Chief Content Officer,
Techstrong Group


In this edition of the Digital CxO Leadership Insights series, Mike Vizard talks to Stefan Sigg, chief product officer for Software AG, about why, despite economic uncertainty, organizations are doubling down on digital business transformation.



Mike Vizard: Hello, and welcome to the latest edition of the Digital CxO Leadership Insights series. I’m your host, Mike Vizard. Today we’re joined by Stefan Sigg, who is chief product officer for Software AG. And we’re talking about a survey they did that kind of highlights everybody’s priorities going into 2023. And what we should expect, Stefan, welcome to the show.

Stefan Sigg: Hey, thank you for having me, Michael.

Mike Vizard: The general expectation is that the economy is soft, and everybody would be pulling in their horns. But your survey seems to suggest that a lot of folks are doubling down on digital transformation. So what’s your sense of where are we in terms of the priority level being applied to these types of projects?

Stefan Sigg: Look, what we see and what the survey also is telling us is that there is no slowdown in competition and competitiveness. And, you know, it’s ever increasing how companies try to find their way of differentiation and competitiveness. And, you know, there are so many ways to do that. But one is clear; there is a common denominator – all those attempts to beat competition, be it the competition that everybody knew from the incumbents, or the new type of competition that comes out of software’s or startups, all is around using data to digitize the product portfolio to connect the product portfolio. In a sense, it is all converging to the question: How can I as an industry player, use software and data to become more competitive to define more business ways to be successful? Now? Can I can I avoid that somebody else? Pasta in adopting software and data, and, you know, outpacing me in on my journey of becoming more digital?

Mike Vizard: Are you seeing folks maybe narrowing the number of projects that they’re trying to drive because there’s nothing like a crisis to focus the mind. And after COVID-19, we saw everybody throw everything at the wall, but maybe now they’re getting behind a couple of projects that seem to be more?

Stefan Sigg: So what we actually are seeing is that companies question a little bit all those projects, where they assumed, oh, I can do it on my own. So that has been, you know, couple of years back, what was very common is that the companies would hire like 10, 20 or 40 software developers and would tell them to do something around digitalization, do something around leveraging better the data that I have, and they would then, hire some infrastructure in the cloud and just get crazy in developing something that usually ended up in failing projects, because the developers got lost. They produced a lot of costs on the cloud side, on the infrastructure side. And it took a long time until the sweet spot came on the horizon where the actual differentiating thing – the industry and companies specific things. So they got developed. So what we’re seeing now is companies coming to us and saying, “Well, you know, all that of, you know, connecting applications, connecting systems, connecting devices, all the integration of data, you know, the ground works is all there, there’s no point re-coding that, you know, on the cloud, we rather take what you guys have, and we concentrate as an industry as a company in and on the last mile of getting our domain expertise on top of, you know, the cloud pipelining, the API management or the integration of systems that that I need to lay the foundation of my new applications.” That said, certainly what we are seeing is a kind of a rationalization in terms of really thinking, what is really the next new thing that I as a company can provide, and what is out there, you know, which is to a certain degree a commodity, certainly something which somebody has thought about programming and coming up with in a way that it is not only there, but it’s also multi-cloud that it is scalable, that it is secure. So these non-functional capabilities of pieces of software that you usually do not pay attention to if you have home grown projects, but eventually will be a big part of your technical debt that falls on you back.

Mike Vizard: Speaking of technical debt, is that out of control? We’ve always had technical debt, but do you think we’re approaching some level of crisis around that particular issue? And what can be done?

Stefan Sigg: Well, I think if you’re not a longer term, software engineering practice, then the trap falling into, in terms of technical debt, that is very dangerous, because you want to see quick results, you want to see quick, successes, and I think only if you plan for a reuse and plan for distributing your deployment many, many times, you think about these kinds of things, at the get-go. As I said before the typical custom build solutions, yeah, they tend to not take care about that; they tend to result into software stacks that are hardwired within given clouds, cloud tech – they tend to be you know, ignorant, ignorant to you know, these things like the abilities of a software stack, in terms of scale, performance, security, accessibility – things like that. So, this is what, in many cases, customers realize only later. And then now what we try to convey is, there is not any more this black and white world. I either buy or I built, it must be both; it’s a combination of builds, and I built the stuff that whether it is a solid platform, a solid stack that I can reuse, maybe it even gives me the possibility to switch cloud stacks – and then I built the stuff that nobody else knows how to build and how to do. And this combination, I think is what we try to get across to our customers – partner with our customers together, with our partners, so that we have both sides of the world a professional software engineer foundation that is built for reuse and built for independency and sovereignty and then the necessary openness that lets the domain expert as a customer come up with, as I said, the last mile to make that truly a solution. And this combination, I think is the way to go, and you know we have now seen so many decades worth where companies, the IT department of companies became more or less orchestrators of standard or standard software, you know, customizing standard software. Now, if you go very far into the history, you would have this kind of software building skill even in non technical companies, but then it got this mass or got a little bit lost over time but now we as a company stand in front of challenges where there is a new need for new domain specific software; software that is not available as standard software, software that you need to make your product smart, that you need to make your services smart, and they are missing clearly the combination of a solid foundation with a standard openness And then the customer spending maybe half or a third of the developer workforce to finalize the solution.

Mike Vizard: As organizations focus more on where they can add value, specifically versus reinventing the wheel, are there particular technologies that you’re seeing that are gaining more traction than others?

