Cloud Wars Live with Bob Evans · 2026-10-30 · 18 min
Key moments - from our scoring
Substance score
39 / 100
Five dimensions, 20 points each
At Oracle AI World in Las Vegas, Bonnie Tinder met with seven customers across diverse industries - including Heathrow Airport, Baylor University, ADT Security, Securitas, and Northwell Health - to understand real-world AI adoption challenges. The key finding across all conversations: clean data and a strong system of record are non-negotiable prerequisites before AI can deliver value. Customers emphasized that security controls, governance, and compliance built into Oracle's suite beat bolt-on tools, particularly for regulated industries. Tinder identified a critical gap emerging in AI maturity: while organizations can now predict problems (employee attrition, security risks, fraud), many struggle to act on those insights. She highlighted the shift from dashboards to "systems of action" - the need to move from awareness to human-led intervention. Performance management and total comp statements emerged as effective low-risk entry points for building organizational trust in AI. While hard ROI numbers remain elusive early in adoption, customers report time savings and productivity gains that create a foundation for future value capture once humans feel empowered to act on credible predictions.
Hosted by Bob Evans. The show scores 52.7/100 on substance across its analysed episodes. See the full show profile.
Clean data and a strong system of record are the top priority; every customer interviewed emphasized that attempting to turn on AI features without this foundation would fail, as Baylor University learned before becoming successful with Oracle applications.
Oracle's native AI includes context awareness across the suite, security controls, provisioning, and governance built-in, which meets compliance and security requirements more effectively than bolt-on tools or self-developed applications.
Low-risk, easily automatable processes like performance management and total comp statements serve as quick wins that build organizational trust and demonstrate AI value before moving to higher-value activities like open enrollment and skills training.
While customers gain time and productivity, converting that into measurable dollar savings requires reinvesting the freed-up time into revenue-generating activities; additionally, humans must first trust and act on predictions, which is still developing across organizations.
Organizations must evolve from simply viewing AI predictions in dashboards to actively making human-led decisions and interventions based on those insights; identifying a problem like employee attrition is only valuable if the company takes steps to prevent it.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode surfaces a handful of real customer-derived themes - clean data as a prerequisite, governance-before-ROI in regulated industries, predictive attrition at Heathrow - but the content is spread thin over 18 minutes with significant conversational padding, and most observations are well-worn enterprise AI talking points rather than novel operator intelligence.
what they said was if we had not had clean data and we hadn't had this concerted effort around ensuring that our data was useful to begin with, turning on all the AI switches would just not have been successful at all
the fact that AI is, is inside the system beats any bolt on tools or any sort of self developed, sort of, uh, you know, AI applications that they could have, um, you know, done on their own, potentially for cheaper
Almost every major claim - clean data first, start small with low-risk use cases, build trust, ROI is elusive - is standard enterprise AI consensus that has circulated widely for two-plus years; there is no contrarian argument, no first-principles reframe, and the guest even leans on clichéd aphorisms rather than fresh thinking.
start with the low risk, uh, use cases that you can Easily automate
trust is earned in drops and lost in buckets
Bonnie Tinder is a legitimate practitioner running a peer-review research firm for enterprise software and came with direct access to seven named Oracle customers, which gives her claims grounding; however she is fundamentally an analyst/researcher rather than an operator who has deployed AI at scale herself, limiting the depth of first-hand operational insight.
I had the opportunity to meet with seven customers since I've been here
I will write detailed write ups on all of these, uh, interviews that I've done so you can look at, um, you know, more detailed analysis
The episode earns credit for naming real companies (Heathrow, Baylor, ADT, Securitas, Northwell Health) and providing a few concrete details like Baylor's headcount and go-live year, but it delivers almost no hard outcome metrics, dollar figures, or quantified productivity gains - the specificity is in the roster of names, not in the evidence of results.
