June 17, 2022
min read

How to engage your organization for data transformation?

Just 2 months after the first discussion with the LightsOnData show, YOOI rejoined the webinar with new data management insights. This time, Nicolas discussed with George and Diana Firican the importance of culture when it comes to digital transformation, the best practices to engage people in the journey and the keys to driving a successful transformation.

The LightsOnData show is one of the top self-media that share the opinions and best practices of data management with global leaders.

This is about how organizational culture can be a catalyst, instead of a culprit, to the data transformation journey. Following are the key takeaways of the interview and some additional information related to this topic. Enjoy!

First of all, how to define data transformation?

It is widely known that relying on data is really at the heart of the "digital" revolution. Because of this, it is also a major requirement for organizations:

  • to leverage data to boost productivity or improve efficiency,
  • to develop new business models
  • to address new markets
  • and more…

Most organizations that are considered "digital-native" are also "data-native". All these characteristics give them their level of responsiveness and agility to take accurate and timely decisions.

But for the organizations who need to perform their digital transformation, data transformation is a new practice and a big shift that is required to enable a successful digital transformation.

And data transformation implies the following challenges:

  • Make data accessible (with the right quality, security, etc.)
  • Make sure that individuals have the tools and the skills to use data to derive insights and support decisions
  • Transform business processes to effectively incorporate data-driven insights
  • And it’s about shifting from a gut-feeling-insights mindset to a based-on-data-insights mindset

So data transformation is about the ability of the organization to manage data more effectively and to extract value from those data.

This is frequently called "being data-driven", even if we should really talk about being value-driven instead.

Data transformation vs. Data strategy?

6 dimensions to succeed with Data Transformation
Illustration by the author

A data strategy is the plan to deliver on the data transformation, this is why it is deeply linked to the business strategy. A strong data strategy is indispensable to delivering value to the organization in the long term.

To succeed with the transformation at the scale of the organization in the long run, there are some dimensions that are really important:

  • Alignment: Due to the complexity of organizations, existing processes and systems, efforts need to be coordinated for reaching well-aligned goals. Alignment is to make sure things are moving together, are well articulated and are pursuing the same value
  • Collaboration: Again, because of the complexity of the business, a collaborative approach is important and ultimately helps uncover insights more quickly and deliver more accurate answers.
  • Innovation: Deploying the transformation also relies on innovation capabilities as well as experimentations, iterations, test and learn approaches
  • Governance: There are more and more regulations, compliance requirements and growing risks (cybersecurity, privacy, ethics, image & social media, etc.). Being able to move safely while moving fast is mandatory - and ultimately this is what D&A governance is about.
  • Literacy: Being data-literate leads to developing critical thinking and helps employees see the bigger picture. Accurate decision-making improves performance and provides a better competitive advantage in the business environment.
  • And the technology of course remains an important enabler, even if there is usually too much focus on that aspect.

Why is it so difficult to engage people in this journey?

NewVantage Partners reports that only 37.8% of companies have achieved to establish a data-driven organization, highlighting a real challenge with evolving the internal culture, skills, organization and processes in order to achieve this shift.

Delivering Data & Analytics is not only a technological issue but also very much an organizational one. You need to build a valid data product, making sure it gets integrated into operational processes and gets adopted by users. In reality, data products are often ignored or walked around.

In terms of obstacles, the challenge in engagement can be divided into two perspectives.

The first one is social perspective with:

  • A skills gap, usually called "Data Literacy": having the ability to read data and understand it is a game-changer.
  • A trust gap: data becomes useful when people use it. The more comfortable and confident people are with data the more they will be able to use it in specific contexts and build stories on their own.

And second is the change management aspect. There is no data transformation without evolving operational processes to incorporate data-supported insights. Change management also means evolving the culture.

Why culture is key in this process?

Culture is what unites individuals within the organization, and it can be defined by shared behaviors and ways of interaction of those individuals.

