4 steps to a successful data-driven cultural shift

Lavish amounts of data, competitors’ pressure and immense opportunities of data-driven business applications made companies heavily invest in data science initiatives. With many reputable software vendors and data science experts on the market, the transformation shouldn’t be that difficult, right? 

Unfortunately for many ambitious business leaders, the main challenge of the data-driven change is related to culture, not technology. According to the 2020 Big Data and AI Executive Survey by NewVantage Partners, only 26.8% of the surveyed organizations report having success with building a data culture. 

In the majority of cases, employees are reluctant to make a shift in their mindset and adapt to a different workflow. Establishing a strong digital culture and revamping organizational processes is what companies should focus on when going for data-based transformation. Let’s look at the four steps that organizations can take to create a culture nurtured with data. 

Start from the top

First things first, the CEO’s role in any transformation can’t be overestimated. In case the CEO isn’t completely immersed in the idea of putting data at the center, every other action taken to empower the cultural shift will most likely fail. The CEO is every employee’s indicator of how important, real and valuable ‘all this nonsense’ is.

If the CEO lacks an understanding of the data science potential, a mixture of coaching and educational sessions is here to help. Headed by Chief Data Officer or even a specialist called from without, top management training is a required step toward a smooth cultural change.

Next, the C-suite should be in charge of internal awareness campaigns. It’s crucial for business leaders to adjust their approach and attitude to communicating ideas to their employees, instilling value in data processes, and placing data in the spotlight during meetings and presentations. However, it’s important to avoid ambiguity when appraising data. Ideas regarding data that are conveyed by leaders should be easily recallable and understandable even by the least tech-savvy staff members.

Run educational programs smartly

As you might expect, educational programs should be at the core of a data-centric transformation. Often overlooked, the lack of a relevant training program can be the sole reason for employees to lose touch. Without relevant knowledge and skills, they can’t understand why exactly the transformation is happening and, most importantly, what opportunities it presents. When done right, training programs are anxiety diffusers and trust enablers.

What’s important not to miss here is that these programs should be as engaging as possible. ‘Learning by doing’ has proven to be the shortest way of becoming proficient in often complex data science concepts. Consequently, by applying this method, each department will be able to craft training programs defined by their employees’ existing skillset and the exact way these people can use data to their advantage.

For example, project managers can be tasked with formulating the problem, which then can be solved by utilizing data analytics. In turn, the sales team can come up with strategies based on the data analysis findings in their specific area. By conducting such simulation exercises, every department can get hands-on experience and have a better understanding of how data can foster better decision-making in their domain.

Another critical success factor of educational programs is their timing. For example, fundamental skills required to grasp data-driven decision making such as self-service BI essentials can be taught before the major change is introduced. In-depth analytics and decision-making training should take all the employees’ headspace right before the time when those tasks will become essential to the company’s operation.

Democratize data

One fundamental requirement for data to bring value is to be accessible. In McKinsey’s report ‘Why data culture matters’, many top C-level executives of established enterprises share very similar experiences with data democratization — the more accessible the data, the more drive, trust and excitement there are in the employees. Without universal access to data within an organization, data loses a great chunk of its potential. 

However, in the context of transformation, granting data access to everyone in the company can be a complex task, as it requires data reorganization, which is an often tedious and slow process. To overcome this challenge, companies can provide access to specific datasets with only the most relevant metrics that align with the current goals. 

Another important aspect of data democratization is to focus on erasing communication boundaries between data scientists and business leaders. One common pitfall to be noticed across many companies’ structures is that their data science departments are separated from the decision-making center. While I don’t want to raise the never-ending ‘centralization vs decentralization’ debate, hybrid models work particularly well in this context. One often overlooked nuance here, however, is that it’s equally important for both sides to move closer to each other. Again, business leaders need to be proficient in data-related concepts and constantly expand their technical vocabulary. 

Make data a habit

Essentially, any cultural shift is a matter of changing habits. In many cases, employees’ habits are defined by their environment and the tools they use. That’s why it’s crucial for an organization to remove cues that trigger old decision-making patterns. As we’ve already established, information sharing and collaboration are pivotal characteristics of successful data-driven transformation.

That’s why it makes sense to reorganize existing office spaces in a way that encourages networking. Ultimately, any tool or template that fosters old decision making and contradicts a data-based approach should be reviewed and be removed or changed. New KPIs also need to be set wisely. At this stage, it’s crucial to not just evaluate how positive the results of data-driven projects are but to reward any ongoing data-based actions and initiatives.

Conclusion

Companies of the new generation have an unfair advantage of being built with data at their core. Older companies, on the other hand, need to completely rethink their strategies and restructure their governance to put data first. While this sounds like a tremendously heavy task, it all starts with small steps in the right direction in the company of the right people. Figure out who of the employees can be your main influencers, the ones who are truly captivated by the idea of building a data-driven culture. While the CEO’s role is vital, spreading ideas personally by word-of-mouth is a far more impactful way of communication compared to blanket announcements.

Building a data-driven culture is all about changing the mindsets and habits, which never happens overnight. It takes time for ideas to take root, and only a continuous day-to-day instilling of these ideas can prove successful.