Cloud & Infrastructure · Legacy Modernization · Hard · 35-45 min

Replacing a 20-Year-Old ERP

Replacing a 20-Year-Old ERP is a hard Cloud & Infrastructure legacy modernization case interview that runs 35-45 min. A manufacturer runs a heavily customized 20-year-old ERP that no vendor will support past three years. A strong answer works through 5 phases: Establish the real driver; Lay out the options; Assess why big-bang ERP programs fail; Address data migration, process re-design (standard vs customized), and organizational change; Recommend an approach with phasing and a fallback, sized against the company's risk tolerance.

Last updated 2026-09-05

The brief

A manufacturer runs a heavily customized 20-year-old ERP that no vendor will support past three years. Finance, supply chain, and production all depend on it, and prior 'big-bang' replacement attempts at peer firms have failed spectacularly. The board wants a modernization recommendation: replace, re-platform, or wrap-and-extend — with a credible plan that doesn't bet the company.

How to approach it

  1. Establish the real driver — vendor end-of-support deadline and accumulated customization debt
  2. Lay out the options — full replacement (new SaaS ERP), re-platform/upgrade, or strangler-pattern wrap-and-extend
  3. Assess why big-bang ERP programs fail — scope, customization, change management — and what de-risks them
  4. Address data migration, process re-design (standard vs customized), and organizational change
  5. Recommend an approach with phasing and a fallback, sized against the company's risk tolerance

What a strong answer does

  • Frames standardize-vs-customize as the central decision — most ERP failures come from re-creating old customizations
  • Considers a phased / strangler approach instead of defaulting to big-bang
  • Treats change management and process re-design as the bulk of the work, not the software license
  • Plans data migration and parallel running explicitly
  • Matches the recommendation to the company's appetite for operational risk

Red flags interviewers score down

  • Recommends a big-bang cutover with no phasing or fallback
  • Assumes the new ERP should replicate every existing customization
  • Treats it as a software purchase rather than a business-process transformation
  • Ignores data migration and the cost of running two systems in parallel

Cases are written in each firm's style, written and reviewed by working consultants; they are not the firms' own published cases.

More Cloud & Infrastructure cases