Baylor University, um, so Texas based university, 8,000 M, uh employees, 20,000 students
Baylor University for instance, went live on Oracle, uh, in 2020. Um, so suite of applications include the HCM. Um, and it took them, uh, several years to, to actually get that foundation built
The host repeatedly restates or extends the guest's own points rather than probing for deeper specifics - he never asks for a single hard number, never challenges the vagueness around ROI, and his interjections often consume air time with his own commentary rather than unlocking new information from the guest.
And, uh, what we're hearing more and more, not about in this specific case, Bonnie, but as you know, the step one, uh, raise awareness of the problem
Yeah, and you can't do that without the data being right, the complexity or the compliance and governance all tied in before that
Computed from the transcript - who did the talking, and the words that came up most.
At Oracle AI World in Las Vegas, Cloud Wars host Bob Evans sits down with Bonnie Tinder, founder and CEO of Raven Intelligence, to explore what enterprise customers are actually experiencing as they implement artificial intelligence. Drawing on conversations with seven Oracle customers across industries, including Heathrow Airport, Baylor University, and Northwell Health, Tinder identifies the critical factors shaping successful AI adoption. Oracle AI Gets Real The Big Themes: Clean Data Drives AI Success: Clean, reliable data emerged as the most important foundation for successful enterprise AI adoption across all seven customer conversations Bonnie Tinder conducted at Oracle AI World. Organizations emphasized that their systems of record needed to be established and optimized before AI could deliver meaningful value. Baylor University, which implemented Oracle applications in 2020, spent several years strengthening its data foundation before expanding AI-enabled workflows. Employee Trust Determines AI Adoption: Trust emerged as a recurring theme.
Transcribed and scored by The B2B Podcast Index.
Speaker A: Foreign. Welcome back to Cloud wars in Las Vegas at Oracle AI World, and delighted to be speaking with our longtime friend Bonnie Tinder, the founder and CEO of Raven Intelligence. Bonnie, good to see you.
Speaker B: Great to see you, Bob, back here in Las Vegas.
Speaker A: Yeah. Yep. Sometimes it feels like, uh, like home, but not really home.
Speaker B: That's. It's our home away from home. Yes.
Speaker A: Yeah. So Bonnie, lots going on here. You, uh, it seems like you really wanted to focus a lot of your time on your discussions with customers. So on the one hand there is a slew of, uh, sort of very advanced sweeping new things that Oracle's doing with AI, but kind of in the real world of what's going on with customers, I think it was great how you focused your time on what's their perception of things and, and how did that go? So let me turn it over to you and what came really through as top of mind for you?
Speaker B: Yeah, well, I had the opportunity to meet with seven customers since I've been here. Um, you know what I think was really impressive this year was just the number of customer examples that they've had of customers, uh, who are using AI in production. Um, I think they said 50 or so sessions. Um, are we featured? And I got to spend uh, some one on one time with uh, seven of those. Um, so I thought I'd talk about a couple of themes that I've heard. Um, and I'll, I'll mention some of the, the customers that I met with. So one of them was Heathrow Airport. Baylor University, um, so Texas based university, 8,000 M, uh employees, 20,000 students, um, ADT security company, um, Securitas, uh, financial services company, Northwell Health was another one. Um, so almost too many to mention at this point, but uh, all of this to say there's some themes that I noted about each of them that I'd like to talk about.
Speaker A: Absolutely. But I think I like it that the set of customers you talk to from all different types of industries.
