Exclusive Coverage: theCUBE at IBM Think | Alvin Francis & Tony Baer

theCUBE.net

Clips
News
More from Exclusive Coverage: theCUBE at IBM Think

Tony Baer

Principal

dBInsight

Alvin Francis

VP Product Management, Business Analytics & Watson Applications

IBM

How IBM’s watsonx BI assistant simplifies the decision-making process

Artificial intelligence should be a huge boon to the business intelligence software market, but according to IBM Corp., companies are struggling to use decision-making tools because they are too complex.IBM recently previewed its watsonx BI Assistant, which employs AI to help users make decisions about their business and simplifies the process of analyzing data systems to obtain insights.IBM’s Alvin Francis (left) and dbinsight’s Tony Baer (middle) discuss how AI is revolutionizing decision-making with theCUBE’s Dave Vellante

play_circle_outline Revolutionizing Decision-Making: Introducing Watsonx BI Assistant, IBM's Interactive Solution at Think 2024

play_circle_outline Revolutionizing Analytics: Enhancing Cognos Reports, Exploring WatsonX BI Assistant for Intelligent Data Apps

play_circle_outline Unlocking the Power of Watsonx: Bridging the Gap Between Insights and Actions with Orchestrate Integration

Info

Alvin Francis & Tony Baer

Tony Baer

Principal dBInsight

Alvin Francis

VP Product Management, Business Analytics & Watson Applications IBM

search

>> Welcome back to Boston. Everybody. Come on inside theCUBE here at IBM Think 2024. My name is Dave Vellante and we're here with Alvin Francis, who's the vice president of product management for Business Analytics at IBM. And Tony Baer, friend of theCUBE, he's a principal at dbInsight, even though he's got way more than one insight. Tony, Alvin, welcome to theCUBE. Good to see you guys.

Tony Baer

>> Good to see you again, Dave.

Alvin Francis

>> Excited to be here. Good to see you.

>> So watsonx BI Assistant, what's it all about? What problems does it solve? Why is it needed?

Alvin Francis

>> Awesome. Thanks a lot. That's a good question, Tony. So first, it's a pleasure to be here. watsonx BI Assistant is a brand new BI solution that we've actually launched here at Think. It's a conversational first interactive product. It is focused on addressing some of the challenges that we have in the BI space right now. If you look at what's happening in BI, organizations spends millions of dollars on tools every year, and we still suffer from adoption issue with BI. Only 30% of users that actually need business insights to make business decisions actually leveraging these BI tools. watsonx BI Assistant is targeted at addressing this adoption issue to ensure that BI is available for the decision makers when they need it, how they need it, in a way that is easily consumable.

>> So it's simplifies the consumption. What's happening in the space today, the BI professional? They burn out on the task and they just can't get to it, or they're so busy, it's like a data scientist they're just wrangling data? Is that the issue?

Alvin Francis

>> Well, the issue is because of the complexity of the existing tools. You hear a lot about self-service, but it's not really self-service for the decision makers, it's self-service for the experts, for the power users. And so decision makers still have to rely on the data analyst to prepare, curate, and create these very complex dashboards so that they can consume. And so this organization, the data analyst organization becomes a bottleneck in that entire process. And so with BI Assistant, watsonx BI Assistant, our goal is to ensure that the decision makers can easily access the business insights they need when they need it without having the dependency of a data analyst.

>> I see. Okay. And, Tony, the BI industry is been created to address the business needs, to understand what's happening, and it's become this very complex hyper-specialized pipeline. What's your take on all this and your take on the IBM announcement?

Tony Baer

>> Yeah. Well, put it this way. We've been through I would say probably three generations of BI. And keep in mind you said it's supposed to serve business. What's it called? Business intelligence. And the original thought, and I remember back in the '90s, back when we were still holding logs and water from the well, is that the whole idea was this is an alternative to green screens. It was when client server came on, so everybody would have a dashboard on their desktop. And that was certainly a step in the right direction. You have to view all of this in terms of what the technology was capable of at the time. It obviously did not deliver on that promise. And that's where basically about 10 years later we got second generation, which was a self-service visualization. And what that did, it still required IT to prepare the data extracts, but now an end user could download these data extracts and start playing with them. Of course, you ended up having multiple versions of the truth. And the fact is there still was a fair amount of complexity, but it was an incremental improvement, a dramatic improvement I should say actually, over 1.0. What we're seeing today now with natural language and generative AI is that we expect to have the same experience that Captain James T. Kirk did on the Starship Enterprise back in 1968, which is essentially to figuratively talk to our computers and maybe literally talk to our computers and ask questions in our conversational language. And now with generative, we can get that back.

>> So is it my understanding that the linear progression here, it was sort of building cubes and there's just a few people who are capable of doing that to a business analyst that might not have been as technical, but it was pretty savvy, could build dashboards and visualization to now you're talking to the machine, that's nirvana. How close are we to that?

