Establishing a Data‑Led Foundation for ERP Transformation
How specialist data analyst capability helped a large professional services firm address the data complexity underlying a major ERP integration across Finance, Risk & Compliance, and HR.
(Finance, Risk & Compliance, HR)
documented decision rationale
treated as one thread
analytics, reporting & AI
The context
A large, established professional services organisation undertook a transformational programme to integrate Finance, Risk & Compliance, and HR into a modern, unified ERP platform. The ambition was not just a technology upgrade but a reset of how data flows, governs and supports the business.
Like many organisations of its scale, the client's data landscape had evolved organically across multiple systems over many years, each designed for different purposes. The result was data that was fragmented, inconsistently governed, and duplicated. It was a firm without a clear system of record.
The methodology
The approach treated data as a first-class asset throughout. Four interconnected workstreams formed the analytical backbone of the programme:
1 — Logical & conceptual data modelling
A system-agnostic logical data model provided a shared language between technical teams and business stakeholders. By defining core entities and relationships independently of how data is currently stored, decisions could be made consistently even where systems differed in structure and terminology.
2 — Embedding SME insight as data logic
Rather than treating SME sessions as separate discovery work, business insight was embedded directly into the data model. SMEs validated interpretations, surfaced implicit rules, and helped distinguish genuine data quality issues from legitimate operational variation, thus converting tacit knowledge into documented, testable logic.
3 — Retention as data logic, not policy text
Retention requirements were translated from policy language into data-level rules: which records are in scope, which attributes drive outcomes, and which systems are capable of enforcing action. This enabled consistent, testable application of obligations rather than manual interpretation case by case.
4 — Migration mapping built on understanding
Migration decisions were derived from understanding rather than assumption. Source data was mapped to ERP target concepts using the logical model, with gaps and mismatches surfaced explicitly. All decisions documented with rationale resulting in an audit trail that survives personnel changes and programme phases.
The Data Analyst's role — distinct from Business Analysis
The engagement was structured as a specialist data analyst role, separate from Business Analysis. This distinction matters and is frequently misunderstood on complex programmes:
| Dimension | Business Analyst | Data Analyst |
|---|---|---|
| Primary focus | Business change and requirements | Data structure, meaning and rules |
| Typical questions | What does the business need? | What does the data represent? |
| Core artefacts | User stories, process models | Logical data models, data mappings |
| Risk mitigated | Change adoption risk | Data integrity and migration risk |
| Continuity value | Requirements ownership | Prevents re-interpretation drift across phases |
The Data Analyst acted as the analytical spine of the data programme: maintaining continuity of understanding across workstreams, translating policy and business meaning into data logic, and ensuring all decisions were evidence-based and traceable.
Data analysis is cumulative. Each analytical decision builds on the last. When the thread of understanding is broken, be it by handover, phase change, or team rotation, programmes pay for it in late-stage rework, migration errors, and compliance gaps that should have been addressed at source.
Outcomes
Positioning for analytics, AI and the future
Well-structured, well-understood, well-governed data is the prerequisite for every data-driven initiative that follows whether it is reporting, automation or AI. Organisations that invest in this foundation now avoid re-doing it under pressure later when the AI use cases are already being demanded.
What the organisation gained
- Governed, system-agnostic data model as programme anchor
- Tacit SME knowledge converted to documented, testable logic
- Retention policy enforced at data level — not left to judgement
- Migration decisions with full audit trail — defensible to regulators
- Data literacy across teams, reducing future external dependency
Applicable to organisations facing
- ERP upgrade or migration in progress or planned
- Retention obligations not yet translated into data logic
- Data meaning held tacitly by people rather than systems
- AI or analytics ambitions stalled by data quality issues
- Previous migration projects that surface surprises too late
Facing a similar challenge?
Whether you're planning an ERP transformation, addressing data quality, or building the foundation for analytics and AI, we'd welcome a conversation.
Book a 20-min clarity call Email us directly