Definition
The BVAC Maturity Stages are the five-stage scoring scale shared across all eight dimensions of theBrand Visibility for Agentic Commerce (BVAC) Framework, developed byGreg Kihlström of The Agile Brand. The stages — Invisible, Discoverable, Comparable, Differentiated, and Agent-native — describe a brand’s progression from structural absence in agent-readable data through full participation in agent-to-agent commerce (Kihlström, 2026).
The scale is one of the framework’s two scoring mechanics, paired with the prerequisite-cap-and-floor logic. Every dimension scores on the same five-stage scale, which makes scores comparable across dimensions and allows a brand to see its profile as a coherent picture rather than as eight unrelated metrics.
A brand can sit at different stages on different dimensions. Most do. The composite framework score is determined by the lowest effective score across all dimensions — a brand is only as agent-ready as its weakest dimension allows.
Why a Shared Five-Stage Scale
The BVAC Framework’s eight dimensions span three distinct dimension types — brand-facing (Identity Legibility, Attribute Completeness, Differentiation Encoding, Brand-Agent Representation, Trust Signal Density), technical-implementation (Protocol Readiness, Latency and Data Freshness), and operating-model (Governance Maturity). They measure different things in different modes. A shared scoring scale across all eight is a non-obvious design choice.
The choice is structural. A unified scale produces three properties the framework depends on:
- Comparable scores across dimensions. A Comparable on Differentiation Encoding means the same maturity position as a Comparable on Protocol Readiness — both describe attribute parity with the category, just expressed in their respective modes. Without a shared scale, the framework would produce eight scores in eight scoring systems and no way to identify the binding constraint on the composite.
- A coherent profile rather than a heatmap. A brand at Discoverable on prerequisites and Differentiated on three strategic dimensions has a recognizable shape. The shape carries diagnostic meaning that wouldn’t survive translation to dimension-specific scales.
- Forced sequencing under the caps. The prerequisite cap and the Trust Signal Density floor cap both depend on stages being comparable. The cap mechanic — “effective score is the lower of diagnostic and prerequisite stage” — only works if the stages are the same scale.
The framework’s scoring discipline scores down on borderline cases. When a dimension sits between two stages, the default is to score to the lower stage and note the partial progress in the gap map. Scoring up risks declaring readiness the brand hasn’t earned, and under the caps that mistake propagates across the strategic tier.
The Five Stages
The five stages function as a shared scale across all eight dimensions. The shared definitions:
Invisible. Structural absence. The brand doesn’t exist in the data form agents require. This isn’t a low score on a graded scale; it’s the absence of the surface entirely. A brand at Invisible on Identity Legibility has no resolvable entity declarations. A brand at Invisible on Protocol Readiness has no standardized protocol surface. A brand at Invisible on Governance Maturity has no defined operating model for agentic commerce. The mode varies; the property — structural absence — is constant.
Discoverable. Present and parseable. Agents can find and read the brand, but it competes on default attributes only. The minimum entry point. A brand at Discoverable has crossed the threshold of existing in agent-readable form but hasn’t yet differentiated from the long tail. For Trust Signal Density specifically, Discoverable is what crossing the floor produces — at least one of the three trust signal types exists in structured form, even if minimally.
Comparable. Attribute parity with the category. The brand can be evaluated against competitors on standard terms. This is the eligibility-for-inclusion stage. The brand has the category-standard data, the structured policies, the basic protocol surfaces, the formal governance structure — whatever each dimension requires for the brand to participate in agent-mediated comparison on equal footing with category competitors.
Differentiated. Unique value propositions are encoded in structured, verifiable form. Premium signals are legible to agents. The brand exceeds category baseline in observable ways. For brand-facing dimensions, this means unique attributes, verified claims, and brand context carried in structured form. For technical-implementation dimensions, this means mature implementations with version concurrency, cross-protocol coherence, and category-leading performance. For operating-model dimensions, this means proactive policy evolution and risk-calibrated authority frameworks.
