What Is Atomic Resolution?

Atomic Resolution is the minimum verifiable unit of completed work that forms the foundation of Resolution as a Service (RaaS) pricing. It is the precise answer to the question every vendor must answer before any outcome-based pricing conversation can begin: what exactly counts as an outcome?

The Three-Criteria Test

An Atomic Resolution must satisfy three criteria simultaneously. A unit of work that fails any one of the three tests is not an Atomic Resolution and cannot be billed as one.

Verifiable. A problem was solved, not merely attempted. The resolution produces an objectively confirmable output: a ticket closed, a document generated, a transaction completed, a compliance check passed. Partial responses, tasks in progress, and tickets closed as “won’t fix” do not qualify. The completion state must be defined before pricing begins, not negotiated after the invoice arrives.

Attributable. The resolution is traceable to the platform’s AI execution, not to a human override, a concurrent manual process, or an external system change. Attribution is the commercial foundation of outcome-based billing. If a vendor cannot prove that its AI produced the result, it cannot defend the charge at renewal. Human escalations, where a human agent completes or significantly modifies the AI’s output, are excluded from Atomic Resolution billing and must be tracked separately.

Finite. The resolution has a clear, predefined endpoint. Open-ended agentic loops, where an AI agent continues processing without a defined completion state, are structurally incompatible with per-resolution pricing because they generate unlimited compute cost under a fixed fee. Each Atomic Resolution type must define its finite boundary before it can be priced: a single ticket thread, a single document, a single transaction record.

Why the Definition Matters Commercially

The distinction between activity and outcome is the difference between a platform that can sustain outcome-based pricing and one that cannot.

“The AI drafted a response” is activity. “The ticket was resolved without human intervention within the defined SLA” is an Atomic Resolution. The gap between those two descriptions is the gap between an invoice that holds up at renewal and one that does not.

Most SaaS vendors, when they first attempt to define their Atomic Resolutions, discover that their natural-language answers describe what the AI did rather than what it achieved. This is not a failure of the product. It is a failure of the measurement infrastructure. The product may be delivering genuine value. The platform cannot yet prove it with the precision that billing requires.

Running the three-criteria test rigorously against each candidate resolution type also reveals something more uncomfortable: some workflows, when examined honestly, can be fully replicated by a general-purpose AI agent without touching the platform at all. Those workflows are not Atomic Resolution candidates. They are product strategy questions. Identifying them before outcome-based pricing infrastructure is committed is among the most valuable outputs of this exercise.

The 48-Hour Quality Gate

A resolved ticket that is reopened by the end user within 48 hours is a presumptive resolution failure. CPAG’s Resolution Guard framework establishes a 48-Hour Quality Gate as a standard post-resolution monitoring window: if a closed resolution is reopened within that window, it is treated as a failed resolution and is not billable at the full resolution rate.

The 48-Hour Quality Gate is an operationalization of the verifiability criterion. It acknowledges that “closed” and “resolved” are not synonymous. A resolution that does not hold for 48 hours was not a resolution. It was a temporary state change that transferred the problem back to the customer.

Building the Atomic Resolution Catalog

Before any pricing conversation begins, a vendor must build an Atomic Resolution Catalog: a documented list of the discrete, verifiable outcomes the platform actually delivers, each tested against the three criteria, each with a defined completion state, attribution rule, finite boundary, and exclusion criteria.

The Vendor Transition Playbook defines the following fields for each catalog entry:

Resolution Name. Plain-language description of the outcome. Example: Tier-1 Support Ticket Resolved.

Completion Criteria. The specific, objectively verifiable state that constitutes completion. Example: Ticket status is Closed AND no reopen within 48 hours.

Attribution Rule. How the platform proves its AI executed the resolution. Example: Agent action log shows 80% or more of the response generated by the model with no human edit.

Finite Boundary. What ends the resolution scope. Example: Single ticket thread; multi-ticket issues split into separate resolutions.

Exclusion Criteria. What does not count as a resolution. Example: Human escalation, partial responses, tickets closed as won’t fix.

Estimated Cost to Serve. Compute plus orchestration cost per resolution, for 1-to-4 Rule Rule validation.

The CPAG standard is five to ten well-defined Atomic Resolutions before proceeding to a hybrid pricing pilot. Fewer than five produces insufficient data. More than ten produces a pilot too complex to yield clean unit economics. A focused catalog with precise definitions outperforms a broad catalog with ambiguous ones every time.

Atomic Resolution and the 1-to-4 Rule

An Atomic Resolution is not only the unit of billing. It is the unit of cost measurement. The 1-to-4 Rule requires that revenue per resolution equals or exceeds four times the AI cost to serve that resolution. Applying the rule requires knowing the cost-to-serve at the resolution-type level, which in turn requires precisely defined Atomic Resolutions.

A vendor who defines resolutions loosely, bundling multiple outcomes into a single billable event, cannot measure cost-to-serve accurately and cannot apply the 1-to-4 Rule. The result is a platform that appears healthy at the aggregate margin level while specific resolution types or specific enterprise customers are generating negative margins at scale.

Atomic Resolution is defined and operationalized as part of the broader Resolution as a Service (RaaS) framework developed by Crown Point Advisory Group. The full category argument is in the RaaS Manifesto. The operational implementation guide, including the Atomic Resolution Catalog template and the Resolution Architecture Audit, is in the Vendor Transition Playbook.