Pyramid Analytics

Overcoming Data Analytics Challenges - Pyramid Analytics

The article discusses how organizations, exemplified by St. Joseph’s Health, can overcome cultural and technological barriers to achieve analytic maturity and data excellence by following five key steps—leadership-driven change, organizational commitment, building trust through gradual transition, proper onboarding for adoption, and maintaining momentum through technology enthusiasm—while emphasizing the need for analytics to be unified, flexible, accessible, and collaborative.

The journey to analytic maturity is long and complex. CIOs worldwide prioritize investment in data analytics, and the ROI of robust data strategies is clear: organizations that make significant use of data to differentiate themselves in the market enjoy a 30% year-on-year growth. However, organizations often struggle with cultural and technological barriers when modernizing data strategies and unlocking real-time analysis capability.

IT, data, and business intelligence professionals gathered to discuss the steps to achieving analytic maturity and data excellence, and how to bring teams and organizations on the journey.

Embarking on the Analytics Journey

Bill Balnave, VP of Global Solutions Engineering at Pyramid Analytics, shared the story of St. Joseph’s Health. The organization struggled with cumbersome daily flash reports—60 pages of charts and numbers with little real value for decision-making. To achieve maturity, Bill outlined five key steps:

  1. 1.Leadership urged change. An internal disruptor, in this case the CEO, recognized the inefficiency of the existing process and catalyzed change.
  2. 2.Organizational commitment. Ownership within the organization ensured the initiative was taken seriously.
  3. 3.Trust through steady evolution. Gradually transitioning to visual analysis while maintaining the old method helped build trust in the data’s source.
  4. 4.Proper onboarding for widespread adoption. Empowering different groups encouraged adoption and fostered an environment for process improvement, such as implementing automation.
  5. 5.Harnessing excitement around technology. Once trust and adoption were established, it was important to maintain momentum.

Ian Macdonald, Principal Technologist at Pyramid Analytics, illustrated the concepts behind the St. Joseph’s Health case study. He discussed four characteristics needed to achieve analytics maturity: unified, flexible, accessible, and collaborative. Moving along the maturity spectrum requires these approaches to be foundational to your data strategy. The typical journey begins with descriptive analytics, moves through diagnostic, predictive, and prescriptive analytics, and ends with cognitive analytics.

Maturity is achieved when data-driven insights are automated using machine learning techniques. This includes:

  • Providing a custom environment for each user based on their experience and business needs
  • Enabling annotation of data and real-time conversations around analytics output
  • Ensuring data accessibility across the organization
  • Minimizing data movement

A foundation of trust in analytics is essential for data-driven success and decision-making.

Building a Collaborative and Mature Data Analytics Culture

Discussion groups explored key barriers to driving maturity in data strategy, building trust in data, defining data excellence, relieving bottlenecks around data science teams, and empowering business users to adopt analytics tools.

Barriers to effective analytics strategies were split into cultural and technological issues:

Cultural Barriers

  • Teams blocking data sharing outside their own groups
  • Lack of desire to build data collaboration
  • Insufficient training
  • Need for leadership from the top
  • A “knowledge is power” attitude, leading to reluctance to share data or techniques

A major issue is the lack of training. Often, organizations invest in analytic technologies but are reluctant to spend on training business users. This inhibits user adoption and ROI. Addressing this requires investment in training and fostering a culture where ongoing data education is part of the job.

Technological Barriers

  • Difficulty predicting costs when moving to the cloud
  • Fragmentation of technology platforms
  • Challenges with data published on third-party platforms
  • Centralized data repository platforms can take over 12 months to deliver

An alternative approach is to use small agile teams to continually improve the data process, focusing on continual improvement rather than perfection.

A widespread issue is the tension between centralized data warehouses (owned by data/BI teams) and decentralized user access. Centralized teams can become bottlenecks, but decentralization can compromise data integrity and create data silos. The challenge is to empower business users with self-service capabilities while maintaining a single source of truth and data integrity.

Although there are many areas to focus on for data and IT professionals, beginning with establishing trust and empowering users is key to achieving analytic maturity.

Originally published at Nimbus Ninety.