Agent-native. The brand operates its own agent, exposes skills via standard protocols, and participates in A2A interactions on equal footing. The fullest stage of agentic commerce maturity. The brand is recognized and cited by major agents as an authoritative source, its protocol surface functions as a category reference, its governance integrates with agent operations in near real time, and its trust signals refresh continuously. Agent-native isn’t an aspirational label — it’s an operating state most brands won’t reach within the next several years, and the framework treats it as describable but not yet broadly achievable.
How Stage Observability Varies by Dimension Type
The same five stages apply across all eight dimensions, but the mode of observability varies by dimension type. The stages describe observable conditions; what is observed differs.
Brand-facing dimensions (Identity Legibility, Attribute Completeness, Differentiation Encoding, Brand-Agent Representation, Trust Signal Density). Stage observability is data-surface observability — schema is present, data is populated, identifiers resolve, claims are encoded with verification anchors. Scoring happens at the SKU level for catalog-dependent dimensions, against the top 20 SKUs by revenue.
Technical-implementation dimensions (Protocol Readiness, Latency and Data Freshness). Stage observability is technical observability — endpoints are exposed, P95 latency is measured, protocol versions are supported, cross-protocol coherence is monitored. Scoring depends on instrumentation: a brand without P50/P95/P99 measurement at protocol endpoints can’t even score the response-latency sub-component because the data doesn’t exist.
Operating-model dimensions (Governance Maturity). Stage observability is procedural observability — playbooks exist, policies are reviewed on cadence, cross-functional coordination operates. The Governance Maturity dimension’s stages lean on procedural observability rather than the data-surface or technical observability used elsewhere, and the framework’s template absorbs this without revision. The stages still describe observable conditions, in a different mode of observation.
The shared scale holds across all three modes. A brand at Comparable on Governance Maturity has the same relative maturity position as a brand at Comparable on Differentiation Encoding or Latency and Data Freshness — category baseline, eligible for comparison — even though one is observed through policies and cadence, the second through structured fields, and the third through P95 measurements.
Diagnostic Score vs. Effective Score
The framework produces two scores per strategic dimension on the same five-stage scale, and the distinction is central to how the framework operates.
Diagnostic score. The stage the dimension reaches in isolation, based on its own assessment criteria. The diagnostic score answers the question: how mature is this dimension’s underlying work, considered on its own merits? A brand can have excellent Differentiation Encoding work — structured USPs, verification anchors, premium signal encoding — and score Differentiated on the dimension diagnostically.
Effective score. The diagnostic score capped by the lower of the two prerequisite stages, and further capped at Discoverable if Trust Signal Density is below the floor. The effective score answers the question: how mature is this dimension in terms of what actually reaches an agent?
The gap between diagnostic and effective is the framework’s readout of how much foundational work is required. A brand at Differentiated diagnostic and Comparable effective on Differentiation Encoding is doing excellent differentiation work that’s being held back by an upstream constraint. The differentiation work isn’t wasted; it’s stranded until the constraint clears.
This distinction matters because it changes which intervention is correct. A brand looking at diagnostic scores alone would invest in the lowest diagnostic-scoring dimension. A brand looking at the diagnostic-effective gap sees that the binding constraint is upstream — typically a prerequisite gap or below-floor trust status — and the correct intervention is to lift the cap rather than to keep building above it.
The prerequisites (Identity Legibility and Attribute Completeness) are scored without an effective score, because they aren’t capped by anything else in the framework. Their score is the cap that applies to the strategic tier.
The Composite Framework Score
The composite framework score is the lowest effective score across all eight dimensions. It carries the framework’s central scoring principle: a brand is only as agent-ready as its weakest dimension allows.
The composite uses lowest-effective rather than average-effective for a specific reason. Agents filter on what they can read. A brand with seven dimensions at Differentiated and one dimension at Discoverable doesn’t behave like a brand at the average of those stages. It behaves like a brand at Discoverable, because the Discoverable dimension is the binding constraint on what reaches the agent. Averaging would smooth over the constraint and misdirect remediation.
The composite stage is the single value that summarizes the brand’s agent readiness. The brand-stage examples:
- A brand with all eight effective scores at Comparable scores Comparable composite — eligible for category comparison across the board.