When developing a data- or insights-driven culture, it is hard to make people understand why they need to work in a new way as they resist change automatically (remember, the Skills and Trust gap).

A few years back, the data team at a major airline built a sophisticated machine learning model to perform predictive maintenance on airplane parts. They gathered historical data, and worked hard in the data lab to train and fine tune the model and achieve a high level of accuracy.

Once finalized, they tried to get it adopted on the field. But the workers did not trust the predictions of the model, and the data team was unable to “explain” the logic and the behavior : “but why should I change that piece that look in good conditions?”. Because of this lack of trust, the model, while very effective, was in the end strongly rejected and as such never used.

To build back trust, the data team scraped the ML model, and went back to work to produce a much simpler and less accurate model, based on statistical calculations and completely explainable. That version ultimately got accepted and used by workers, and the machine learning version never got re-introduced before many years.

As the saying goes: "culture eats strategy for breakfast". The most difficult part is probably to avoid individuals sticking to old habits and not embracing new tools, processes, etc.

Everyone want changes but nobody wants to change

Generally, there are 3 obstacles to cultural change:

  • Adoption resistance
  • Organizational structure challenge
  • Leadership commitment

Shaping culture takes a lot of time and effort but it undoubtedly matters in this data transformation journey.

Are training programs the solution to teach people how to manage, use and read data?

Organizations need to show individuals how they can evolve and support them in this change. In that context, training programs are required as part of the equation, as a great way to provide support and as a way to communicate widely on the change initiative.

But too often, training is seen as a magic bullet, used as the only way to educate people and to try to address the data literacy issue. "They have been trained so they are data literate". And it really does not work.

The problem with large training programs is that they are mostly useless when they are not connected to concrete and short-term applications. You send individuals to training classes, and 10 months later when they face a data-related question, they have already forgotten almost everything.

Good training happens when:

  • It directly connects to concrete challenges individuals are facing - meaning they are already facing them and realize that they miss something to address the challenge
  • It can be put in practice instantly in their context - so that they can almost directly apply the learnings to their challenge and so that it gets assimilated in the long term
  • Individuals are motivated to learn, which is tightly connected to the first one

Otherwise, training programs are just "awareness programs", and generate frustrations:

  • The frustration of losing time for some, because they don't see how it relates to their context & their daily work
  • The frustration of being too theoretical for some, because they don't see how it can solve their problems
  • The frustration of having to train individuals who changed jobs just after

We can firmly say that training is needed, but only without missing the expectations:

  • Initial awareness sessions can help to prepare minds, and put things in motion. But they won't benefit beyond that.
  • Avoid upfront training in large batches and don't expect they are really the ONE solution to drive the culture change.

What is the solution to efficiently uplevel data skills across an organization?

The key approach here is to embark individuals on the transformation journey with concrete projects!

The ultimate goal is to build Data & Analytics capabilities within the organization. This can rely on hiring new people with the required skills and/or upskilling current employees. In both cases, to attract new talents (and retain them) or to support existing ones, providing the right context is key.

So, the approach here is to start from the end :

  • Find ways data can serve the business strategy, and that can serve as an example
  • Having a clear value-driven data strategy, that defines the goals, and prioritizes initiative with a clear expected value that supports the transformation
  • Have a bias for action and focus on delivering concrete initiatives
  • Onboard users from the very start, so that they fully share the current status as well as the future & expected one
  • Build the foundations & culture on the go
  • Identify the existing gaps (technology, processes, skills) in that context and integrate those in the roadmap
  • Don't forget about governance: risks, ethics, etc.

It is very much like applying product management principles to Data & Analytics.

We see this as the only way to align the efforts and really engage individuals in the change. All members across the organization can understand the reason for the change and be aware of the challenges. Eventually, they will be able to benefit from targeted training that will enable them to succeed in the current initiative.

Users will also be in a better position to accept new tools and new processes (and will be able to contribute to their evolution to ensure they are fit and efficient).

