The Knowledge Tiers

Not all knowledge ages the same way

2–3 minutes

Product knowledge isn’t one thing. It has different shelf lives, different ways of breaking down, and different governance, depending on how long it’s meant to last.

Most organizations don’t think about how long knowledge is meant to last when deciding how to manage it. That’s what makes the investment imbalance so hard to see.

Organizing knowledge internally by durability reveals where most companies are under-invested and why.

The Three Tiers

Ephemeral knowledge is generated in real time, in the flow of work. Slack threads, meeting decisions, git pull request review comments, verbal clarifications from a subject matter expert, design rationale captured in a Figma comment. This tier contains some of the richest context in the system — the reasoning behind decisions, the tradeoffs that were weighed, the alternatives that were rejected. It also vanishes before anyone decides it was worth keeping. No pipeline promotes what matters into a more durable form. The capture problem is rarely owned by anyone.

Versioned knowledge is tied to a specific product version or release cycle. API reference documentation, release notes, changelogs, user guides, migration guides. This is the most visible tier because it has deadlines, ships alongside the product, and its absence isn’ticed immediately. It’s where DDLC pipelines, CMS and CCMS investment, and review workflows concentrate. Relative to the other two tiers, it’s well-managed. But only this tier is.

Evergreen knowledge is intended to remain valid over time. Architecture overviews, core concept glossaries, onboarding guides, style and terminology guides, API conceptual guides. This content is created carefully at the start, then left alone. Nobody retires it. It’s rarely audited. It sits alongside current content with no signal to readers that it may have become obsolete or out of date. The gap isn’t in the creation but in the governance that never follows.

The Investment Imbalance

Companies concentrate tooling, process, and governance in the versioned tier because it’s directly tied to shipping. It’s visible. It has deadlines. The other two tiers are treated as self-managing — ephemeral knowledge is considered disposable, and evergreen knowledge never makes it onto a maintenance schedule.

Neither holds. Ephemeral knowledge contains context that, once lost, can’t be reconstructed. Evergreen knowledge drifts silently until it misleads rather than guides.

The imbalance isn’t primarily a resource problem. It’s a design problem. The system was never designed to govern all three tiers, only one.

How the Tiers Interact

The tiers aren’t independent. Ephemeral knowledge is the raw material that feeds the versioned and evergreen tiers when capture pipelines exist to move it. Versioned knowledge is built on evergreen foundations. When those foundations drift, versioned content loses its conceptual anchor.

A knowledge system that manages only the versioned tier while leaving the other two ungoverned is managing the output while ignoring the inputs and the infrastructure those outputs depend on.

Each tier is examined in depth in its own article: