Faglig dypdykk: Hvordan Møller ble datadrevet

Hvis «data eller dø» er sant også for din virksomhet, hvordan unngår du det siste utfallet? Det handler om endring. Her er en historie om datadrevet endring i Møller Mobility Group. Artikkelen er på engelsk.

Deloittes Geoffrey Van IJzendoorn-Joshi, Kari Rødås (tidligere Møller, nå i Ruter) og Hege Line (CTO i Møller Digital) skriver om en omfattende digitaliseringsprosess i Møller Mobility Group.

Foto: Sleepycats (Pixabay CC BY 0)

The change management behind data management

Organisations must be data-driven to survive the new age: it is data or die. 67.9% of major companies in 2019 have a Chief Data Officer to become more data-driven. They thus utilise the role to better enhance data and analytics to drive innovation and business transformation.

Only 31% are data-driven

However, many organisations still struggle to become data-driven.

Only 31% of major companies see themselves as a data-driven organisation. Culture is a major factor: “Cultural challenges remain the biggest obstacle to business adoption.”

Additionally, the Data Management Body of Knowledge also mentions that one must have a collaborative culture and change management capabilities when setting up an organisation focusing on data management.

What is data management?

Data management is the governance and operationalisation of plans, policies and standards, which contribute to an improved data capitalisation and control.

Data management is a prerequisite in order to be data-driven and agile enough to tackle any new requirement on data, such as better data protection, enhanced analytics, machine learning or robotics.

Embedding data management in your organisation is a herculean job. It becomes less herculean when:

  1. Acknowledging that data management is a change management challenge;
  2. Using proven data management frameworks to structure the change; and
  3. Onboarding data management professionals with change management acumen who know how to use these frameworks.

We believe that implementing a data governance without any of these key success factors would end up in a failure.

«We believe that implementing a data governance without any of these key success factors would end up in a failure.»

A benchmark from MMG

This article shares how Møller Mobility Group (MMG) utilised these factors to make the organisation data-driven. By doing so, we also addressed a core pain, a sub optimal data model, and used it as an accelerator to forge a data management dream team.

By sharing our key success factors, we aim to help current and future data management professionals in their journey towards setting up a data management organisation, and to provide a benchmark.

Three enablers of data management

More and more organisations acknowledge the need to be data-driven and to implement a data management team with a Chief Data Officer. One or several data management enablers typically trigger this need. Three such enablers are:

  • customer centricity: the need to improve the service to the customer.
  • compliance: this is historically a powerful enabler. For example the General Data Protection Regulation was a strong compliance enabler for data management. Before the GDPR, the financial sector in Europe had its compliance boost due to the financial crisis, which caused central banks to demand stress test reports with high data quality.
  • performance indicators: A third example of an enabler is the need for an accurate representation of the company’s performance so that leadership can make the right strategic decisions. There is a good reason that CEOs and strategic directors keep Chief Data Officers close for a reason in mature data-driven organisations.

Tandem works best

Data management works best when working in tandem with business, and not as a purely IT-led initiative, as business determines the data management enablers that trigger the need to be data-driven and to set up a data management organisation.

Compliance and customer centricity are MMG’s main enablers for data management. These enablers made data management a burning platform topic, and ensured top leadership buy-in, thus making them business mandated.

The real challenge

Key Success Factor 1: Acknowledge change management as the real data management challenge

All self respecting organisations embark on the digital journey. However, oftentimes, this leads to the realisation that a huge legacy exists in the current IT landscape, and that your data ─ the content in your systems and applications ─ is not the gold you were hoping to find.

In other words, to understand the need of being data-driven is one thing. It is a different matter to help the organisation understand the amount of labour and the size of the change to uncover the gold in your data.

At MMG, we also journeyed towards a change to be data-driven and to implement a data management organisation ─ data governance ─ ultimately improving compliance and customer service.

Lean and easy

The most serious competitors are for us not the traditional industry competitors. Instead, they are the tech giants that may move into our industries, like Amazon, Facebook, and Google: all innovative data-driven champions of self service. As a consumer, we expect self service to be lean and easy, otherwise we choose the competitor.

These “new corporates” have built their businesses around data with huge success. They synchronously built their business along with much of the basis for their self service focused data management. Leadership was fully involved in the digital journey from day one. Change and data went hand in hand for these companies.


