Application Portfolio Rationalisation

Most large estates contain more applications than anyone believes, and a substantial fraction do overlapping things — four ticketing systems, three reporting tools, two CRMs from an acquisition that completed six years ago. Nobody chose this; it accumulated, and no individual decision was wrong.

The standard analysis plots each application on two axes: business value and technical quality. High value and high quality is invested in. High value and poor quality is the modernisation backlog. Low value and high quality is tolerated or consolidated. Low value and poor quality is retired, and this quadrant is where the money is.

The disciplines that make the exercise credible rather than a slide. Use evidence — actual usage telemetry, licence consumption, support ticket volume, incident history and total cost including infrastructure, licences and the people who maintain it — because self-reported business criticality is always high. Include the hidden estate of departmental tools and spreadsheets that run real processes. And treat total cost of ownership per application as the number that drives the conversation, since it is usually two to three times what anyone expects.

The step that determines whether any value is realised is decommissioning, which is unglamorous, gets cut when the programme runs late, and is the only part that produces the saving. Fund it explicitly and track retirements as the metric, not assessments completed.