Problem

Major programmes do not fail because teams lack data. They fail because the data that matters is fragmented across documents, schedules, risks, requirements, assurance evidence, delivery plans, operating models, and lessons learned. Even where data exists, reporting often strips away the relationships, assumptions, lineage, and consequences that give the data meaning.

This creates two related problems: poor quality data and poor reporting context. Information may be incomplete, inconsistent, stale, or structured differently across functions, while reporting tends to aggregate it linearly into registers, dashboards, and status packs. These artefacts are useful, but they can flatten the programme into separate lists rather than showing how the system behaves.

Challenge

Senior leaders need to understand what has changed, what is connected, what is exposed, and what decision is required. Conventional reporting can struggle to answer these questions because the most important project information sits in relationships: dependencies, ownership, evidence chains, common causes, downstream effects, and cross-portfolio patterns.

The challenge is to move from reporting data to reasoning over connected project knowledge. A knowledge graph helps by modelling the real-world things in a programme — documents, risks, people, systems, contracts, milestones, requirements, and decisions — and the relationships between them.

Solution

Kaleido developed graph-enabled applications to demonstrate how fragmented project information can become a connected decision system. We have trialed graph builds in Neo4j, Palantir Foundry, and proprietary applications for bespoke use cases and AI integrators, testing how connected project data can be modelled, queried, and made usable through different enterprise and bespoke technology routes.

Two Kaleido-developed examples show the capability:

  • Connected document intelligence: an application that maps complex engineering and project documentation, identifies stale or inconsistent references, traces the downstream effect of document changes, and lets users ask natural questions across the document estate.
  • Capital portfolio intelligence: an application that connects structured project controls data — including risks, owners, mitigations, projects, schedule activities, controls, and standards — to reveal portfolio-level patterns, dependencies, and interventions across delivery.

Both examples show the same pattern: connect the minimum useful set of source information, model the relationships that matter, use AI to help users retrieve the right evidence, and present outputs in an application that supports real project decisions.

Conclusion

This case study demonstrates Kaleido’s ability to move beyond static reporting into decision intelligence. By combining knowledge graphs, AI-assisted retrieval, controlled agents, and project controls expertise, we help clients understand not only what the data says, but how the system fits together and where action is likely to have the greatest value.