Decision Invisibility (BVAC Framework)

Definition

Decision invisibility is the exclusion of a brand from a purchase it would otherwise have competed for, decided inside an AI agent, before the customer evaluates anything, and without any corresponding signal in the brand’s own analytics.

The term was introduced by Stibo Systems in 2026 to describe the phenomenon by which a brand is excluded from a customer’s consideration set because its product data is unstructured, incomplete, or ambiguous to an AI agent (Molino Sánchez, 2026). The agent interprets the request, resolves the candidate set, filters on the attributes it can read, and returns a short list of purchase-ready options. A brand the agent cannot resolve or cannot compare is not ranked low; it is absent from the set the customer ever sees. Nothing about that absence presents as a failure the brand can investigate. Traffic doesn’t fall, no campaign underperforms, and the lost sale was never attributable in the first place.

Decision invisibility is the failure mode that the Brand Visibility for Agentic Commerce (BVAC) Framework, developed by Greg Kihlström of The Agile Brand, was built to diagnose and address (Kihlström, 2026). The framework treats decision invisibility as the practical consequence of failures in the data surface an agent reads — particularly its two prerequisite dimensions, Identity Legibility and Attribute Completeness — and provides the assessment methodology and remediation library that move a brand out of below-threshold status.

Why Decision Invisibility Matters

The scale of the shift makes decision invisibility a measurable risk rather than a hypothetical one. Three data points set the context:

• Adobe Analytics recorded generative-AI-referred traffic to US retail sites growing roughly 690% year over year across the 2025 holiday season (Adobe Analytics, 2026).
• McKinsey estimates conversion from AI-generated product recommendations at about 4.4 times that of traditional search (McKinsey, as cited in MetaRouter, 2026).
• Capgemini’s consumer research finds that 58% of consumers have already replaced traditional search engines with generative-AI tools as their primary route to product recommendations (Capgemini Research Institute, 2025). agilebrandguide

The share of discovery that now passes through an agent is large enough that exclusion at that layer is no longer a marginal loss, and it’s growing faster than most brands are revising their assumptions about where the decision forms.

The implication for marketing organizations is direct. The stages of the funnel still exist; they have been compressed into a single interaction the customer doesn’t observe. The agent interprets intent, filters the field, and returns a small set of purchase-ready options. Whatever is in that set constitutes the consideration set. Whatever is outside it is, for that purchase, absent. The work that earns attention from a human shopper — narrative, design, persuasive merchandising — is largely invisible to the system that now determines whether the brand reaches that shopper at all.

How the Exclusion Happens

Before an agent can weigh anything a brand would consider strategic, it has to resolve two questions: which brand this is, and which product it is evaluating. When identifiers drift across channels, when the entity declaration is thin, or when a return policy sits unmarked in unstructured HTML, the agent doesn’t pause to investigate. It defaults to whatever resolves cleanly — typically a marketplace listing — or it omits the brand.

The behavior driving this is risk aversion. An agent that recommends a product it cannot stand behind incurs the costlier error, so when an expected attribute is missing or ambiguous it skips the product rather than guess. The exclusion isn’t a judgment that the brand is worse than the alternative; it’s a judgment that the brand cannot be assessed, and from outside the agent the two are indistinguishable.

Controlled testing reinforces where the leverage actually sits. Sabbah and Acar evaluated eight standard promotional mechanisms across four models and more than 16,000 simulated consumer purchase decisions, and the one signal that consistently moved selection upward across every model and category was structured ratings, while the reasoning models were the more skeptical of overt persuasion (Sabbah & Acar, 2026). The finding is consumer-scoped and doesn’t extend to every category, but its operational implication for decision invisibility is general: the remedy for being skipped is structural, not creative. agilebrandguide

The mechanism reduces to a sequence:

1. The agent receives a category query carrying the user’s mandate.
2. It resolves the candidate set from the data sources it can read.
3. It filters the set on the attributes its risk model requires.
4. It ranks the survivors on the dimensions it can weight.
5. It returns the small set of purchase-ready options.

Decision invisibility happens at step 2 or step 3. A brand the agent can’t resolve is dropped at step 2. A brand the agent can resolve but lacks expected attributes for is dropped at step 3. Neither drop produces a signal the brand can observe.

What Makes It Different from Other Visibility Problems

Every other category of marketing problem announces itself through a metric that moves. Bounce rate rises. Campaign engagement falls. Conversion declines. The triage process that allocates marketing attention to the dashboards that change works because problems present as signal changes.