Speaker B: Yes, yes, definitely. Uh, well represented across the board from not for profit financial industries, manufacturing, um, you know, uh, and you know, security companies as well. So a really good balance of different industries and different use cases for AI as well. Um, I would say number one in every conversation that I had, the foundational key to success was clean data. Um, and this idea that the system of record had to come first. Um, and before they could really feel AI became useful, that initial starting point had to be in. So Baylor University for instance, went live on Oracle, uh, in 2020. Um, so suite of applications include the HCM. Um, and it took them, uh, several years to, to actually get that foundation built. But in since 2022, they've really sought to increase the number of workflows that they're doing within their system and now have turned on several different areas, um, of AI for their employees. But what they said was if we had not had clean data and we hadn't had this concerted effort around ensuring that our data was useful to begin with, turning on all the AI switches would just not have been successful at all. So clean data number one, um, number two, um, was the fact that many of these firms felt a great deal of confidence turning on AI within Oracle because of the context awareness that it had across the suite, as well as the security controls and provisioning and governance built in. Um, you know, many of these companies or organizations are regulated industries. So this idea of compliance and security comes before being innovative and progressive. And so the fact that AI is, is inside the system beats any bolt on tools or any sort of self developed, sort of, uh, you know, AI applications that they could have, um, you know, done on their own, potentially for cheaper. Uh, so that was the second, you know, really big thing across every, every different organization was the fact that, you know, governance and security came first before any return on investment can even be, you know, discussed.
Speaker A: Okay, yeah, yeah, it makes sense. And it's the combination of those things. Right. You know, uh, we've heard different tech companies for a while saying, hey, if you want to move in the direction of AI, you got to get your data clean. And there's probably a lot of companies that have been trying to get their data clean for other reasons, but either felt, ah, uh, you know, we don't have the time, we don't have the budget, uh, we don't have the expertise, the technology is too hard. But I think this, uh, the uh, intermittent arrival of AI in every industry has become a real differentiator. It's been a forcing function
Speaker B: that
Speaker A: I think has made this happen. But then, uh, if you're going to get that right, as you said, Bonnie, you got to get the data right, but then also ensure that you got the, uh, compliance and governance tied in there as well.
Speaker B: Yeah, for sure, for sure. Um, you know, I asked, uh, every single one of these customers, what's a, what's a lesson learned? What would be advice you'd give, uh, another company that isn't as far along in the AI journey? And you know, overwhelmingly I heard, you know, start with the low risk, uh, use cases that you can Easily automate. Um, you know, some of those uh, you know, the low hanging fruit, so to speak, the easy stuff, um, many of them used the performance management um, process as low hanging fruit. It's a point in time, sort of um, high friction event that you can help managers, uh, and employees prompt through the performance management process, which there's always a lot of pain uh, associated with that, a lot of manual updates and things like that. Um, and that was something that you know, definitely Heathrow, uh, Airport and Baylor both felt like was a quick win, that they could show their uh, you know, employee popular population that hey, there's a lot of um, you know, a lot here that we can do with, with AI. So they start small, use that, use total comp statements as a way that employees um, could sort of start and engage with the system and then they moved on to higher, um, you know, higher value type of activities such as open enrollment, um, and uh, you know, skills and training uh, within the system. Um, so the, the I guess the, the lesson is start small, low risk and then you know, easy wins. Build trust within the organization.
Speaker A: Bonnie. I was going to say it's sort of like the a uh, foot in the door but then I was thinking camel's nose under the tent. But I, I'm not sure if. Anyway, let's, let's get my things there but, but get, get something done. Get a win on the board.
Speaker B: Yeah, yeah.
Speaker A: Okay.
Speaker B: And you know trust is such a huge word, a huge theme throughout this you know, conference. Um, you know, so many of these customers, you know, know that when they introduce AI to their organization, um, you know it's. What is it? Trust is earned in drops and lost in buckets. And so, so many of them want to ensure that when they're rolling out these programs that they are embraced, they're adopted. Um, and that you know they, they really earn the trust of their, their employees and continue to build on that with future, um, you know, future enhancements.
Speaker A: Yeah, and you can't do that without the data being right, the complexity or the compliance and governance all tied in before that.