Alvin Francis

>> Well, actually we have demos that are both today. I think if you look at some of the technologies that we leverage at IBM, it's built in watsonx. So you could say it's powered by some of the models that we have in watsonx, which gives us the capabilities to have this interaction. I would say that technology has matured over the last few years, so we are very close. I would say that just in terms of the accuracy that we have seen in terms of the results, we are very confident that we'll deliver on that promise. The approach that we took is not necessarily just, hey, let's just make it a conversational, but we wanted to declutter BI. So if you think of some of the tools that are there, they're delivering on BI, but they're still very complex and very intimidating for the average user. So it's not just the fact that it's conversational, but it's very simple. Removing all the need to learn a BI tool. So imagine never having to learn a BI tool ever again. You just converse in your natural language and you get the information that you need.

>> I would love that. What happens to sort of old analytics like Cognos? How does that fit?

Alvin Francis

>> So the strategy that we are using is to ensure that watsonx BI Assistant is embeddable. As you can imagine, we have a very large customer base of Cognos, and most of our customers use Cognos for Cognos reports. With watsonx BI Assistant we're going to enable these customers to derive more value from Cognos. So for example, they'll be able to ingest the metrics in the reports and now converse with the data on which the reports are built on. So you could say that these products are going to continue to co-exist, but we are able to use BI Assistant to modernize the Cognos report experience for customers.

>> So Tony, that doesn't go away, but I can still see a lot of potential innovation, maybe disruption when you start... We've API-ified the metrics. We're always talking about the semantic layer to go bring analytics and bring transactions now into the equation, although that seems to be more distance every time you and I come-

Tony Baer

>> There's always a hope.

>> We're still hoping for that. But this idea of feeding intelligent data apps and obviously BI is a huge part of that, how do you see it all playing out?

Tony Baer

>> Put it this way, this is where I think IBM was pretty savvy in terms of what they did. Had this had been, say, five years ago, whatever, IBM's default approach would've been let's just add a conversational interface to Cognos and just make it look simpler. What I give IBM credit for is they rethought this and said, look, where is this going? And yes, we need to basically bring our existing user base, our Cognos install base, up to date with this. But we also need to reach out to the new generation of, I should say, distributed apps, distributed data, distributed intelligence. And so therefore, let's make this embeddable. This doesn't have to be its own application. IBM is not going to be in the BI... I mean, yes, they're in the BI tools business with Cognos, but that's yesterday's business. Today we need intelligence on the data that we're getting and it's not just essentially the traditional analytics, but we need to use machine learning and in some cases generative to extend this and add more broader context.

>> You're saying it's a different... You're right-

Tony Baer

>> It's a different world....

>> back then when IBM bought Cognos it was part of an IBM transformation.

Tony Baer

>> Yes.

>> Steve Mills was architecting. Now you're saying it's using AI to enable business insights or maybe even eventually business actions. And that's the IBM sort of value add here. Is that right?

Alvin Francis

>> Well, that's part of it. So just building on that concept of business actions, there is a chasm that exists today between the insights and actions. A lot of organizations believe that because they derive insights from the data, they're leveraging these insights to make business decision or take actions. Part of our strategy is actually to integrate watsonx BI Assistant with watson Orchestrate. So directly from the user experience, insights can be used to automatically drive actions and close that chasm between insights and actions.

>> I love that, by the way. And this ships when?

Alvin Francis

>> It's Q3. So we are actually signing up customers right now for private preview, and we had, as you can imagine, a lot of traffic at the booth. A lot of customers have signed up. We'll co-create with a number of our key customers and we will continue to ensure that we are obviously addressing the use cases. You talk about how close we are in terms of delivering on the promise, but we want to make sure that we're working closely with our key customers, identifying some of the key pinpoints, solving it, and so that when we in Q3, we're going to hit it out of park.

>> So let's say it ships this fall a year from then, let's say think 2025, what do you want to be able to say in May of 2025 at IBM Think, if that's when it is, that you can't say today day?

Alvin Francis

>> So our focus is obviously to drive adoption. And so the way we will measure success and hopefully at Think next year is to look at how organizations are adopting BI Assistant to actually transform the entire BI organization and how they leverage data to make decision. So what I can't say today obviously is the number of customers that we will secure, but at Think next year, we should be able to pop the champagne in terms of saying that we've been successful. Obviously this product will continue to evolve. There's still a lot to be done in terms of how we transform BI. We are fairly confident that we're on the right path and it'll evolve from a capability perspective, leveraging maybe more powerful models, in some cases smaller models that will integrate to address specific use cases. But by Think next year we should have obviously continued to lead the curve and be differentiated.

>> And real quick, Tony, what do you want to be able to say a year from now in this space that you can't say today?