- A brand with seven scores at Differentiated and one at Discoverable scores Discoverable composite — held back by the one dimension.
- A brand below the trust floor with strong work elsewhere scores Invisible composite — the floor caps the entire strategic tier at Discoverable and the prerequisite cap or the floor cap produces an Invisible effective somewhere.
The composite changes when the binding constraint changes. A brand that lifts its lowest effective score from Discoverable to Comparable moves the composite to Comparable — unless another dimension is also at Discoverable, in which case the composite holds at Discoverable until that one moves too.
This is why the framework’s remediation roadmap is dependency-ordered rather than horizon-ordered. The horizon view shows what’s easiest; the dependency view shows what lifts the composite. The two are often different gaps.
How the Caps Work in Practice
The cap mechanics are easier to see through worked patterns than through abstract definition. Three illustrative patterns:
Pattern 1: The strong work above weak foundations. A brand has invested heavily in agentic commerce. Differentiation Encoding scores Differentiated diagnostically — every USP is encoded, verification anchors are in place, premium signals are legible. Brand-Agent Representation scores Comparable diagnostically — Agent Card published, multiple skills invocable, authentication in place. But Attribute Completeness sits at Discoverable: MerchantReturnPolicy isn’t marked up and several category-standard attributes are missing. The prerequisite cap holds every strategic dimension’s effective score at Discoverable, regardless of the diagnostic scores. The composite is Discoverable. The remediation isn’t more differentiation work — it’s the unglamorous schema work that lifts the prerequisite cap.
Pattern 2: The floor that swallows the tier. A brand has cleaner prerequisites — Identity Legibility and Attribute Completeness both at Comparable. Differentiation Encoding scores Comparable diagnostically. Protocol Readiness scores Comparable diagnostically. But the brand sits below the Trust Signal Density floor: no structured Review entities, no certification surface in schema, no sameAs authority anchors. The floor caps every other strategic dimension’s effective score at Discoverable, even though the prerequisites would have allowed Comparable. The composite is Discoverable. The remediation is crossing the floor — typically a single 90-day schema action that lifts the cap from the entire strategic tier at once.
Pattern 3: The binding constraint above a clean foundation. A brand has cleared the prerequisites (both at Differentiated) and crossed the floor (Trust Signal Density at Comparable). Five strategic dimensions score Comparable effective. Governance Maturity sits at Discoverable — no defined ownership, no authority scope, no cross-functional coordination. The composite is Discoverable, because Governance Maturity is the binding constraint. The remediation is the operating-model work that lifts the weakest strategic dimension, which then exposes the next binding constraint.
The cap mechanics produce a remediation sequence that’s fixed by dependency: prerequisites first, floor second, strategic dimensions in binding-constraint order. Work performed out of order produces diagnostic improvement with no effective improvement.
How to Score on the Maturity Scale
Scoring is per dimension, with each dimension’s specific stage definitions applied. The framework’s assessment methodology specifies the scoring sequence and the data requirements per dimension.
The scoring sequence:
- Score the prerequisites first. Identity Legibility and Attribute Completeness each receive a stage score. Prerequisites have no effective score — their score is what caps the strategic tier.
- Check the Trust Signal Density floor. Before scoring strategic dimensions diagnostically, run the floor threshold check. If the brand sits below the floor (none of the three structured signal types present), record below-floor status. Below-floor status caps strategic effective scores at Discoverable.
- Score each strategic dimension diagnostically. Apply the dimension-specific assessment method to produce a diagnostic stage score independent of caps.
- Apply the caps. The effective score for each strategic dimension is the diagnostic score capped by the lower prerequisite stage, then additionally capped at Discoverable if Trust Signal Density is below the floor. The binding cap is the lower of the two.
- Compute the composite. The composite framework score is the lowest effective score across all dimensions.
Scoring discipline:
- Score down on borderline cases. When a dimension sits between two stages, score to the lower stage and note the partial progress in the gap map.
- Score at the right granularity. Catalog-dependent dimensions score at the SKU level (default: top 20 by revenue), not at the aggregate level. SKU-level scoring catches the variance that aggregate-level scoring masks.