Stefan Sigg: Yeah, so I think customers nowadays, they expect from their software members that the software they are providing is adaptive to the current state of the art of cloud deployment, right? So I think it’s not sufficient anymore to have an installable piece of software, put it into a container and claim to be this cloud software. That is not state of the art right now. I think, even if companies continue to run on prem data centers, they do that in a very similar way, as the cloud infrastructures, have that in professional childhood as well, meaning Kubernetes clusters, and deployment of containers in Kubernetes clusters rather than having you know, installers running and making hardwired connections to the infrastructure. So, that is one part that I think is irreversible. Then, of course, and I talked about it already, the openness is very important. And the ability to integrate with existing assets around the domain that you want – that is so important. And I think the the integration that we are seeing is where there is a strong demand integration into applications. So you surely have an SMP system around, you surely have a theory around, surely have an HR system around, but then also, if we more going into supply chain or production environments, you know, there’s a whole world of MES systems, of scheduling systems, of of logistics systems out there. So whatever companies aspire to become in terms of becoming a software company, it is not an isolation; there are a lot of integration points that need to be connected. And then of course, nothing, nothing exists without a professional way to move data, to integrate data, to establish data pipelines. Typically, data pipelines into cloud based data lakes, which are cost effective ways to store big amounts of data, and then, you know, having a solid way to get data out of it again, for those typical analytics purposes. So there’s a convergence of getting the integration points connected and the data integrated with a strong momentum. And that’s, by the way, the reason why we have brought up our our friends from StreamSets on board, to live up to the synergies of you know, traditional application integration middleware, and modern data integration, based on the type of data that nowadays is more and more dominating; maybe not so much perfect as much as relational, but much more, much more semi structured data, the whole world of JSON data, for example, is becoming more prevalent every day. And what is the final word on a concept that I really talk to customers a lot about is the notion of system of records. You know, over many, many years, more or less, the ERP was the dominant system of record within organizations. But nowadays, more and more, you know, you find very important datasets that are not managed by earpieces. Alright. And that brings up the question. So what is my new kind of system of record? Think, for example, about the whole world of ERP. Yeah, so it’s only a fraction of the data is really managed in an ERP, the other parts of the data is coming from whatever, and you need to have very strong data integration capabilities to at least integrate the data as if it was a system of records. And then take it from there and build your solutions.

Mike Vizard: You cannot walk down the street these days without somebody talking about AI in ChatGPT. So what’s your sense of what the impact of this will be as we go to build applications?

Stefan Sigg: So I mean, it’s mind boggling. I mean, we have been very fast. I think we were the first in our industry to come up with a ChatGPT connector that is really usable by dragging and dropping integration into integration flows. And we have been discussing that with early customers and customer advisory boards. And the number of ideas were just, you know, coming, like without demand. So and what you’re using there is, you know, one very clear use case is the incredible capabilities of ChatGPT type of AI to come up with text with language. So, many times you have the task to craft a good text out of some contextual data, and make it a ticket for example. Yeah. So that is totally automatable, Yeah, so you drag and drop your connector, you feed it with some keywords out of a, let’s say, an insight from an analytics task. And you ask it, “Okay, now write a ticket statement about this issue that we have found integrated with ServiceNow.” Or, you know, send out invitations or give contextual fitting feedback to customers who have answered an NPS. So there are so many cases where a generated API connector plays a fantastic role in, in very natural integration flows. And we have only seen the tip of the iceberg here. We’ve only seen this and, if you will, now, we have now a new angle: Looking at what we do in integration. So usually, integration is also a certain type of asking, answering a question. And one of the interesting tedious tasks of making two applications talk to each other, which are not designed for talking to each other is finding out a good way to map certain subsets of the fields. Well, guess what? There’s a high probability that you know, the fields from a system and the field pharmacists already have been integrated, somewhere out there, and somebody has done it and wrote something in GitHub or Stack Overflow and you can find that. So rather than having a question level AI on automatic. So the field X is the field somebody, and can you please suggest to me how to connect and how to map those? And the results are incredible. So you see, we are generating almost everyday new use cases where this little connector plays a decisive role in new types of integration flows, the data or application integration.

Mike Vizard: So do you think we’re approaching a moment where maybe the pace of innovation is faster than businesses can absorb?

Stefan Sigg: Look, I think in the end, it all boils down to relatively simple principles. What do I mean? What do I mean by that? Take for example the notion of getting a job done. When do people adopt products or organizations to adopt products? And when do they dismiss products and join or hire a new product? So this is all when they figure out there is a better way to get the job done. And nowadays, you know, AI is is a reason why new solutions can get the job. There is a way to get a job better than before. Because you can reach out to this gigantic world of existing knowledge and ask a question. And what you get back is not a bunch of websites, but actually bands. That’s what you were looking for, right? You’re not asking Google to give me a bunch of websites; you want to have an answer. And it’s only now that you realize how many times did you use Google and not to find a website but to get an answer? And that is what ChatGPT and generated AI tools can do. So much data. And I bet you that there is a whole plethora of jobs to be done, that can be done better. Because you know, you have this ability to get answers, definite answers to questions where the assumption is somebody did it already out there. But I think the speed of adoption is given by how convincing is a new solution for getting the job done.

Mike Vizard: All right, folks. Well, you heard it here. The pace of innovation is picking up and we all gotta adjust one way or another. Stefan, thanks for being on the show.

Stefan Sigg: Thank you very much. Mike. It’s been a pleasure.

Mike Vizard: And thank you all for watching the latest edition of the Digital CxO Leadership Insights series. I’m your host, Mike Vizard. You can find this and other episodes on the website. We invite you to check them all out. And once again, thanks for watching.