Chris Hansen, Chief Corporate Development Officer of Clariant, provides insight into Clariant’s digital initiatives and AI utilization strategies. A fundamental focus on system globalization, data integration and master data quality gave Clariant a solid foundation for implementing powerful generative AI tools such as Clarita, their internal large language model. Chris sees more potential still in further leveraging these technologies across the business to connect past insights with present actions to optimize for the future.
Read the Transcript.
Ali Amin-Javaheri: Hello and welcome everyone. I’m your host, Ali, CEO and co-founder of Knowde. Our goal with this series is to share perspectives on the changing digital landscape in our industry. I’m thrilled to welcome Chris Hansen, Chief Corporate Development Officer at Clariant.
As you all know Clariant, I won’t go into it too much. They’re one of the world’s leading specialty chemical companies. They pride themselves on an overarching purpose of greater chemistry between people and the planet. They provide innovative and sustainable solutions to many customers all around the world across many different industries.
Chris heads corporate strategy, procurement, investments, operational excellence and digital initiatives for the company. I don’t know if there are any other roles, but I think we’ve covered them all.
Chris, thank you so much for making the time. Before we jump into today’s conversation, I’d love for you to share a little bit more about your background.
Chris Hansen: Hi Ali, very glad to. First of all, thank you very much for having me here.
I’m originally from Denmark. I started my career in management consulting and spent five years with McKinsey before joining Clariant in 2009. I actually just celebrated my 15th anniversary with Clariant.
In my first years at Clariant I developed and rolled out our excellence programs across the company’s business units. In 2013 I got the chance to move to Asia—first to Shanghai and later to Singapore—to lead one of our business units there. In 2020 I returned to our headquarters here in Switzerland and led another business unit for a couple of years until I took on my current role.
As you just said, it’s a mix of things. Often I refer to myself as ‘the Chief of Other’ since I’m now overseeing what no one else did. The role includes ELT (executive leadership team) and overseeing our corporate strategy, procurement, company investments, operational excellence and digital initiatives. Digital and genAI have really become my favorite hobby in recent years.
Ali: I would imagine you spend a disproportionate amount of your time on all the digital stuff — at least that’s what we’re going to hone in on today.
I’ve known Clariant for many years, and I would say that they are one of the leaders in adopting all things IT and data management. Can you talk to us about some of the systems that Clariant has in place and how the company has evolved from this perspective over the years?
Chris: I think at Clariant we were early to recognize the importance of a robust IT infrastructure and data management. Our journey really began almost three decades ago — pretty much with the creation of the company — and we rolled out SAP globally. For more than 10 years now we have had one global SAP instance and a global data lake, which is really our single source of truth. It sounds simple — but with my consulting background and in talking to friends, I’m aware that having one single source of truth in the company when you look at the numbers is not always a given.
For a number of years now, all of our data and applications have been in the cloud. That’s something we’ve done together with AWS. I think it’s important to say that the motivation was always a strive for simplicity and cost efficiency combined with looking to really understand and address the needs of the business. I don’t think we ever introduced new technologies out of an aim of being at the technological forefront — it was always because they supported our strategy, they supported simplicity and cost efficiency, and they addressed business needs.
I think the exciting thing is that this strategy and decades of hard work by our IT colleagues have put us in a very nice situation when it comes to introducing and leveraging generative AI in our company now — particularly, the fact that we have global systems and all our data joined in the cloud really gives us a good starting point.
Ali: I’ve been doing this for a long time and I don’t know that I’ve ever heard a producer say that they have a single source of truth. Just from that perspective alone, I applaud you.
For other companies that are looking to move in that direction and get their data in a better place, any sort of advice you would give — any sort of first steps?
Chris: I think the best course will always depend on a given company’s context and strategy, and of course, on the company’s legacy IT environment. A lot of companies, especially in our industry, have gone through a lot of acquisitions and as a consequence of that they end up with very fragmented IT landscapes. We have been in a situation over the past decade where we have — fortunately or unfortunately — sold more businesses than we have acquired and that has made this a little simpler for us as well.
My general advice — irrespective of the starting point — would be to really focus on data integration into a data lake or the like and to work on the quality of your master data and other data. From what I hear around the industry there are some very flexible solutions out there nowadays that can enable this in a way that wasn’t possible even a few years ago—so that this legacy IT architecture that is often a stumbling block early on becomes less of a hindrance.