Decision invisibility is invisible by construction. Three properties distinguish it from familiar visibility problems:

• The brand was never in the set, so there’s no “before” to compare against. A brand that ranks fifth and drops to tenth shows declining performance. A brand that was filtered out at step 2 of the agent’s process shows nothing — no degradation, because no presence.
• The lost sale wasn’t attributable in the first place. Discovery and comparison occur inside the agent. Even if the customer eventually arrives, they arrive through a direct or organic path that carries no signal about the agent’s recommendation. The brand sees direct traffic that looks normal.
• The triage process actively works against detection. Marketing organizations allocate attention to the dashboards that move. Decision invisibility doesn’t move the dashboards. The triage logic that focuses on degrading metrics filters this problem out of attention.

The hazard this creates is specific. The absence of a signal isn’t evidence that the problem is small. Under agentic discovery, the absence of a signal is the expected presentation of the problem.

The Conditions That Produce It

The BVAC Framework places two dimensions ahead of every strategic consideration — Identity Legibility and Attribute Completeness — because they are the conditions under which an agent can evaluate anything at all. Their failure modes are precisely the conditions that produce decision invisibility, and none of them is a branding failure in the conventional sense.

Identity Legibility failures that produce decision invisibility:

• GTIN drift. The same product reads as several because identifiers differ across channels.
• URL fragmentation. Uncanonical variant pages split the entity signal.
• Name collision without disambiguation. The agent resolves to the wrong entity.
• Marketplace authority. The brand’s marketplace entity is more complete than its own, so the agent cites the reseller.

Attribute Completeness failures that produce decision invisibility:

• Absent or vague MerchantReturnPolicy markup. Risk-averse agents deprioritize.
• Missing category-standard attributes. Measured against what competitors publish rather than what the brand prefers to publish.
• Stale inventory or pricing. A recommendation withdrawn after it surfaces trains the agent to distrust the source.
• The marketplace completeness gap. The marketplace listing has more complete attributes than the brand’s own catalog, so the marketplace becomes the authoritative source.

Each of these is a data condition, and each produces the same outcome: silent removal from the set. None of them is a creative problem, and none of them resolves through better positioning or richer brand storytelling. The framework’s prerequisite cap formalizes the consequence — the effective score on every strategic dimension is held at the lower of the two prerequisite scores, so a brand can do strong differentiation, trust, and protocol work and remain invisible because the agent never reached any of it. Decision invisibility is what that cap looks like from the outside: excellent strategic work returning nothing, with no indication of why.

The Attribution Gap

The attribution gap is the structural reason decision invisibility persists rather than self-correcting. Discovery and comparison happen inside the agent; the customer, if they arrive at all, arrives later through a direct or organic path, and the influence that determined the decision leaves no trace in the merchant’s analytics (Hanna, 2026). Two dynamics make the gap particularly hard to close:

• Query fan-out. The agent draws from several sources without the user clicking any of them. The brand may have been considered, compared, and excluded, with no surface visit recorded.
• Zero-click shortlisting. The agent returns a short list of options without a search results page the user reviews. The brand sees neither the queries it lost nor the comparisons it was excluded from.

The financial logic compounds this. Because the loss is unattributable, it doesn’t enter the brand’s return-on-investment calculations, and a cost that cannot be priced is reliably under-prioritized against costs that can. The inputs that set the urgency — agent adoption, and competitor and marketplace data quality — are outside the brand’s control. The one variable the brand does control is its own legibility, and it’s the variable being postponed.

The BVAC Framework’s measurement loop addresses the attribution gap by operating on directional signals rather than last-click attribution. Substitution rate — how often agents recommend a competitor in place of the brand for queries where the brand is a legitimate candidate — falls after the Trust Signal Density floor is crossed or after Attribute Completeness gaps close. This is the reversal signal for decision invisibility. The signal won’t establish a causal link to a sale; it shows the eligibility surface widened.

How to Detect Decision Invisibility

Because the failure doesn’t surface on its own, detection has to be deliberate, and it has to come before remediation. The BVAC Framework’s assessment methodology provides the structured detection instrument.

The detection sequence:

1. Identify the top SKUs by revenue. Default scope is the top 20.
2. Assess them against the structured data an agent reads, not the marketing site. Whether the brand and product resolve to stable entities. Whether identifiers are consistent across the site, feeds, and marketplaces. Whether MerchantReturnPolicy and the category-standard attribute set are present in the markup.
3. Run agent query simulations. Run category buying prompts across three to five agent interfaces (ChatGPT, Gemini, Perplexity, plus category-specific assistants). Record which brands appear in the consideration set, which differentiators surface, and which sources the agent cites.
4. Identify which source the agent cites. Where an agent cites a marketplace over the brand’s own catalog, that’s the failure in operation rather than a hypothesis about it.
5. Compare against competitors. A brand absent from the consideration set while competitors with similar or weaker product positioning appear is the diagnostic signature.

