All recorded decisions organized and searchable so that relevant prior decisions can be surfaced when related situations arise. The right people know what has been decided, why, and by whom, without having to ask.
Every organization makes decisions constantly. The problem is not the volume of decisions. It is that most decisions leave no findable record. A policy gets established in a meeting. The meeting notes are shared with some of the people who needed to know. The policy governs behavior for years. Nobody can find where it came from or who authorized it. When a related situation arises, the organization either rediscovers the old decision by accident or makes the same decision again as if for the first time. The Decision Index captures significant decisions: the ones that change direction, establish policy, commit resources, or resolve contested questions. Each record holds what was decided, when, by whom, with what rationale, and what alternatives were considered. The rationale is the part most organizations skip. It is also the part that makes the index useful six months later.
The Decision Index is a core dimension of the Awareness phase in the Ragsdale Framework for Autonomization, developed by Marc Ragsdale. Prospus implements it as part of structured AI transformation engagements. Kaamfu operationalizes it as a native layer inside the Autonomous Operating Environment, giving AI agents access to organizational decision history.
Decisions are made in conversations and meetings with no structured record. Retrieving a prior decision requires asking the people who were in the room. Policy conflicts surface as disputes rather than lookups. AI agents have no access to decision history.
Significant decisions are recorded in a structured, searchable index. Prior decisions can be surfaced when related situations arise. Policy conflicts are resolved by looking up what was decided rather than by re-litigating the original question. AI agents surface relevant prior decisions to inform current ones.
| Metric | Before | After |
|---|---|---|
| Retrieval | Retrieving a prior decision requires knowing who was in the room and hoping they remember. A new team member needs to understand why a vendor was chosen over alternatives. The decision was made before they joined. Nobody can point them to a document. |
Prior decisions are surfaced by querying the index. The record includes what was decided, why, and by whom. |
| Repeat Decisions | The organization makes the same decision multiple times because nobody can find the record of the first time. A policy on contractor access to production systems is established, forgotten, and re-established three times in four years. Each iteration creates a different policy. |
Before a significant decision is made, the index is queried for prior decisions on related topics. Repeat decisions are flagged. |
| Policy Conflicts | Contradictory policies exist because different decisions were made by different people in different contexts with no visibility into what already existed. Two teams operate under contradictory expense approval policies. Both policies were established by legitimate decision-makers. Neither decision-maker knew the other existed. |
New decisions are checked against prior decisions in the index. Conflicts are visible before they compound. |
| AI Context | AI agents making or supporting decisions have no access to organizational decision history. Every decision is made without context. | AI agents query the Decision Index before supporting decisions. Prior decisions, rationale, and alternatives are available as context. |
| Rationale Preservation | Decisions leave records. Rationale does not. Future readers know what was decided but not whether the conditions that justified it still apply. | Every decision record includes rationale and alternatives considered. Future readers can assess whether prior decisions should still govern current situations. |
A score of 10 on the Decision Index means this dimension is fully resolved and no longer a constraint on the phases that follow. Here is what that requires in practice.
Define the threshold. Not every decision goes in the index. Establish what qualifies: direction changes, policy establishment, resource commitments, and resolution of contested questions.
Build the record structure. Every decision entry captures the decision statement, date, decision-maker, rationale, and alternatives considered. Rationale is required, not optional.
Establish the retrieval architecture. Tag decisions by domain, topic, and the situations they govern. The index is only useful if prior decisions can be surfaced by situation, not just by date or author.
Integrate the index into the decision-making process. Before significant decisions are made, the index is queried for prior decisions on related topics.
Make the index queryable. Inside Kaamfu for organizations on the platform, or as a standalone natural language interface for those that are not.
Decisions get recorded but rationale does not. A record that says what was decided but not why is a list, not an index. The value of the Decision Index is in the rationale and alternatives fields. Without them, the index can tell you that a decision was made but cannot tell you whether the conditions that justified it still apply.
The Autonomy Diagnostic scores the Decision Index on a 0 to 10 scale. Each point reflects a specific observable state in your organization.
No structured decision record exists. Prior decisions are retrieved by asking the people who were in the room.
Some decisions are recorded in meeting notes but with no structure, no rationale, and no retrieval architecture.
Significant decisions are documented informally. Rationale is absent. Retrieval requires knowing when the decision was made.
A decision log exists with what was decided and when. Rationale and alternatives considered are not captured.
Decisions are recorded with decision-maker and rationale. Retrieval architecture is minimal. Searching by topic requires effort.
A structured Decision Index exists with rationale and tagging. Most significant decisions are captured but coverage is inconsistent.
Significant decisions are captured consistently with decision statement, rationale, alternatives, and tagging. Prior decisions can be surfaced by topic.
The Decision Index is complete and current. Policy conflicts are checked against the index before new decisions are finalized.
The Decision Index is queryable and referenced by AI agents before decisions are supported or made.
All significant decisions are captured with full rationale, alternatives, and traceability. AI agents surface prior decisions to inform current ones. Repeat decisions and policy conflicts are structurally prevented.
Direction changes, policy establishment, resource commitments, and resolution of contested questions qualify. The threshold needs to be defined for the organization. Too broad and the index becomes noise. Too narrow and it misses the decisions that matter.
A decision record without rationale tells you what was decided but not whether the conditions that justified it still apply. Six months later, when a related situation arises, the rationale is what determines whether the prior decision should govern the new situation or whether circumstances have changed enough to warrant a new one.
Meeting notes capture everything that happened. The Decision Index captures only what was decided, why, and by whom, in a structured form that is searchable by topic and situation. Meeting notes are a source. The Decision Index is the structured record built from that source.
Before a significant decision is made, the index is queried for prior decisions on related topics. If a prior decision exists, it surfaces automatically. The organization can then assess whether the prior decision still applies or whether circumstances justify revisiting it.
AI agents supporting or making decisions need access to organizational decision history. Without the Decision Index, every decision is made without context. With it, agents can surface prior decisions, flag contradictions, and provide rationale from prior decision-making before a new decision is finalized.
Access governance defines who can see what. Confidential decisions are captured with restricted access. The index knows they exist. Only those with appropriate access can see the content.
Superseded decisions are marked as superseded with a link to the decision that replaced them. The record is preserved for historical context. The index surfaces the current governing decision first and the superseded decision as context.
One person or role is responsible for maintaining the index, enforcing the record structure, and auditing whether significant decisions are being captured. Without a defined owner, the index reflects only the decisions that someone happened to document.
A useful index can be built quickly if the organization starts from the decisions that currently govern its operations and works backward. The goal is not a complete historical record. It is a current record that captures the decisions shaping organizational behavior today.
A score of 7 or above means significant decisions are being captured with rationale and can be surfaced when related situations arise. The Autonomy Diagnostic will tell you which dimensions need the most attention first.
The Autonomy Diagnostic scores every dimension of the Ragsdale Framework and tells you exactly where to focus first.
Take the Diagnostic