In other words, training programs must be fully embedded into the initiative's value chain, the same way the tech and process roadmap should be. The real challenge is to help people obtain necessary skills when they REALLY need them.

In some cases, we see organizations who provide big upfront training programs or other enablement initiatives. Then, they rely on “vanity metric” to measure successes: How many people went through? How many people passed the test? Etc.

These metrics are pure vanity because they fulfill the ego of the organization and give a false sense of progress toward the target. Yet, in reality, they don’t reflect any actual progress. Valid metrics that matter are always related to the goals: What part of decisions are based on data-driven insights? Are individuals able to actually derive new insights from data in a self-service way? What part of datasets are searched for and reused? Etc.

By engaging users in initiatives early on and along the full lifecycle, provide adequate training in that context to help them along the way, we maximize the chances that initiatives will be not only Useful, but also Useable and Used!

What should an organization do to sustain the change/perpetuate the initiative?

Once you have achieved some successes and managed to transform some specific areas, the challenge becomes 1. to scale it and 2. to make it stick.  This means we do not want to see people reverting to old practices after some time - a.k.a. the elastic principle.

To recap :

  • You need to have a strong data strategy, that gives the direction and supports continuous value delivery for the enterprise.
  • In that context, “continuous” is really important as you need frequent "wins" to create the Wow effect and to serve as an example or as a trigger for the rest of the organization.
  • The data strategy has to be aligned with the enterprise strategy while also aligning with individual objectives. Everyone needs to see that there is something in there for them.

Then, in terms of culture, you need to make sure to onboard individuals in order to scale the different expertise and ensure the collaboration across those experts that are all required to deliver D&A initiatives.

And finally, you need to build communities to ensure the first ones on board (the “early adopters”) will help with scale in the organization, provide help with peers, etc. Communities gather and animate people with similar interests, and help spread messages. They are a strong tool to engage individuals and give them the opportunity to “grow and shine” by animating and infusing expertise in different parts of the organization.

Data transformation is more about people than technology, but how technology can help people in data transformation?

While we have restated several times that the focus should almost never be on technology, we can't ignore that some technology shifts have been essential enablers for some approaches:

  • The new "workplace" tools like Slack, Teams, Zoom, Meet, etc. made remote work a real thing
  • "Big data" platforms like Hadoop initially opened new data processing capabilities and whole new use cases became possible
  • Machine Learning opened new techniques to derive insights from big data and opened the way to many new use cases,
  • Self-service data discovery and BI platforms widened the depth & scale of analysis that can be provided to end-users,
  • etc.

In the same way, this is why we have built YOOI: to really help with animating the strategy, ensuring and tracking the alignment with value, enabling the end-to-end governance of D&A projects… and help engage communities with a single cockpit that gives visibility, supports communication, and enables collaboration, etc.

Having strong digital platforms to support data transformation is really a game-changer. Relying on "homemade" tools, usually, Excel- and Powerpoint-based, makes it really hard to scale, to engage everyone, to stop spending excessive time chasing the different participants, stakeholders, etc. !! So while those might be a good way to start, be aware of the limits and shortcomings over time.

So, yes, tools and technology are not an end but are important enablers, especially to scale and to be efficient when it comes to empowering users, keeping alignment, facilitating collaboration and ensuring governance.

How to make sure the culture is strong and “working”?

Regularly assess D&A maturity

Looking at the transformation as an iterative process is key. It means having the typical Observe-Orient-Decide-Act (OODA) loop to adjust the course of action: continuously iterate processes, improve workflows and collaboration, and explore new opportunities in terms of capabilities.

Regularly check how many people are on board and how many are not and/or are resisting, and identify ways to bring them onboard.

Measurable goals to get tangible impacts

According to Gartner, the CDOs who successfully demonstrate ROI from their D&A investments are nearly twice as likely to be effective at consistently producing clear business value for the organization.