The detection isn’t a creative exercise. It’s a structured-data audit and a controlled simulation run. The framework’s Layer 3 assessment methodology provides the procedural specification; the Layer 5 measurement loop provides the ongoing directional signals (AI share of voice, citation rate, substitution rate) that detect drift between full annual assessments.

The most important diagnostic property: detection has to happen before the brand’s revenue declines, because by the time decision invisibility shows up in revenue, it has been compounding silently for the months or quarters during which agent share of discovery was growing.

How to Address It

The BVAC Framework’s remediation sequence is fixed by dependency rather than by perceived urgency. Decision invisibility addresses through the framework’s prerequisite-first, floor-second, strategic-third order:

1. Prerequisites first. Identity Legibility and Attribute Completeness gaps are the conditions that produce decision invisibility most directly. Resolving them moves the brand from being unfilterable to being eligible for inclusion. Most prerequisite remediation falls in the 90-day horizon — schema and markup work that fits within existing systems.
2. Trust Signal Density floor second. If the brand sits below the trust floor (no structured Review entities, no certification surface, no sameAs authority anchors), crossing the floor is the next move. A single 90-day schema action against existing reviews, certifications, or authority recognition typically crosses it.
3. Strategic dimensions in binding-constraint order. Once the prerequisites and floor are resolved, the strategic dimensions return their diagnostic value. The framework’s action paths library specifies the remediation horizon for each sub-component.

A worked case illustrates the pattern. A national brand with strong unaided awareness sells its flagship product through its own site and through two large marketplaces. On its own catalog the product carries marketing copy, lifestyle imagery, and a returns page written for human readers in unstructured HTML. On the marketplaces the same product carries a complete normalized attribute set, structured ratings, and an explicit return window, because the marketplace enforces those fields. An agent asked for a recommendation in the category resolves the product most cleanly through a marketplace, compares it there, and returns it as a marketplace listing — or omits the brand’s own surface from the citation entirely. The brand’s analytics show stable direct traffic and healthy marketplace sales. Nothing in them indicates that the agent never considered the brand’s own catalog, that margin shifted to the marketplace, or that the brand’s entity authority is accruing to a reseller. Identity Legibility and Attribute Completeness both sit below Comparable on the brand’s own surface; the prerequisite cap holds the entire strategic tier down regardless of how strong the positioning is; and the composite is Invisible while every metric the brand watches looks normal. The lesson isn’t that the brand is performing poorly. It’s that the instrument the brand uses to know whether it’s performing poorly doesn’t observe this failure.

Comparison to Similar Concepts

Concept	Focus	Relationship to Decision Invisibility
Zero-Click Search	Search results that don’t produce a click-through to a site	Zero-click search is one mechanism of decision invisibility; the brand may be cited or omitted with no click signal either way
Filter Bubble	Algorithmic narrowing of content exposure	Filter bubbles narrow human-facing content; decision invisibility narrows agent-facing inclusion before the human sees content
Algorithmic Aversion	Human reluctance to follow algorithmic recommendations	Adjacent and inverse — algorithmic aversion concerns the human side of the decision; decision invisibility concerns the algorithmic side
Narrative Flattening	Agent reduction of rich brand content to default attributes	Narrative flattening is a Differentiation Encoding failure mode; decision invisibility is the broader exclusion problem that prerequisite failures most often produce
Authority Erosion	Marketplace becoming the cited trust authority for a brand	Authority erosion is one specific mechanism by which decision invisibility happens — the agent cites the marketplace and omits the brand’s own surface
Attribution Gap	Inability to trace agent influence to a sale	The attribution gap is the structural reason decision invisibility persists rather than self-correcting; they pair

Decision invisibility is the failure phenomenon; the BVAC Framework, the floor mechanism, and the prerequisite caps are the diagnostic and remediation instruments. The phenomenon was named by Stibo Systems; the framework that operationalizes diagnosis and remediation was developed by Greg Kihlström.

