Skip to content

Methodology

This section outlines the methodology used to evaluate the organization’s data landscape, define strategic priorities, and design a governance framework fit for a modern data-driven enterprise.

Note

  • The assessment involved data discovery, inventory, quality audits, and stakeholder interviews.
  • Strategy development was informed by industry best practices and aligned with organizational priorities.
  • A governance framework was designed to ensure compliance, data integrity, security, and reusability.

Environment Discovery

A dual-track approach was adopted to ensure a comprehensive and balanced evaluation:

  • Technical Analysis: Assessment of existing data stores, pipelines, infrastructure, and practices from both the Data Office and Nissan Digital.
  • Business Discovery: In-depth interviews and workshops with key stakeholders across multiple departments and levels of the organization.

This combined approach enabled both validation and exploration, cross-checking known practices while uncovering overlooked data needs and isolated initiatives.

Crucially, business priorities were intentionally placed above technical constraints. Resource availability was treated as a variable, not a blocker, so the resulting strategy could be designed from a value-first perspective.

Interviews with executives and functional leads were conducted to confirm alignment with enterprise goals, validate assumptions, and bridge communication gaps between technical and business teams. This ensured a shared understanding of both needs and constraints across the board.

Business Needs Assessment

Understanding the business context is foundational to defining any sustainable data strategy. Rather than starting with tools or architecture, we anchored the methodology in actual business challenges and value drivers.

The primary business case that catalyzed the creation of the Advanced Data Office (ADO) was marketing-centric: consolidating and enriching customer data to enable consistent, reliable, and actionable insights. This objective underpins the Customer 360 (C360) initiative.

Additional business priorities identified include:

  • Purchasing Efficiency
  • Marketing Cost Control (e.g., FMI / VME optimization)
  • Supply Chain Resilience
  • Sales Efficiency
  • Financial Forecasting & Reporting
  • Vehicle Design Optimization
  • Customer Experience Enhancement
  • Warranty Claims Management

These needs formed the foundation for designing the data architecture and use cases. In short, business use cases dictate technical design, not the reverse.

Data Discovery

The purpose of the discovery phase was to map out the data ecosystem, what data exists, who produces it, who uses it, and how it flows.

Key findings include:

  • A significant share of data is managed manually via Microsoft Excel files (.xls, .xlsx, .csv).
  • Many datasets are siloed within departments, rarely shared, and lack visibility outside the producing unit. (cf. QX80 remote issues discovered 3 months after the facts which should have been picked on simply by looking and monitoring the Telematics vs waiting for TCS to compile complaints)
  • The absence of a centralized data catalog makes discovery labor-intensive and reliant on repeated stakeholder interviews.

Discovery was supported through extensive outreach across teams, including senior managers and non-managers, to piece together a more complete picture of the data ecosystem.

[Read More]

Data Quality Evaluation

To evaluate data integrity, a practical and highly relatable approach was taken: using the author’s own customer record as a known-good baseline.

This method revealed discrepancies and fragmentation across systems and highlighted areas requiring remediation.

Additional techniques included:

  • Entity-Relationship Diagram (ERD) analysis: verifying the presence of primary/foreign keys, examining record cardinality, and checking for null or low-volume fields.
  • Depth Analysis: Assessing modality richness and attribute consistency.
  • A real-world project provided by the CFO was used as a benchmark to test feasibility, agility, and quality of basic analytics on existing datasets.

[Read More]

Resource Mapping

Beyond tools, the human network proved to be the most critical resource in uncovering hidden processes and data dependencies.

Additional resources used during this assessment include:

  • ADO internal team knowledge
  • Confluence documentation
  • Historical vendors and projects (e.g., Maritz, CGI)
  • JIRA for tracking existing and past initiatives
  • Bitbucket repositories for accessing legacy code and integration logic

Collaborations and informal interviews were also conducted with staff from Tech Mahindra, Maritz, and CGI to gain perspective on both historical context and ongoing operational nuances.

This methodology establishes the foundation for actionable recommendations in the next sections, ensuring that proposed solutions are practical, business-aligned, and technically achievable.