However, change and data have been asynchronous for most traditional companies. New technologies were adopted, while no centralised business-led data management vision evolved along with these new technologies. Data continued to be handled and supervised in silos, mainly by IT.

Traditional companies therefore have to put more effort in playing catch up to become data-driven.

The first step and relatively easy part is to put “becoming a data-driven organisation” on the agenda, embedding data as an asset in the strategy.

And then the hard work starts.

«Traditional companies therefore have to put more effort in playing catch up to become data-driven.»

The hard work

One example of that hard work is the technical implementation of the supporting tools. But perhaps the biggest challenge is the implementation of a change in culture and processes, with new roles and responsibilities that are necessary, but not directly straightforward for everyone.

We argue that data management is, first and foremost, not a tech challenge, but a change management challenge. Our first key success factor in MMG was treating data management as a change management challenge.

Key Success Factor 2: Making good use of data management frameworks helps structure and smoothen the change

Typically, when starting a data governance initiative, elements of data management already exist in the organisation, although not yet holistically coordinated. MMG was not much different. A framework can act as a catalyst to structure and smoothen this change towards a more mature data management, as it provides a frame of reference people can use to speak the same business language.

Miscommunication happens when people speak different languages. Here “language” is the framework we use: the terminologies and principles that we see as “the truth”, which we developed throughout our studies and careers.

Different truths

Professionals develop different “truths”. A major risk factor is when people are not aware that they have these different frameworks, terminologies or “truths” in the back of their minds. People then get frustrated when they think they understand each other at first, but actually ended up misunderstanding each other due to similar but different usage of terminologies. Misunderstandings lead to resistance, which complicates the data management change.

No one framework is a perfect fit at face value for any organisation. It requires adaptations to fit to the organisation’s specific needs. Gartner’s 7 building blocks for Master Data Management was used in MMG to provide a structure to the implementation of our data management initiative.

Potential confusion

By referencing Gartner’s framework, we directly witness a potential cause for confusion. Other frameworks see “Master Data Management” as an inherent part of a larger Data Management scope, similar to that which falls within Gartner’s MDM Building Blocks.

At MMG we started with a data strategy, which resulted in an MDM project, with which we set up a data governance to have good control over our data management.

Data strategy, master data management, data governance and data management: these terms are used interchangeably by some, and seen as related, but different topics, by others. And each of these terms can mean something different when talking with different professionals.

Frameworks are guidelines

It is evident that language, knowledge of frameworks, and agreeing about which frameworks to use in which context, is crucial.

One important learning from this: Always see frameworks as nothing more than guidelines, and never rigidly follow them to the letter.

Instead, adapt them to the environments and flavours of the organisation at hand. Make it your own. However, not using frameworks at all, with a total lack of awareness of different word usage for different data management frameworks, or even blatantly scoffing and dismissing the entire concept of frameworks, is a major pitfall and a recipe for full on confusion.

It especially causes confusion, as the dedicated profession of data management is still relatively new and could use a clear frame of reference.

Gartner’s 7 building blocks

Gartner’s 7 MDM Building Blocks helped structure how to implement a data management implementation in MMG, whereas the Data Management Body Of Knowledge (DMBOK) was the framework used to define and differentiate between the topics of interest within the data management profession, to define data management and its sub components.

DMBOK defines the needed capabilities end helps categorise deliverables in MMG.

We elaborate further on Gartner’s 7 MDM Building Blocks:

Gartner's 7 MDM building blocks
Gartner’s 7 MDM building blocks

Vision and strategy

Gartner’s MDM building blocks start with Vision and Strategy, as is the case with most frameworks. A stronger customer focus and simplification of the digital environment are examples of a good starting point.

The data strategy focuses on how you want to use data management to achieve the goals articulated within the vision.

Metrics decided by business provide a way to measure the success. They can be data quality KPIs, if available. Otherwise, they may be business KPIs, such as the number of meetings or project time it takes to build a data model.

As the framework shows, the building blocks follow an iterative flow, meaning that more iterations lead to improvements on the KPIs, but also a higher maturity on how to measure success.

The next four

The next four building blocks then form the foundation for how you want to operationalise. Governance sets the work environment, and ensures the right policies and templates are in place.

The People building block determines which stakeholders are to be involved in which roles, potentially including a change of structure. Process tackles business requirements throughout the data flow through the systems.