Measurable and achievable goals give clarity across the organization and initiate discussions around D&A transformation. This helps animate the transformation and extract more value from D&A. This is what I call achieving a value-driven mindset.

What would be the recommendations or best practices for a great start in this journey?

The first thing is to accept and understand that this is a journey, that it will take time, and that it needs to be progressive. Most attempts to go faster with too large upfront programs fail to deliver because they face that transformation need.

  • A clear and value-driven data strategy built to support the enterprise strategy and that enables the organization to focus on progressive value delivery, quick wins, and measurable outcomes - to engage users and align objectives
  • A good visibility on data maturity to identify gaps and what is required to enable that strategy (tools, processes, people)
  • Have a CDO to lead and conduct the transformation, make the link with the executive committee and business leaders, ensure leadership engagement, and have the leverage to get engagement and budget
  • Identify and prioritize concrete initiatives that are achievable quickly in the current context
  • Deploy the right organization, tools and governance to enable to move fast (iterations, experimentations, collaboration, reuse), while moving safe (privacy, cybersecurity, ethics, environmental and social responsibilities) - supporting the identified initiatives
  • Rely on communities and targeted training programs to support individuals engaged in initiatives

A focus on delivering progressive and tangible results, with concrete projects, is really important in that context.

These efforts will gradually give light to new innovative ideas and help evolve the culture as time passes.

Is having a CDO optional or crucial to engage people in data transformation?

Leading these changes with a dedicated person, and being able to adopt a holistic approach as a Chief Data Officer does, is the right move if you want to initiate a deep data transformation. In French, we would say that the CDO is the "chef d'orchestre" (the conductor)!

It’s a position extremely demanding in terms of expectations since CDOs are managing the whole roadmap to make data transformation happen and work.

CDOs are considered business leaders. They need to map their data strategy on the business objectives and become a change agent using data as an asset to achieve business needs and outcomes. They coordinate with other executive members and business leaders to get their buy-in and engage them in the transformation. They animate the overall program and report on its progress to the executive committee.

The CDO has a unique leadership to establish this transformation, and support the business ambition.

In contrast to the changes a CDO can bring, here are two archetype situations that organizations might be facing before such a function gets created:

  • Data is embedded into each business line, each struggle with skills, rely on heterogeneous technologies, and operate with inconsistent governance, leading to poor outcomes and creating growing risks for the organization
  • Data is a pure IT function, that focus on delivering technology platforms and foundations, but fails to embark on the business and deliver tangible value

How YOOI is helping in this data transformation journey?

CDOs need to have a holistic view to drive change within an organization, animate and engage all the different parties, and report on progress.

Supporting CDOs and the data transformation are our main goals at YOOI, providing a platform that enables data leaders to pilot all their Data & Analytics initiatives: from defining and deploying their data strategies to encouraging collaboration within their organization, and tracking expected vs realized value creation through those investments.

By connecting with underlying technical tools, managing the alignment and building the bridge between business and IT, YOOI is the cockpit of the Data transformation.

It is the right platform to:

  • Drive a data-infused and value-first culture across the organization by empowering all levels of workers to communicate with data, building knowledge and making decisions
  • Streamline business processes and workflows, which helps everyone thrive with data and helps scale the governance efficiently
  • Scale literacy efforts across the organization by enabling the development and animation of data communities thanks to great collaboration features
  • Demonstrate tangible return on investment by consolidating information, to ensure resources are prioritized and used, and to monitor the value created at every level of each initiative

Have more questions? Reach out to YOOI through request a demo or get in touch with Nicolas on LinkedIn.

Feel free to replay the webinar with George Firican on the LightsOnData show where we answered some other spontaneous questions from our audiences.

If you missed our first LightsOnData show, "How to drive value from your data strategy?", click here for the replay.

Don’t hesitate to contact our team to learn more and schedule a demo. We’re always here to help!

People also view...