- Record model versions. For dimensions assessed through agent query simulation (Differentiation Encoding, Brand-Agent Representation, Trust Signal Density), agent behavior shifts across model releases. Recording the model version turns a release into a recalibration event rather than unexplained drift.
- Use the same procedure for self-administered and consultant-led assessments. The shared scoring rubric ensures consistency. Consultant-led mode brings deeper diagnostic methodology; the stage definitions remain the same.
Comparison to Similar Maturity Models
| Model | Focus | Relationship to BVAC Maturity Stages |
| Capability Maturity Model Integration (CMMI) | Process improvement maturity across organizational capabilities | Both use a staged progression; CMMI applies to general organizational capability, BVAC applies specifically to agent-readiness of the data and operating surfaces |
| Digital Experience Maturity | Stages of digital customer experience capability | Adjacent; BVAC scores a different surface (agent-readable data) than digital experience maturity (human-readable experience) |
| Experience Maturity | Stages of customer experience capability | Adjacent; both stage progression toward fuller capability, with different scopes |
| Marketing Operations Maturity Models | Stages of marketing operations capability | Operating-model adjacent to BVAC’s Governance Maturity dimension specifically |
| Digital Shelf Maturity | Stages of product content quality and availability | Overlaps with BVAC’s Attribute Completeness and Identity Legibility prerequisites; BVAC extends to strategic dimensions digital shelf maturity doesn’t address |
The BVAC Maturity Stages are distinct from most general maturity models in three ways. First, they’re agent-specific — the stages describe maturity in agent-mediated commerce specifically, not general digital or experience capability. Second, they combine with the prerequisite cap and floor mechanics to produce diagnostic and effective scores, not just a single staged readout. Third, they apply uniformly across all eight dimensions despite the dimensions measuring different things in different modes.
Best Practices
- Score every dimension on the same scale. The cap mechanics depend on stages being comparable across dimensions. Resist the temptation to invent dimension-specific scoring rubrics.
- Score down on borderline cases. When a dimension sits between two stages, score to the lower stage and note the partial progress in the gap map. Under the caps, scoring up declares readiness the brand hasn’t earned and propagates the mistake across the strategic tier.
- Distinguish diagnostic from effective scores in the readout. Presenting only effective scores hides the strategic work that’s been done; presenting only diagnostic scores hides the gap that’s holding it back. The framework’s scorecard format expresses both.
- Read the composite as the binding constraint. The composite stage isn’t a summary statistic — it identifies the dimension that’s holding the brand back. Remediation targets that dimension, not the average.
- Treat Agent-native as describable but not yet broadly achievable. Most brands won’t reach Agent-native within the next several years. The stage is defined so that it functions as a true endpoint, not as an aspirational label brands can claim.
- Re-score after each remediation block. The composite changes when the binding constraint changes, and the binding constraint changes as work lands. Treating the maturity assessment as annual without intermediate re-scoring loses the dependency-ordered signal.
- Use the stages for executive communication. The five-stage scale is more communicable than dimension-by-dimension detail. Executives can absorb “we’re at Discoverable composite and the binding dimension is Trust Signal Density” in a way they can’t absorb a heat map of eight scores.
- Don’t conflate dimension types when comparing scores. A Comparable on Governance Maturity is observed procedurally; a Comparable on Latency and Data Freshness is observed through P95 measurements. The stages are equivalent in meaning, not in source data.
Future Trends
- Stage definitions tightening as agent capability advances. The Agent-native stage is currently defined against what’s possible at the leading edge in 2026. As agent capability advances, the bar for Agent-native is expected to rise, and the bar for what counts as Comparable will rise with it.
- Stage calibration by vertical. The universal core’s stage definitions hold across categories. Vertical overlays adjust stage emphasis and add category-specific sub-component weighting (the Consumer DTC overlay’s required-signal rule on Trust Signal Density, the Regulated overlay’s treatment of disclosure attributes). Stage definitions themselves remain stable; the calibration above the stages varies.
- Model-version-indexed scoring. As agent behavior shifts across model releases, scoring is expected to become more explicitly model-version-indexed. A brand at Comparable on Differentiation Encoding under model version X may score differently under model version Y, and the framework’s measurement loop already records model versions for this reason.