Lastly, the Infrastructure dimension provides the technical requirements on the tooling that must support the new data management environment. Gartner’s 7 MDM Building Blocks thus provided MMG multi-dimensional guidelines for its data management initiative.

MMG’s conscious use of tailored data management frameworks was another key success factor. It reduced confusion, provided a benchmarked guidance, and as such increased the acceptance of a dedicated data management organisation.

Increasing team value

Key Success Factor 3: Data management professionals well versed in change management increase your team’s value

The framework is an enabler to smoothen the transition. At face value, however, it needs something more. It is easy to focus on the technical solution and on the details. It is also easy to forget about the business needs and the amount of energy that goes into finding the business needs from the variety of different stakeholders, each with their own sometimes conflicting interests.

It is one thing to acknowledge that a change is needed, more difficult to do the actual compromise.

Change must continuously be communicated, implemented, followed up, adjusted, communicated, implemented, again followed up, and so on. In other words, this requires data management professionals who are experienced in the use of change management.

Change managers and data management professionals well versed in change management increase the value of the Gartner framework. This was MMG’s third key success factor for data management.

«It is one thing to acknowledge that a change is needed, more difficult to do the actual compromise.»

Living the success factors

The key success factors come to full fruition through a diverse team “living it” on testing grounds.

Yes, to change an organisation into a data-driven organisation may be hard, and sometimes even tiresome . A resistance for change is normal. However, with good change management, we innovate through the people that take part in the project.

Firstly, the project itself can be a good testing ground for the data governance you wish to implement. In MMG, a sub optimal data model led to the evolution of a governance, for the effective development of a more sustainable data model.

As the project progresses, a certain style is found that best befits the organisation. People are made aware that this is a serious change, and that stakeholder and resistance management is necessary: all hands on deck for the change for data management.

The project resources also adapt the frameworks to their taste in the project context. The frameworks come to full use as they are tested in the project.

Improving change management skills

In addition, the team members get to learn and improve their change management skills. Decisions made in the project still have a compartmentalised impact.

People feel they are in a safe environment, where they can make mistakes and adapt their roles, if a core principle in the project is that the project’s governance is a try out for the final data governance.

In the Nordic cultural context, as is the case in MMG, it is perhaps even more important to ensure that all stakeholders first have the chance to adapt their new roles and environment to their liking, before it becomes formal.

In our experience, also beyond MMG, a data governance becomes real through the roles defined within the project by the ones “living it”. The project resources’ self defined governance is useful to effectuate the data management organisation as it is to be implemented.

Secondly, once the project is a representative and diverse team, in the truest meaning of the word “team”, the different types, styles and opinions result in a more robust approach, better tested frameworks and even more change savvy resources.

«… a data governance becomes real through the roles defined within the project by the ones “living it”»

High fives

While one person may lean towards getting things done, another needs more time to reflect. They must familiarise themselves with each other’s styles, and they need to realise they are not alone in their mission.

Take time to support and cheer on your team as you go along. Your team members are the missionaries, the teachers, the coaches, and the gatekeepers.

They provide the quality, the drive, and positive energy. This is not a given. The energy and the support need to come from somewhere. Forge an alliance to stand in the storm together.

Do some high fives!

«Your team members are the missionaries, the teachers, the coaches, and the gatekeepers.»

An oiled machine

In our case, we formed our data governance by starting with a small organisation. In a cross functional and cross border team of data stewards, business representatives, system representatives, enterprise architects, coaches and top management, we scheduled weekly meetings and daily workshops.

With five months of continued training and practicing, we became an international oiled machine with on point fast decision making and prioritisation.

One important business stakeholder became a full fledged data steward hero and champion to our project. We had both energy and execution power in the room.

High five!

What’s next?

We utilised our three key success factors for data management: change management awareness, smart framework usage, and change savvy data management professionals.

We then took our core pain – a sub optimal data model – and by redesigning this model, we had the ultimate exercise to become the data management dream team.

We are still in the process of implementing the data management organisation, but with the generated momentum, we have a team that is ready to take on future data challenges.

Sources and further reading:

Are you interested in new ideas for data governance and its evolution? Check out Ryan Gross’s thoughts on The Rise of DataOps.

Denne fagartikkelen er skrevet av en av Dataforeningens ildsjeler. Har du noe du ønsker å dele med andre medlemmer, ta kontakt for å diskutere en mulig publisering på våre nettsider.

Oppslagsfoto: SleepyCats fra Pixabay