Available at: https://dacs.datavaultalliance.com/documents/dacs-manifesto/

Doctrine statement: Defensible Analytics requires an organization to be able to show why a material decision was permitted to rely on the evidence it used, for the stated use, under the conditions that existed at the time.

Status: v1.18 Release Candidate. This manifesto explains the doctrine behind DACS. It is an informative companion doctrine and is not the formal conformance specification.

Founding Authors and Committee Leads:
Daniel Linstedt – CEO, DataVaultAlliance Holdings, LLC
Cindi Meyersohn – COO, DataVaultAlliance Holdings, LLC

When a material decision is challenged, the organization should be able to reconstruct the evidence, meaning, identity, relationships, context, authority, controls, and permitted use that supported it.

DataVaultAlliance maintains the authoritative DACS Specification and official DACS publications. Redistributed copies may become outdated. Verify the current authoritative version at dacs.datavaultalliance.com before relying on this publication.

1. Executive Statement

Enterprises do not lack data or AI. They lack accountable decision paths.

Modern organizations can usually show where data came from, how it moved, what a model returned, which API was called, or what an automated workflow executed. They are often less prepared to show why a material decision was permitted to rely on the identities, meanings, relationships, context, authority, and evidence embedded in that execution.

That distinction matters because a technically successful decision path can still be based on ambiguous business meaning, unstable identity, an invalid or misoriented relationship, stale context, missing authority, or evidence that was available but not eligible for the stated use.

The problem is not that catalogs, lineage, semantic layers, governance programs, data-quality controls, observability tools, AI platforms, or guardrails are ineffective. The control gap appears when their presence or output is treated as sufficient proof that a material decision path had the evidence, authority, controls, and conditions required for permitted use.

DACS stands for Defensible Analytics Control Specification.

DACS is a vendor-neutral, implementation-neutral control specification for decision-evidence readiness within a stated decision scope. It is not a product category, software platform, consulting slogan, replacement architecture, data model, governance program, or AI tool.

DACS defines testable controls for determining whether the material evidence behind a decision may be relied upon for a stated use and whether the basis of that decision can later be reconstructed, inspected, challenged, and defended.

Defensible Analytics is the intended outcome: a decision path whose material evidence, meaning, identity, relationships, context, authority, control results, constraints, and permitted use can be reconstructed for the scope in which the decision was made.