In the end, it’s still quality data that is key for quality insights and for an IT-enabled impact. To hone down my recommendation, it would be to really invest in data quality and then to leverage it. That is probably the most important thing that I hear. When it comes to leveraging our data and leveraging our technology, it comes down to people and change management. I’ve seen so many times where a lot of money is put into exciting and promising technology without creating the desired impact, because people need to be taken along in this. People need to overcome their fears and their reservations. In the end, they simply just need to change their habits. We are all humans and I think change is hard for most of us. We need to change how we work—how we do things.
Ali: I’m thrilled that you mentioned all that. Across the industry I’ve seen so many digital projects fail because of the reasons you just suggested—they see some shiny object and they want to bring it into the organization, but they forget about laying the foundation correctly in order to be able to enable said systems and make sure that people adopt said systems.
We’re constantly evangelizing around the industry that if you want to transform your org or implement this system or that system, it has to start with the data—because if that’s not there, I don’t know what you’re going to push into the system. You’d be surprised as to how many people miss that. But it’s also an area of great complexity. If it was easy to fix the data problem and get it to the place that Clariant has it, people would have already done it. It’s hard—it’s hard.
I know that we share a passion for the power and potential of genAI. I’ve seen a lot of your LinkedIn posts. I know how proud you are of what you’re developing. Again, I applaud it because I think you’re one of very few companies that has stepped out and said, we want to take an ownership in this space and we want to lead—it may not be perfect on day one, but we have the courage to step out and do it.
And so I’d love for everyone listening to hear about Clarita.
Chris: Of course, I hoped you would ask that question.
So the question is, who is Clarita? I’d say for many people at Clariant today, they talk about Clarita as being their favorite colleague.
Clarita is our internal large language model. She’s based on Claude. In the end, she’s simply designed to help everyone in their daily work lives and even in their daily private lives. She’s competent, she’s knowledgeable, and she’s always in a good mood no matter what you throw at her.
We conceived Clarita in response to the emergence of powerful generative AI tools in late 2022, particularly the launch of chatGPT. I think all of us remember this sort of ‘wow’ moment. I remember over my Christmas break that year thinking about how this could be transformational and the enormous opportunities that this technology could really present, but also already reflecting on the risks—data protection risks, particularly—in simply using these public AI platforms.
This was early on in discussion; we wanted to do something. After that Christmas break I came back and talked to our CEO and said, “listen, I would like to take this forward” and he said, “do that.”
And there we went. We talked a lot about it early on, but it was quite clear to us that we wanted three criteria covered here. The first was security, the second was fast implementation, and the third was flexibility—we wanted something that could be adapted going forward.
On the security side, it was very interesting because we knew nothing—not in the beginning. So I thought, okay, let me reach out to consultants. In early ’23 I talked to a lot of consultants across the industry and in different industries about the ideas and everyone was very eager to support. No one knew very much at that point. What struck us was that very few took the security concerns seriously enough in the beginning. It was more a notion that either you have to wait or you need to accept that you share your IP and other confidential information with the world. That was a no-go because our IP is of crucial value to us and we would not want to risk that getting out into the public space.
For implementation speed I like to think in terms of weeks, not months or years. We wanted to get going. I already mentioned flexibility—we knew this was just the beginning, right? We needed something that could be developed upon when the technology developed and new use cases emerged.
One more thing to add: we wanted this to be available to all of our employees. This was fundamental in the beginning. We did not want to create a small pilot with a small group and then never get to impact. So our fundamental criteria was that we wanted this open to everyone from day one. In the end we found a good partner in AWS. We had worked with them for years on moving everything into the cloud. We got going with them I think in June and six weeks later Clarita was there as a solution, accessible to everyone—including all the boring stuff. This is always what takes the longest: access rights, single sign on, all this that I normally refer to as a ‘plumbing’—but it needs to be there to make sure everything works.
It was interesting: the first solution was clearly inferior to what people could do with chatGPT at that time. But we put in place a lot of opportunities to actually collect feedback and we caught up quite quickly. We realized really how important communication, training, and change management more broadly were also in this case.
That is also what led us to name our genAI solution Clarita—to give her an amicable face. It was about mitigating the fear and the hesitancy of dealing with yet another tool and instead make this something fun and spark the curiosity of colleagues.
I think in hindsight this was a super important thing for for the adoption and acceptance of our genAI solution.