- Agent-native as competitive surface. As more brands reach Agent-native on individual dimensions, the composite Agent-native — all eight dimensions at Agent-native simultaneously — is expected to remain rare and to function as a durable competitive position rather than as a checkpoint most brands reach.
- Stage-based benchmarking. Category-level benchmarks for stage distribution (what percentage of brands in apparel sit at Comparable, what percentage at Differentiated) are expected to develop as more brands run the assessment. The stage scale’s uniformity makes cross-brand comparison feasible in a way dimension-specific scales would not.
FAQs
1. Who created the BVAC Maturity Stages? Greg Kihlström, martech futurist and Principal at The Agile Brand, developed the maturity stages as part of the Brand Visibility for Agentic Commerce (BVAC) Framework, introduced in 2026.
2. What are the five stages? Invisible, Discoverable, Comparable, Differentiated, and Agent-native. The stages describe a brand’s progression from structural absence in agent-readable data through full participation in agent-to-agent commerce.
3. Why a shared scale across all dimensions? A shared scale makes scores comparable across the framework’s eight dimensions, supports the prerequisite cap and floor mechanics (which depend on stages being on the same scale), and produces a coherent maturity profile rather than eight unrelated metrics.
4. What’s the difference between diagnostic and effective scores? Diagnostic is the stage a dimension reaches on its own assessment criteria. Effective is the diagnostic capped by the lower of the two prerequisite stages (Identity Legibility, Attribute Completeness) and capped at Discoverable if Trust Signal Density is below the floor. The gap between them is the framework’s readout of how much foundational work is required.
5. How is the composite framework score determined? It’s the lowest effective score across all eight dimensions. The framework’s principle is that a brand is only as agent-ready as its weakest dimension allows; averaging would smooth over the binding constraint and misdirect remediation.
6. Why score on the lower stage in borderline cases? Scoring up declares readiness the brand hasn’t earned, and under the caps that mistake propagates across the strategic tier. Scoring down to the lower stage and noting the partial progress in the gap map preserves the dependency-ordered remediation signal.
7. How does stage observability differ by dimension type? Brand-facing dimensions are observed through data-surface conditions (schema present, data populated). Technical-implementation dimensions are observed through technical measurements (P95 latency, version concurrency). Operating-model dimensions are observed through procedural conditions (playbooks exist, policies reviewed on cadence). The stages are equivalent in meaning across all three modes.
8. Does every brand need to reach Agent-native? No. Agent-native is a true endpoint that most brands won’t reach within the next several years. Most strategic value comes from moving the composite from Invisible or Discoverable to Comparable or Differentiated. Agent-native is the upper bound of the scale, not a near-term target.
9. How often should a brand re-score on the maturity stages? The full assessment runs at least annually. Focused re-assessments on specific dimensions run after major remediation blocks complete, because the composite changes when the binding constraint changes. The measurement loop (Layer 5 of the framework) provides directional signals that trigger focused re-assessments between annual cycles.
10. Can a brand be Agent-native on one dimension and Invisible on another? Yes, and it’s a common pattern early in agentic commerce maturity. A brand with a sophisticated deployed agent (Agent-native on Brand-Agent Representation) and no structured trust surface (below floor on Trust Signal Density) is exactly this profile. The composite framework score in this case is Invisible — the floor cap and the lowest-effective rule combine to produce the lowest stage as the summary.
Related Terms
- Brand Visibility for Agentic Commerce (BVAC)
- Agentic Commerce
- Identity Legibility
- Attribute Completeness
- Differentiation Encoding
- Brand-Agent Representation
- Trust Signal Density
- Trust Signal Density Floor Mechanism
- Protocol Readiness
- Latency and Data Freshness
- Governance Maturity
- Decision Invisibility
- Attribution Gap
- Experience Maturity
- Digital Experience Maturity
Sources
- Kihlström, G. “How Purchase Decisions Now Form Before the Customer Is Involved.” The Agile Brand, May 2026. https://www.gregkihlstrom.com/martech-futurist-blog/purchase-decisions-form-before-customer-involved