The impact of training and learning in IFF’s journey to become a data-driven, global enterprise.
IFF supplies end-markets with “innovative solutions that allow them to create the products consumers know and love.” In 2021, it merged with DuPont Nutrition and Biosciences, becoming a 12-billion dollar global enterprise. IFF’s centralized R&D function has over 1600 people working at ten different sites across the food and beverage, fragrance, home and personal care, and health and wellness sectors. To stay relevant across such a broad scope of industries as a global enterprise, IFF needed to exploit all the benefits of digital technology.
The company’s Digital transformation and global R&D lead, Dorthe Malmqvist, recently spoke with GoDataDriven about the impact of training and learning in IFF’s journey to become a data-driven, global enterprise.
Q: Where did you start the transformation? On the people- or tools- side?
We started with our people — specifically our scientists. Within R&D, we’re a super-specialized organization. Our scientists are the key players. We look at it as a change management effort because the only ones who actually know what’s going to help our scientists do better are the scientists. We needed to find out how to motivate and inspire them to drive this forward as an integrated part of everything we do. The scientists needed to own it because it was about their area of expertise. So, then it was the leadership’s responsibility to build a framework to enable the scientists. Part of that framework is training, so we created the Digital Advanced Analytics Academy.
Q: Many people at IFF have been there for a long time. Have you had trouble convincing them of the change of direction regarding data & analytics?
Very early in the process, one of our very senior people thought this data democratization was a little scary. But it brings about a lot of transparency: in R&D, we focus on developing new products and methods. We can do that by learning fast.
As soon as our senior people saw how quickly we could generate knowledge and insights through data & analytics, they were all in. They didn’t question the “why,” We formulated it clearly from the beginning, so our direction was clear from the start.
Interestingly, the initial push in this new direction didn’t come from the leadership but the biotech team. It’s a very data-heavy area, so biotech was the first to recognize the potential value of data and its business case. For them, the starting point was healthier data centrally stored instead of little islands of Excel all over the globe.
Q: Was it difficult to get people off Excel?
Yes, it was challenging. Many people said, “Well, I’m only testing things out by myself.” And it’s true; it’s an excellent tool for these tasks. But as soon as it takes shape, you must get out of Excel. It’s super easy to work with in the short term, but it’s very clunky and negative to keep using it for everything over the long term. We’re getting better at showing the value of robust solutions because domain areas with healthy data move forward faster.
Q: You mentioned training: what training do you offer and for whom?
We started with the minimum number of roles needed to drive these solutions forward. We began with leadership, making sure they would walk the talk. We wanted them to be champions in words and actions and to understand what it’s all about.
Next, we focused on the business translators, anyone that could take an idea and turn it into a data product. They run the projects for us, interpret the results, and so on.
Lastly, we focused on our data professionals.
We have a lot of data scientists across the globe who don’t always work together as a community. But by putting them in classes together, they now feel more connected and encouraged to reach out to each other. Training them has created alignment between colleagues who can now learn with and from each other. It made a major difference.
Also, everyone should understand the world of data in general. So we added an inclusive “crash course” to ensure everyone understood our direction.
Q: You started your data journey a few years ago. In the meantime, Covid-19 struck, making it difficult to get people in the same room, which was very problematic for the training industry. How did IFF handle it?
IFF is, in many ways, a global company, so we’re used to working remotely. It may not allow for the same kind of connection and regular contact, but operationally, remote working allows us to get going faster and create new bonds around the globe.
What’s more, we get very high ratings on our courses. The average satisfaction level is 4.1 (out of 5), so I’d say avoiding travel was a win for us.
Q: High ratings are one way to look at it, but how do you measure the broader impact of learning?
We’re measuring the impact in two ways. One, we take a survey six months after people complete the training. We ask them whether they’ve applied the learnings, and that score is pretty impressive.
We also measured the number of analytics project proposals we’ve received since implementing the academy and saw that the numbers have doubled. The quality has also greatly improved. In the past, the proposals were not really data and analytics oriented. Now, we have trouble assigning resources because they’re all so exciting!
My vision was to have eyes, ears, heads, and hearts in every corner of our organization on the lookout for data and analytics opportunities — and then bring them to the core team’s attention so our organization could continually improve. I am seeing that vision realized!
Q: Do you have any tips? Or things that were harder than you expected?
When we started, we explained the big “why” to everyone, setting up lots of training sessions. Today, the appetite for training is still phenomenal: we have waiting lists for the courses for almost the entire year — and this is the third year of the academy. So, that’s a good sign. Participants are going back to their teams, talking about the training, and saying it was a good investment of their time — and some training courses are 32 hours, so it’s not exactly something you do during lunch.
But the biggest challenge is to keep the learning alive. You need to support people and make sure they start applying what they’ve learned as soon as possible.
Q: What is the role of early adopters, and how do you capitalize on them to ensure the potential and momentum are not wasted?