Speaker B: 100%. 100%. Um, you know the other thing that I heard um, and this was conversation with Heathrow Airport is uh, you know they were able to through um, the fdi, the um, the, the, the data intelligence, um, that they're using through Oracle to identify at risk employees. And um, so you know they were able to look at that and potentially predict attrition. So that was an interesting use case for how they're using analytics and, um, and AI, uh, to look at their employee base. And so, you know, it sort of begs this question. I think the next phase of AI is, you know, we're able to now predict things.
Speaker A: Yeah.
Speaker B: And the Magic 8 Ball is getting pretty accurate, you know, given, uh, all of the context that it now has. But, you know, in the case of Heathrow Airport, so I guess the next question was, well, what did you do about that? Yeah, right. You can predict that these, um, you know, employees could potentially turn over. So what did you do to ensure that that didn't happen? And so that, I think, is the larger question I think with every single customer is, okay, now we've gotten these tools turned on, so is that moving the needle, like, now, you know, who's going to turn over? So are you able to be proactive and ensure that they don't? Or is it just that, you know, sooner
Speaker A: that
Speaker B: you're going to have a problem on your hands? So all of this to say, you know, I think, you know, that's one example of many with predictive, um, insight that we have through AI. But I think the larger question is going to be, okay, you can lead a horse to water.
Speaker A: Yeah.
Speaker B: Like, how are your decisions now going to change as a result of that? So.
Speaker A: And, uh, what we're hearing more and more, not about in this specific case, Bonnie, but as you know, the step one, uh, raise awareness of the problem. Step two, get the agency AI, you know, seen over by humans and sort of prescribed by humans to go in, start to take the action about that. You know, does this person live too far away? Can't fix that? Does this person have a different feeling that maybe some intervention or steps taken would keep this person. But you need to hear about this thing right. Of the. The turnover, the cost of getting a new employee into the. Into the Commodore, into the airport and then trained so forth like that. It. It's, uh. And the drag that puts on other people who are doing, you know, that training, stepping away from their regular jobs to do this. So, uh, it. This sounds like, you know, an ideal case for somebody like that to say technology led to a very specific benefit and ROI for us.
Speaker B: Yeah. Oh, yeah, absolutely. And, you know, in order to. For the ROI to be there, you actually have to, at some point, like, do something.
Speaker A: Yeah. Uh,
Speaker B: with the knowledge that you've gained and that insight. And if you don't, you know, all the great information in the world can do. Do you. No good, essentially.
Speaker A: Yeah.
Speaker B: So,
Speaker A: and. And if they don't do something and say See why we spend all that money on AI? Uh, you know, all we did was get a heads up early on these people leaving.
Speaker B: Yeah,
Speaker A: no, but that, that's, that's got to be though, that type of thing. I heard a few times over these past couple of days, people talking about we had systems of record and we need to advance that or evolve that to systems of action.
Speaker B: Yeah.
Speaker A: Identify the problem, take steps to address it,
Speaker B: from dashboards to actual decisions. It's one thing to have a dashboard.
Speaker A: Uh-huh,
Speaker B: Right. But you know, if I look at my dashboard and I see I'm speeding and I don't do something about that. Um. Well, you know, that's right. That's right. Who are we going to blame? Hallucinating.
Speaker A: That's right.
Speaker B: Ah,
Speaker A: not your fault.
Speaker B: No, no, no, of course not. Yeah. So, um, and I guess the, the last point sort of related to this is that, um, you know, I think at all of these conferences, not, not just, you know, AI World, but many of these conferences that we attend were all of these new AI tools are coming to light. We're looking for these hard, hard ROI numbers right now. And it's a little bit elusive.
