the basics. As data becomes more accessible, advocacy, discoverability, and collaboration will define success. Because without a strong foundation, even the smartest AI can’t deliver meaningful results.

Explore the full catalog of TAG Data Talk conversations here: TAG Data Talk with Dr. Beverly Wright – TAG Online.