Phase 03 / 08 TRANSFORM

Analyse

Understand legacy, master processes, sharpen requirements

The Analyse phase produces the complete functional and technical understanding of the starting position: for legacy systems, agentic reverse engineering reconstructs architecture, business logic and data models; in parallel, business analysis examines the as-is processes and requirements engineering specifies the requirements for the agentic target solution.

Position of the phase in the TRANSFORM delivery model

Process model

The phase delivers a consolidated, traceable knowledge base: reconstructed system documentation, analysed as-is processes with weak points, and a quality-assured, prioritized requirements model following the IREB standard.

System inventory & scoping
Reverse engineering (for legacy)
Business analysis of as-is processes
Requirements engineering
Consolidation & gap analysis
1

System inventory & scoping

Capture the affected systems, interfaces, data assets and process domains; define analysis depth and reconstruction goals per system based on the roadmap prioritization.

2

Reverse engineering (for legacy)

Agentic reconstruction of architecture, module structure, business rules and data models from source code, databases and operational artifacts — methodically aligned with the Chikofsky & Cross taxonomy and the horseshoe model.

3

Business analysis of as-is processes

Process capture and evaluation using process mining and BABOK techniques; identification of media discontinuities, automation potential and human-in-the-loop points.

4

Requirements engineering

Elicitation, documentation, validation and negotiation of functional and non-functional requirements per IREB/ISO 29148; modelling with use cases, user stories and to-be BPMN processes.

5

Consolidation & gap analysis

Merge system knowledge, process analysis and requirements into the knowledge base; gap analysis between as-is and the agentic to-be as direct input for the Design phase.

Methodology mix

Reverse engineering, business analysis and requirements engineering

MethodPurpose in the TRANSFORM contextReference
Reverse engineering taxonomyConceptual and procedural frame for design recovery and re-documentation of legacy systems; scientific basis of whitebox reverse engineering with SLIP.Chikofsky & Cross (1990)
Horseshoe model (architecture reconstruction)Three-level reconstruction from code via function to architecture view; structures the transformation from legacy to target architecture.Kazman, Woods & Carrière (1998)
Process miningData-driven reconstruction and conformance checking of real process flows from event logs; objectifies the as-is process analysis.van der Aalst (2016)
BABOK v3 — business analysis techniquesCurated technique set (incl. stakeholder analysis, value stream analysis, root cause analysis) for the functional mastery of processes.IIBA (2015)
IREB CPRE / requirements engineeringSystematics for eliciting, documenting, validating and managing requirements; ReqPOOL's binding quality standard.Pohl (2010); IREB
ISO/IEC/IEEE 29148International standard for requirements specifications; defines quality criteria such as verifiability and consistency.ISO/IEC/IEEE 29148:2018

Platform support

ReqPOOL Suite: SLIP (Reverse engineering — whitebox) · App2Spec (Reverse engineering — blackbox) · Requirements Manager (Agentic requirements management) · reqChecker (Requirement quality)

SLIP — whitebox reverse engineering

Reads any code base (SAP/ABAP, COBOL/PL/I/JCL, IBM i, NATURAL/ADABAS, Java and more) into source-proven system documentation — every statement verifiable down to program and line in the source code.

SLIP — knowledge graph & AI assistant

Queryable graph of programs, data flows, interfaces, batches and data structures; the agentic assistant answers impact questions in seconds and delivers transformation recommendations per module.

App2Spec — blackbox observation

An AI agent operates the application like a user or observes real users; screens, business rules and end-to-end processes become an evidence-linked AS-IS specification with process diagrams.

Requirements Manager — specification engine

Workshop videos, protocols and ideation documents become structured specification documents; the result is the AI coding intent as a machine-readable implementation order.

reqChecker — requirement quality

Automatic quality checking of requirements with structured feedback and testable rewording proposals — target: 100% testable requirements before the baseline gate.

Platform usage best practices

  • Combine whitebox (SLIP) and blackbox (App2Spec) and validate the code view against the usage view — deviations are analysis findings, not noise.
  • Define reconstruction goals per system before the analysis run (depth, views, artifacts) — full analyses without a target picture create knowledge graveyards.
  • Only adopt extracted business rules with source code or screenshot evidence; the business side confirms the functional meaning, never the engine alone.
  • Manage requirements exclusively in the Requirements Manager — no parallel copies in office documents.
  • Run reqChecker before every review meeting and clean up findings beforehand; reviews should assess content, not formalities.
  • Maintain traceability from day one: every requirement references a process, a system source or a stakeholder decision.

Artifacts & outcomes

  • Reconstructed system documentation (architecture, business rules, data models)
  • As-is process analysis with process mining evidence and weak point catalogue
  • Quality-assured requirements model (functional & non-functional)
  • Traceability matrix requirement ↔ process ↔ system
  • Gap analysis as-is vs. agentic to-be

Quality gate — Analyse

Transition to the next phase happens through a formal quality gate (go/no-go). Gate criteria are documented in the gate review and signed off by the engagement lead.

  • System documentation accepted by architecture and business
  • Requirements checked against ISO 29148 criteria and prioritized
  • Critical business rules functionally confirmed and evidenced
  • Gap analysis released as design input by the engagement lead

Scientific deep dives

Chikofsky & Cross: Reverse Engineering and Design Recovery (IEEE Software 1990)

Foundational taxonomy of reverse engineering; the field's most cited reference.

https://doi.org/10.1109/52.43044
Kazman, Woods & Carrière: CORUM II / horseshoe model (WCRE 1998)

Connecting reengineering and architecture models; basis of architecture reconstruction.

https://doi.org/10.1109/WCRE.1998.723185
van der Aalst: Process Mining — Data Science in Action (Springer 2016)

Standard reference on process mining from discovery to conformance checking.

https://doi.org/10.1007/978-3-662-49851-4
Pohl: Requirements Engineering — Fundamentals, Principles, and Techniques

Comprehensive scientific reference on requirements engineering.

https://link.springer.com/book/9783642125775
IREB: CPRE syllabi & handbooks

Certification standard for requirements engineering; ReqPOOL's quality benchmark.

https://www.ireb.org/en
ISO/IEC/IEEE 29148:2018 — Requirements Engineering

Standard for processes and quality criteria of requirements specifications.

https://www.iso.org/standard/72089.html
IIBA: BABOK Guide v3

Body of knowledge of business analysis with its technique catalogue.

https://www.iiba.org/career-resources/a-business-analysis-professionals-foundation-for-success/babok/

Best-practice guide: Analyse phase

Compact checklist covering process model, methodology mix and platform best practices for use in your engagement.

Download the best-practice guide (PDF)