Speaker A: It's
Speaker B: a lot of it elusive. Right. I think what I heard from all of these customers initially was they've gotten time back, they've gotten higher productivity, um, and there's a lot of things that have made their job easier, such as, you know, these performance evaluations, automating that process, taking that pain away. Um, but does that mean that I can then turn that into, well, that saved me a million dollars a year, uh, because it gave me the time back. Unless you can reinvest that time and some revenue generating activity like it's, it's hard to come up with that, that dollar figure. You know, with that said, the foundation is there and I think what we will see in at this time next year is now as trust is earned and you start looking at those dashboards and see accurate measures, uh, there's comfort level that now humans in the loop will feel empowered to say, all right, now I'm going to start acting on, uh, these predictions because I know them to be credible. And I think the ROI foundation is there. It's just a little bit early to realize that just yet.
Speaker A: Hey, Bonin, let me toss in one sort of corollary to that again. Uh, throughout this event here at AI World, I've heard, um, a number of, uh, situations where businesses are finding that the first sort of, um, the first step in their ROI journey is About AI is going to be really good at finding problems. Like in this case, the potential people who are going to leave the company.
Speaker B: Yeah,
Speaker A: um, there was a lot of talk about with security. Right. Uh, and AI's ability to go through all your systems and identify, hey, you know, the good news is I've identified all the problems. The bad news is there's a lot of problems. But um, we hear about it in things like preventive maintenance, potential, um, fraud, you know, number of cases like this. So as I think, uh, you know, we gotta have sort of courage and stiff backbones because AI is going to surface a lot of ugly stuff from what you've been talking about here with these great customers to some other things. And then quickly, I think there's got to be a real tight connection between. Here's the awareness of problem. Now here are the steps I can take to address those in whatever part of the company it is.
Speaker B: Yeah, yeah. Really and truly. And you know, I mean you could look at the problems that it surfaces and get depressed about it or you can like actually do something to ensure that those problems don't escalate and snowball. And it's uh, this ability that we have to now peek around the corner that we haven't had in the past. I know, you know, I do it on a personal basis and, and in, you know, the work that I do as well is, you know, there's predictions out there that help us to steer the ship and you know. Right. You know, write things, um, before they become a bigger problem. So I think it's the same thing with all these customers.
Speaker A: Okay. Okay, well, Bonnie, thanks. Interesting, um, insights. And again, love the range of customers you had, the, the issues they were facing and the industries they're in. Good
Speaker B: stuff. Yeah. And I will write detailed write ups on all of these, uh, interviews that I've done so you can look at, um, you know, more detailed analysis of what each one of these customers said. But these in aggregate were sort of the themes.
Speaker A: Okay.
Speaker B: Yeah,
Speaker A: Bunny, where could people find those, those write ups?
Speaker B: Raven intel is, uh, my company, ravenintel.com or definitely check it out on uh, LinkedIn. They'll be posted under, uh, under my name.
Speaker A: Well, great. Bonnie, always a pleasure. Good to see you here. Uh, and thanks for these insights that you've given us and I hope there's safe travels home after AI world.
Speaker B: Likewise. Thanks so much, Bob.
Speaker A: All right, thanks. And folks, thanks to all of you for being with us here on this episode of Cloud Wars. Been talking with Bonnie Tinder, the CEO of Raven Intelligence. We'll see you next time.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
#407 - How Martin Gourdeau refused the commodity race and added $1M ARR in 9 monthsThe Remarkable SaaS Podcast · on System of record68 / 100
Workday CIO and CTO on Why the Agent Era Belongs to PlatformsFuture of Work · on System of record68 / 100
AI Deployments Need Less Speed, More Velocity, with Raju MalhotraSaaS Scaled · on System of record65 / 100
The Real Reason Your L&D Team Is Overwhelmed with John PeeblesThe HR L&D Podcast · on System of record65 / 100
DM Radio Live from Accelerate: Why Context Is the Missing Link for Enterprise AIDM Radio · on System of record58 / 100
Episode 813: A Conversation with Lori Hock, CEO of OpptlyThe Future of Work Exchange · on System of record55 / 100Microsoft, Oracle, Google, AWS: $2.3 Trillion Backlog + RPO64 / 100OpenAI Partner Network Signals New Era of Enterprise AI Deployment35 / 100