Tony Baer

>> What I want to see essentially is that where we're doing... I'm thinking a lot of the stuff that you're talking about with the Orchestrate, where we're talking turning insight to action. You guys are promising that, I want to see that as reality next year.

>> Guys, great analysis. Congratulations on the announcement. Thank you so much for coming on, theCUBE.

Tony Baer

>> Thanks Dave.

Alvin Francis

>> Thank you very much. Thanks for having us.

Tony Baer

>> Good talking .

>> We're wrapping up. You can hear the conference hall is loud. People is getting ready to go out tonight. This is Dave Vellante. Keep it right there. More action from IBM Think 2024 you're watching theCUBE, the leader in enterprise tech coverage. Right back.

Alvin Francis & Tony Baer

search

>> Welcome back to Boston. Everybody. Come on inside theCUBE here at IBM Think 2024. My name is Dave Vellante and we're here with Alvin Francis, who's the vice president of product management for Business Analytics at IBM. And Tony Baer, friend of theCUBE, he's a principal at dbInsight, even though he's got way more than one insight. Tony, Alvin, welcome to theCUBE. Good to see you guys.

Tony Baer

>> Good to see you again, Dave.

Alvin Francis

>> Excited to be here. Good to see you.

>> So watsonx BI Assistant, what's it all about? What problems does it solve? Why is it needed?

Alvin Francis

>> Awesome. Thanks a lot. That's a good question, Tony. So first, it's a pleasure to be here. watsonx BI Assistant is a brand new BI solution that we've actually launched here at Think. It's a conversational first interactive product. It is focused on addressing some of the challenges that we have in the BI space right now. If you look at what's happening in BI, organizations spends millions of dollars on tools every year, and we still suffer from adoption issue with BI. Only 30% of users that actually need business insights to make business decisions actually leveraging these BI tools. watsonx BI Assistant is targeted at addressing this adoption issue to ensure that BI is available for the decision makers when they need it, how they need it, in a way that is easily consumable.

>> So it's simplifies the consumption. What's happening in the space today, the BI professional? They burn out on the task and they just can't get to it, or they're so busy, it's like a data scientist they're just wrangling data? Is that the issue?

Alvin Francis

>> Well, the issue is because of the complexity of the existing tools. You hear a lot about self-service, but it's not really self-service for the decision makers, it's self-service for the experts, for the power users. And so decision makers still have to rely on the data analyst to prepare, curate, and create these very complex dashboards so that they can consume. And so this organization, the data analyst organization becomes a bottleneck in that entire process. And so with BI Assistant, watsonx BI Assistant, our goal is to ensure that the decision makers can easily access the business insights they need when they need it without having the dependency of a data analyst.

>> I see. Okay. And, Tony, the BI industry is been created to address the business needs, to understand what's happening, and it's become this very complex hyper-specialized pipeline. What's your take on all this and your take on the IBM announcement?

Tony Baer

>> Yeah. Well, put it this way. We've been through I would say probably three generations of BI. And keep in mind you said it's supposed to serve business. What's it called? Business intelligence. And the original thought, and I remember back in the '90s, back when we were still holding logs and water from the well, is that the whole idea was this is an alternative to green screens. It was when client server came on, so everybody would have a dashboard on their desktop. And that was certainly a step in the right direction. You have to view all of this in terms of what the technology was capable of at the time. It obviously did not deliver on that promise. And that's where basically about 10 years later we got second generation, which was a self-service visualization. And what that did, it still required IT to prepare the data extracts, but now an end user could download these data extracts and start playing with them. Of course, you ended up having multiple versions of the truth. And the fact is there still was a fair amount of complexity, but it was an incremental improvement, a dramatic improvement I should say actually, over 1.0. What we're seeing today now with natural language and generative AI is that we expect to have the same experience that Captain James T. Kirk did on the Starship Enterprise back in 1968, which is essentially to figuratively talk to our computers and maybe literally talk to our computers and ask questions in our conversational language. And now with generative, we can get that back.

>> So is it my understanding that the linear progression here, it was sort of building cubes and there's just a few people who are capable of doing that to a business analyst that might not have been as technical, but it was pretty savvy, could build dashboards and visualization to now you're talking to the machine, that's nirvana. How close are we to that?

Alvin Francis

>> Well, actually we have demos that are both today. I think if you look at some of the technologies that we leverage at IBM, it's built in watsonx. So you could say it's powered by some of the models that we have in watsonx, which gives us the capabilities to have this interaction. I would say that technology has matured over the last few years, so we are very close. I would say that just in terms of the accuracy that we have seen in terms of the results, we are very confident that we'll deliver on that promise. The approach that we took is not necessarily just, hey, let's just make it a conversational, but we wanted to declutter BI. So if you think of some of the tools that are there, they're delivering on BI, but they're still very complex and very intimidating for the average user. So it's not just the fact that it's conversational, but it's very simple. Removing all the need to learn a BI tool. So imagine never having to learn a BI tool ever again. You just converse in your natural language and you get the information that you need.