We made the most of our visionaries, especially our super-engaged data scientist in the Netherlands. He had ambitions at the corporate level as soon as he came on board — perhaps for the first time. He spent a lot of time cleaning up other people’s data and knew we could do much better.
So, we capitalized on people like him — the leaders who dared to step up to the topic and task — even when they didn’t have a data background. We also started hiring people with data competencies who became part of the core team. It’s phenomenal to have these people.
Q: These people often bring in tools or tech they know and use to replace things such as Excel. In many companies, IT is afraid or wary of new tools. How did it work out for IFF?
In the beginning, we struggled with IT. Getting the foundation and the infrastructure in place was a big undertaking. However, that changes pretty quickly. R&D is the value creation engine of our company, so it was easier to voice our needs and be heard. Now the IT organization is really supporting us.
Q: Do you have suggestions or a take-home message for companies that want to embark on a similar journey?
Yes, I have a few of them.
First, start with the “why,” as it serves as guidance for everything you will do.
Lay out the strategy; that’s the vision with a concrete roadmap describing how you will achieve it. Include learning programs and how to measure them. And, be explicit with your leadership: they should recognize the contributions of the people who invest in this journey. All of our data scientists need time to focus, so how will data and analytics help them do a better job?
Q: What do you see as the future of IFF in these areas?
Well, first, we’re expanding the academy. After the initial data science training, our scientists explored more advanced techniques, such as deep learning. So we now offer a course about deep learning applied to Natural Language Processing [NLP, very relevant in the field we operate].
We are also embarking on a more extensive journey with data management: we have islands of digital solutions, but we want to ensure everything fits and can work together. We also realize there’s much more we can do to help differentiate and keep excelling.
Questions from the audience:
Q: How do you measure the adoption of the learnings, and what is the level of waste?
As I mentioned, we measure adoption in two ways; one is by the number of project proposals involving data & analytics that we receive. These are important for developing the data products our organization needs to improve. We also measure adoption with a questionnaire six months after each training session. It asks participants if they have applied what they learned in their everyday work. What we see from their responses is yes, they are using these new skills in their daily tasks.
As for the waste, we assigned people for the first iteration of each course instead of letting them subscribe. We wanted to start quickly, get feedback, improve and iterate.
So, in the beginning, it was not a perfect fit for everyone, but people were free to join other courses afterward, and this continues. It means we have people that follow the data science course, then take the business translator course and the other way around.
This approach has the added advantage that it doesn’t scare anyone away.
And to increase inclusiveness, we opened up the program not just to R&D but also to operations, that is, people working at our manufacturing sites. Long story short: we had some waste, but you must accept that. Just like our business unit leaders are critical for our success, they need to accept that we spend a certain amount of time on these efforts instead of their R&D projects.
We were transparent from the beginning and encouraged them to attend our data and analytics leadership training to get them on board.
Q: What is the purpose of this journey?
Essentially, it enables us to do better, faster, new R&D for people and the planet.
Additionally, as we explained to our people, we wanted to stay relevant as a business and on an individual level. That’s why we needed to embark on this journey. We needed to accelerate. I don’t know if it was easier with R&D people because that’s what they always do, but I think they have been highly receptive.
Q: R&D-heavy organizations are usually secretive when it comes to sharing knowledge. However, you started sharing what IFF was doing with other companies at the beginning of its journey, and these companies shared their plans and pains. Can you tell us a bit more about that?
In some areas, we are secretive because we deal with trade secrets. We can’t just share them around. But this transformation is about people and processes. We think we need to expand this ecosystem, and some of our external vendors helped us connect with some very interesting companies that helped us on our way.
That’s why we’re more than happy to give back.
Q: In my company, C-level executives set the “why,” and they have an idea about the “what” and “how.” But they delegate execution, and some misalignment ensues in effort, budget, and resources. How can we deal with that?
You go back to your executives and tell them you can’t do the “why” without the rest. You tell them it’s not possible to achieve their goals without funding. And if they’re not going to fund, it’s not prioritized or strategic. That’s why you need leaders who support this in words and action.
Q: When embarking on this journey at an R&D-heavy company, how do we ensure leadership starts with people instead of trying to do everything simultaneously?
If they’re R&D heavy, it should be pretty obvious. We have unique scientists who need to buy into any tool you’re trying to implement. If they don’t, they won’t adopt it. So you can’t just change them. From an R&D perspective, it’s then easy. Other functions may use more standardized solutions, so you need less focus on the people’s side. For instance, you might have 80% standardized and 20% customized in an HR-focused organization. But we’re the opposite: there is no way around those scientists. They need to drive it and have it be an integrated part of it.
Data Literacy Enables Better Decision-Making
Organizations underwrite the need to extend data-driven operations to develop a sustainable business. This need requires people in every aspect of the business to develop their proficiency to leverage data.
Get in Touch!
Contact Marcel van Denderen, if you want to know more about the possibilities to tackle the challenges in your organization with Data Literacy. He’ll be happy to help you!