International Paint (AkzoNobel)

The company

AkzoNobel is the global firm behind recognised brands such as Dulux, Sikkens and Interpon powder coatings. As a leader in the field of paint and coatings, it is active in a variety of industries, including aerospace, high value infrastructure, consumer products, interior decoration, mining, and power.

Through its subsidiary company International Paint, AkzoNobel's offering also extends into the marine, protective and yacht markets comprising industries like shipping, offshore wind and oil and gas. It employs 35,000 people across the world and operates in more than 80 countries.

International Paint is on a mission to enhance its data maturity.

The challenge

At the heart of International Paint's pursuit for innovation lies a keen interest in pushing the boundaries of its data science capabilities. Marie Dale, Data Portfolio Leader within the research and development team, sees a distinct challenge in leveraging the company's solid foundation in data to explore new technologies that promise long-term value for their products and the organisation. Despite their existing competencies, International Paint grapples with the conundrum of needing to prove the value of further investments in data science to secure additional resources, while simultaneously requiring those very resources to uncover and demonstrate potential value.

Marie defines the problem "We are on a journey of data maturation, trying to become more mature with our use of data... It is a cycle between needing more people to do more work, but to justify more people you must demonstrate value"

To address these, challenges, Marie explored the potential of outsourcing some of the data science work, all while ensuring that those external partners could quickly and efficiently gain the domain knowledge necessary to deliver impactful results.

The goal

Reflecting on the goal of the project, Marie said "The focus was on how we would manage external data science contractors to get the most value out of them. We wanted to create a framework that identifies the problem and defines the process for importing, cleaning, and wrangling data. Our goal was learning how best to share domain knowledge with an external person as efficiently as possible".

The project aimed to develop a robust framework for managing external data science contractors, effectively. The framework would ensure external teams could deliver high-quality data science work with minimal time spent on acquiring domain knowledge.

The objectives were to:

• Enhance data documentation and governance practices.
• Foster ethical and transparent use of data.
• Implement a data-driven scrum process for project management.
• Facilitate skills transfer and upskilling with International Paint's R&D data science team.

The result

In collaboration with the National Innovation Centre for Data, International Paint created a comprehensive process framework that not only streamlined external collaborations but also created a significant internal impact. The project emphasised soft skills such as data documentation, ethical data use, and the adoption of data-driven scrum methodology.

The approach taken improved project efficiency and contributed to a deeper understanding and trust in data processes.

The tangible outcomes of increased data maturity may unfold over years, but the foundation laid by this project is expected to yield significant benefits in terms of operational efficiency, innovation, and decision-making.

Building a data maturity framework

Data Science Tools and Techniques

To support International Paint with this problem, the team at the National Innovation Centre for Data took a different approach to their normal delivery.

International Paint's plan for utilising the project outcomes with the National Innovation Centre for Data focuses on enhancing data maturity and streamlining external collaboration.

The project started by delivering the Success with Data Science workshops followed by a discussion on how the methods could be adapted to fit their business practices.

The final step was for a member of the National Innovation Centre for Data technical team to act as a contractor to test out the workflow and identify ways it could be improved.

Reflecting on the project, Marie said "We've adopted some of these learnings and embedded them in internal policies to enable that skill transfer amongst our wider data community" broadening the project's impact beyond its initial scope with the aim to upskill teams with new methodologies and insights.

Marie also notes the significance of rolling these practices out to wider teams, mentioning that "we've got some of our colleagues coming over from Singapore in two weeks' time. We have started to share some of this stuff with them and have been working through it together” stressing the intention to foster a collaborative environment where the skills and frameworks developed through the project are shared across different segments of the organisation.

The emphasis on iterative learning and application is evident "As we've been doing the project we've been implementing things as we've gone on and evaluating what works well. For others we have returned and realised we need to tweak certain aspects or change how we do things" showing an agile approach.

Marie also anticipates future collaborations and skill-building opportunities. "We have other projects that we’d like to partner with the National Innovation Centre for Data on. The focus on upskilling means that we can make sure that someone within the data science team has the skills apply learnings to multiple different projects in the long term.”

Highlighting the benefits of skills transfer for data maturation in large organisations, Marie said "Ethics and transparency around using data was a big thing. Evaluation checklists that considered legal and ethical compliance elements of the data modelling were an important learning from this project. It is not just about teaching others how to build and run a data model, but also teaching them how to evaluate a model.”