Find One Valuable Redundancy
Repeated access routes can protect knowledge from specialist barriers when the overlap serves a real work need and is maintained deliberately.
Redundancy is often treated as obvious waste: two copies, two routes, two people who know the same thing. I am not interested in defending duplication in general, but some overlap increases exposure and internalization because people are more likely to encounter and understand knowledge when it appears where their work already happens.
The exercise is deliberately small. Spend ten minutes identifying one valuable piece of information trapped in a single system that requires specialized skills to access. Who else would benefit from seeing it? Where do those people already look for information? Which system could surface the same material in a different context? Document the information, the present barrier, and one potential redundant location.
The word valuable carries most of the weight here. This is not a license to copy every record into every application until no one knows which version is current; the redundancy should already help work move, or clearly prevent a repeated failure. It might be two retrieval routes into a project record, overlapping expertise that protects a critical decision, or parallel records serving different legitimate audiences.
Imagine a maintenance lesson held inside a specialist project system. The facilities team cannot reach it without a particular account and a vocabulary they do not use. They may instead look in a maintenance library, a handover record, or a local work-order system. The existing single route preserves the lesson for the original project team while making it effectively invisible to the people who will encounter the condition later. A second route, maintained with clear provenance, can make the same knowledge available where the next user will actually search.
Tracing the two routes matters. Who uses the project system, and what do they need from the record? Who uses the maintenance library, and what part of the record changes their work? What breaks if the second route does not exist? The answer might be repeated diagnosis, a delayed repair, an unnecessary request to a former team member, or a solution reinvented because the first solution stayed behind a specialist barrier.
Do this twice a week for four weeks. Eight iterations produce a portfolio of integration opportunities. That portfolio is more useful than an abstract debate about whether redundancy is efficient, because each case identifies a piece of knowledge, a barrier, an audience, and a possible new location. Together, the cases can become a systems-integration roadmap.
Resistance will appear. Stakeholders may say that duplication creates inefficiency. Sometimes it does. The response is not to declare redundancy always good; it is to document the cases where the absence of overlap causes people to reinvent solutions because they cannot find the existing knowledge. Usage patterns prove the value more clearly than a theoretical argument about clean architecture.
The record must distinguish copies from routes. A route may point from a maintenance system to an authoritative project record, while providing the contextual information that makes the connection discoverable. A maintained copy may be appropriate when the user needs a local working version. In either case, the design should identify which record is authoritative, what is synchronized, who owns the connection, and what users should do when the two surfaces disagree.
Redundancy also applies to expertise. If one person is the only route into a process, the organization has an access problem even if every document is perfectly organized. Training another person, pairing roles, or creating a shared practice may be a valuable redundancy because it keeps work moving when the original expert is unavailable.
Before the familiar argument about duplicate storage begins, write down the cost of losing the route. The answer has a useful way of clarifying whether we are discussing waste or resilience.
The exercise ends with one record, one barrier, and one second route that a real colleague would use during ordinary work. Trace who uses each route, name what fails without it, and identify which record remains authoritative. That evidence makes the distinction between valuable overlap and ordinary confusion much easier to argue—and, if the second route proves useless, much easier to remove.