Ali: I get that you purposefully opened up to the entire organization. Were there certain functions that you wanted to achieve something for first? For instance, I’ve heard the use case around making the field sales team more productive by putting the product catalog and product knowledge at their fingertips.
Did you think through these sort of use cases?
Chris: I would love to say that we did, but honestly in the beginning we said, let’s get it out there and let everyone play around and try this out. The strategy and the structure developed over the first six months, really—then our thoughts around the whole thing became much clearer.
One of the things that came out of that was some specific solutions. So beyond what we call now the ‘core Clarita,’ which is the general genAI chatbot, we have specific solutions for different employee groups. We have our Clarita sales assistant, and what she does is link all the relevant customer, order, and product information for our sales colleagues so they have everything in one place. She can find the different pieces of information that they need and help them prepare for customer meetings super efficiently in a high quality way.
It’s also interesting—this was our vision, what we really wanted: we wanted our salespeople to be able to treat any customer like a key account by having all this information. It works, and people are doing this.
But what really got the buy-in was the fact that she also helps them complete some very mundane tasks that they hate. Our salespeople need to follow up on overdues regularly. This is not a preferred task and it used to be complicated: going into the systems, finding the data, and so on. Now they just ask Clarita to give them an overview and put that into an email that they just need to copy into Outlook, and it’s done—and the data quality and everything is much better than it was before.
So it’s these small things that are actually much simpler tasks that help build credibility as well. Then people start using Clarita for the next thing, and they open up and say, this is actually helpful for me.
Ali: So you’re starting to feed order data, CRM data, and all of that into the system at this point?
Chris: This is all in our sales solution, and our general Clarita nowadays has access to anything that you have on your own OneDrive: from your emails to any SharePoints you have access to.
So for many people, this is also just a super productive search tool: “I think I had a document talking about this and that two years ago, please help me find this” —and she’ll come up, talk about what this was, and give you a reference to your documents. You just click a link and there it is. How you do it takes some getting used to, but it is super, super powerful.
Another way we utilize data and some of the different tools for the R&D community is the R&D assistant we have in place. We have a so-called ‘rack solution’ that links up to better documents—internal and external documents that contain all the knowledge we have on our core technologies. By integrating that, we really support our scientists in their product innovation and troubleshooting.
A third way of utilizing data is on the operations side. Here, we have vast amounts of data that we have collected over many years and thousands of dashboards. Clarita is integrating all this and translating it into forward-looking recommendations—otherwise, a lot of time is spent looking at the past while the present is just happening without being optimized. So it’s super, super powerful.
In operations, I would also say that the change management task is very large for us. We have around 70 sites globally, and you need to be there with the people. You cannot just send out an email with a link and say, “Here’s the tool, use it.” You need to be there, and you need to adopt it individually. So there’s a major task here, and it will take us years to have it in place everywhere.
Ali: That’s amazing. I think your vision for where you’re taking this is spot on, and it should hopefully encourage others to follow suit.
I’m curious to know, what’s getting you excited next? You’ve got your hands full with everything you just mentioned, but is there another big idea that’s brewing in the back of your mind?
Chris: There’s a lot of things brewing in my mind. I think if you really take a couple of steps back, I’m super positive about the potential here and we see a lot more potential still in deploying genAI and AI, especially in our operations and supply chain. We see in many ways Clarita being this integrating interface between the digital world and our people, between past insights and present actions.
And again, back to operations, Clarita is using past knowledge to optimize how we run the next order on the schedule. We initially focused on reducing energy consumption because it’s a very positive topic. How can Clarita help us do this? We have operators sitting in a plant in Indonesia, speaking Bahasa with Clarita in the local language, getting the recommendations, and asking clarifying questions in order to really understand what they have to do. They see the impact. They see directly how they are saving energy and they get the credit for it, which is also very important. It is their achievement. Clarita is not taking the credit for said achievement—she is working as their assistant.
Ali: Thanks so much for these insights. Like I said a couple of times, I think you have sped ahead of many others in this industry and it is so encouraging for me personally to see how much this industry is evolving.
Everyone loves listening to these short informal conversations with people like you who help shape this industry. So thank you for everything you’ve done and continue to push and I’m curious to see where you take Clarita next.
Chris: Thank you very much, Ali. It was a great pleasure.
Ali: Thank you, Chris.
Discover Knowde’s Executive Interview Series
Interested in more insights from the executives driving digital transformation forward within their enterprise organizations? View the full web series, or follow along on Spotify and Apple Podcasts.
