Understanding Data Democratisation

In 2006, an influential article titled "Competing on Analytics" by Thomas Davenport and Jeanne Harris was published in the Harvard Business Review. This work ignited conversations around leveraging analytics for competitive advantage, prompting companies to invest heavily in business intelligence software, big data infrastructures, and AI-driven tools. However, the outcomes have often fallen short of expectations. A Deloitte survey revealed that, fourteen years later, only 10% of U.S. firms effectively utilized analytical insights, with Microsoft Excel remaining the most common analytics tool.

Transforming into a data-driven organization is more complex than it appears. Assessing your company’s positioning on the data maturity scale requires analyzing four critical domains:

• Data: Quality data is essential for effective AI.
• Skills: Is your workforce equipped with data literacy?
• Tools: Does your infrastructure support analytics at scale?
• Culture: A legacy culture resistant to data insights can hinder progress.

At my organization, a major bank, we currently rate 2.5 out of 5 on the maturity scale, striving to reach level 4. Globally, the average sits around 2.2, indicating a vast majority of employees lack analytics proficiency. Data democratisation serves as a bridge to close this data literacy gap.

Data democratisation represents a collective effort to enhance data maturity across all departments. Rather than relying solely on specialized teams of data scientists, my bank is committed to equipping all 40,000 employees with the tools, skills, and culture necessary for self-service data access, automating routine tasks, and prioritizing data-driven decision-making over instinct.

Imagine if each employee could save just one hour a week through automation — that would amount to approximately 2 million hours saved annually, translating to around $150 million available for other investments.

Recognizing the importance of both routine tasks and high-level analytics projects, data democratisation fosters a culture where both everyday contributions and groundbreaking projects are valued equally.

What does data democratisation look like in practice? Below are five strategies, illustrating advancements in tools and culture.

1. Unified Analytics and Auto-ML Platforms

   The 2020s have seen a surge in low-code analytics and machine learning platforms that consolidate visualization, analytics, and automation into one interface. Notable examples include Alteryx, Dataiku, DataBricks, and Snowflake.

My bank utilizes Alteryx and Dataiku, which streamline processes for our finance team. With intuitive drag-and-drop interfaces, they can quickly set up audit pipelines without needing to code, significantly reducing the time spent on these tasks.

How to make data more accessible to business users (in the right way) - YouTube

2. Integrated Productivity and Analytics Ecosystems

   This approach focuses on creating a tightly integrated suite of applications. Microsoft exemplifies this with its ecosystem, which includes Office 365 and Power Platform — a suite of low-code tools for data modeling, app creation, automation, and chatbot development.

The ease of use and integration within Microsoft 365 enhances adoption rates, allowing employees to streamline processes like data requests through automation, significantly saving time.

3. Advanced Big Data Tools

   Many companies have struggled to realize the potential of big data due to poor tools and quality issues. At my bank, we faced challenges with initial big data tools, but transitioning to a cloud-based solution on Microsoft Azure has improved our data integration and governance capabilities.

With Azure, we’ve created a more cohesive big data experience that integrates analytics tools and data governance seamlessly.

4. Fostering a Supportive Culture

   The most significant hurdle in achieving data-driven maturity is not technology but rather the organizational culture. To overcome resistance to change, my bank initiated a Data & Digital Enablement program, offering employees opportunities to learn about data fluency and leadership through various engaging formats.

We also host an annual TechX conference and Data Week, reinforcing the theme of "Data for All," showcasing the importance of data accessibility across the organization.

5. Establishing Data Marketplaces

   A critical issue in many organizations is a lack of visibility, trust, and timeliness surrounding data. Companies are shifting toward decentralised data mesh architectures, allowing business units to create and share trustworthy data products across the enterprise.

My bank is developing a user-friendly data marketplace, reminiscent of platforms like Netflix, where employees can easily access curated data and publish their own assets, thus treating data as a valuable product.

What's Next?

The most impactful analytics often stem from domain experts, and as we move forward, the importance of democratizing data access will only grow. By empowering citizen data analysts, organizations can enhance data quality, drive insights, and create a culture of continuous improvement.

As generative AI continues to evolve, it promises to simplify analytics further, allowing anyone to leverage data insights with ease. The landscape of analytics is rapidly changing, and it's an exhilarating time to be involved in this field.