>> I would love that. What happens to sort of old analytics like Cognos? How does that fit?

Alvin Francis

>> So the strategy that we are using is to ensure that watsonx BI Assistant is embeddable. As you can imagine, we have a very large customer base of Cognos, and most of our customers use Cognos for Cognos reports. With watsonx BI Assistant we're going to enable these customers to derive more value from Cognos. So for example, they'll be able to ingest the metrics in the reports and now converse with the data on which the reports are built on. So you could say that these products are going to continue to co-exist, but we are able to use BI Assistant to modernize the Cognos report experience for customers.

>> So Tony, that doesn't go away, but I can still see a lot of potential innovation, maybe disruption when you start... We've API-ified the metrics. We're always talking about the semantic layer to go bring analytics and bring transactions now into the equation, although that seems to be more distance every time you and I come-

Tony Baer

>> There's always a hope.

>> We're still hoping for that. But this idea of feeding intelligent data apps and obviously BI is a huge part of that, how do you see it all playing out?

Tony Baer

>> Put it this way, this is where I think IBM was pretty savvy in terms of what they did. Had this had been, say, five years ago, whatever, IBM's default approach would've been let's just add a conversational interface to Cognos and just make it look simpler. What I give IBM credit for is they rethought this and said, look, where is this going? And yes, we need to basically bring our existing user base, our Cognos install base, up to date with this. But we also need to reach out to the new generation of, I should say, distributed apps, distributed data, distributed intelligence. And so therefore, let's make this embeddable. This doesn't have to be its own application. IBM is not going to be in the BI... I mean, yes, they're in the BI tools business with Cognos, but that's yesterday's business. Today we need intelligence on the data that we're getting and it's not just essentially the traditional analytics, but we need to use machine learning and in some cases generative to extend this and add more broader context.

>> You're saying it's a different... You're right-

Tony Baer

>> It's a different world....

>> back then when IBM bought Cognos it was part of an IBM transformation.

Tony Baer

>> Yes.

>> Steve Mills was architecting. Now you're saying it's using AI to enable business insights or maybe even eventually business actions. And that's the IBM sort of value add here. Is that right?

Alvin Francis

>> Well, that's part of it. So just building on that concept of business actions, there is a chasm that exists today between the insights and actions. A lot of organizations believe that because they derive insights from the data, they're leveraging these insights to make business decision or take actions. Part of our strategy is actually to integrate watsonx BI Assistant with watson Orchestrate. So directly from the user experience, insights can be used to automatically drive actions and close that chasm between insights and actions.

>> I love that, by the way. And this ships when?

Alvin Francis

>> It's Q3. So we are actually signing up customers right now for private preview, and we had, as you can imagine, a lot of traffic at the booth. A lot of customers have signed up. We'll co-create with a number of our key customers and we will continue to ensure that we are obviously addressing the use cases. You talk about how close we are in terms of delivering on the promise, but we want to make sure that we're working closely with our key customers, identifying some of the key pinpoints, solving it, and so that when we in Q3, we're going to hit it out of park.

>> So let's say it ships this fall a year from then, let's say think 2025, what do you want to be able to say in May of 2025 at IBM Think, if that's when it is, that you can't say today day?

Alvin Francis

>> So our focus is obviously to drive adoption. And so the way we will measure success and hopefully at Think next year is to look at how organizations are adopting BI Assistant to actually transform the entire BI organization and how they leverage data to make decision. So what I can't say today obviously is the number of customers that we will secure, but at Think next year, we should be able to pop the champagne in terms of saying that we've been successful. Obviously this product will continue to evolve. There's still a lot to be done in terms of how we transform BI. We are fairly confident that we're on the right path and it'll evolve from a capability perspective, leveraging maybe more powerful models, in some cases smaller models that will integrate to address specific use cases. But by Think next year we should have obviously continued to lead the curve and be differentiated.

>> And real quick, Tony, what do you want to be able to say a year from now in this space that you can't say today?

Tony Baer

>> What I want to see essentially is that where we're doing... I'm thinking a lot of the stuff that you're talking about with the Orchestrate, where we're talking turning insight to action. You guys are promising that, I want to see that as reality next year.

>> Guys, great analysis. Congratulations on the announcement. Thank you so much for coming on, theCUBE.

Tony Baer

>> Thanks Dave.

Alvin Francis

>> Thank you very much. Thanks for having us.

Tony Baer

>> Good talking .

>> We're wrapping up. You can hear the conference hall is loud. People is getting ready to go out tonight. This is Dave Vellante. Keep it right there. More action from IBM Think 2024 you're watching theCUBE, the leader in enterprise tech coverage. Right back.