Best Practices

• Detect before you remediate. Decision invisibility doesn’t show up in dashboards. Detection requires deliberate assessment of the structured data surface and controlled simulation of agent queries. Skip this step and remediation targets the wrong problem.
• Audit the data surface, not the marketing surface. The brand’s own perception of its visibility comes from reviewing the human-facing site. The agent’s view comes from the structured data. The two diverge regularly and the second is what matters.
• Treat marketplace authority as the structural problem it is. When an agent cites a marketplace over the brand’s own catalog, that’s the failure in operation. The fix is encoding the brand’s own surface, not improving marketplace presence.
• Sequence prerequisites first, floor second, strategic third. The framework’s dependency order is non-negotiable. Strategic work above an unresolved prerequisite or below the floor returns nothing in effective score.
• Use directional measurement, not last-click attribution. Substitution rate, citation rate, and AI share of voice from controlled agent query simulations are the reversal signals. Last-click attribution can’t measure what the attribution gap hides.
• Don’t wait for revenue to decline. By the time decision invisibility shows up in revenue, it has been compounding silently for the months or quarters during which agent share of discovery was growing. The detection-then-remediation sequence has to run ahead of the revenue signal.
• Treat the absence of a signal as expected presentation. Under agentic discovery, the lack of a moving metric isn’t reassurance. It’s the diagnostic signature. The triage process that focuses on degrading metrics has to be supplemented with a structured assessment that detects what the dashboards can’t.
• Address the controllable variable. Agent adoption and marketplace data quality are outside the brand’s control. The brand’s own legibility is the one variable it does control, and it’s the one most reliably postponed.

Future Trends

• Decision invisibility growing in absolute terms. The growth in generative-AI-referred traffic to retail sites (roughly 690% year over year across the 2025 holiday season per Adobe Analytics) is unlikely to reverse. As agent share of discovery rises, the exclusion at the agent layer grows proportionally.
• Marketplace structural advantage compounding. Marketplaces hold a structural advantage in agent-mediated discovery because they enforce data standards and carry assortment and review volume that an individual catalog does not (Mirakl, 2026). The surface an agent reaches for when a brand’s own data is ambiguous is the same marketplace that erodes the brand’s margin and entity authority. Every month a brand’s own legibility is deferred, the default it is being measured against gets stronger.
• Detection instruments standardizing. Agent query simulation, controlled re-runs against fixed query sets, and substitution-rate measurement are expected to develop into standard instruments the way SEO measurement did over the past two decades. Brands without these instruments will be detecting decision invisibility increasingly late.
• The attribution gap partially closing. Brand agent observability (logs, transcripts, outcome data from the brand’s own agent in A2A interactions) provides first-party transaction data where analytics traffic provides none. The gap closes partially as brand agents deploy.
• Vertical-specific decision invisibility patterns emerging. B2B procurement-query exclusion, regulated-category compliance-attribute exclusion, and consumer DTC review-signal exclusion are expected to develop as distinct sub-patterns of decision invisibility with vertical-specific diagnostic and remediation specializations.

FAQs

1. What is decision invisibility? The exclusion of a brand from a purchase it would otherwise have competed for, decided inside an AI agent, before the customer evaluates anything, and without any corresponding signal in the brand’s own analytics. The brand is filtered out at the consideration-set assembly stage and the customer never sees it as an option.

2. Who first described the concept? Stibo Systems introduced the term in 2026 to describe brands being filtered out of agent recommendations because their product data is unstructured, incomplete, or ambiguous (Molino Sánchez, 2026). The BVAC Framework, developed by Greg Kihlström of The Agile Brand, is the diagnostic instrument that operationalizes detection and remediation.

3. How is this different from low search rankings? Low rankings produce signals — declining traffic, falling conversion, visible competitive movement. Decision invisibility produces no signal. The brand was never in the set the agent assembled, so there’s no “before” against which to compare. The dashboards stay flat while market share erodes.

4. Why can’t I see this in my analytics? The attribution gap. Discovery and comparison happen inside the agent. If the customer arrives at all, they arrive later through a direct or organic path, and the influence that determined the decision leaves no trace in the merchant’s analytics. Query fan-out and zero-click shortlisting mean the brand sees neither the queries it lost nor the comparisons it was excluded from.

5. What causes decision invisibility? Failures in the data surface an agent reads — particularly in Identity Legibility (GTIN drift, URL fragmentation, name collision, marketplace authority) and Attribute Completeness (absent MerchantReturnPolicy, missing category-standard attributes, stale data, marketplace completeness gap). The BVAC Framework’s two prerequisite dimensions catalog the conditions that produce it.

6. How fast is this growing? Adobe Analytics recorded generative-AI-referred traffic to US retail sites growing roughly 690% year over year across the 2025 holiday season. Capgemini’s consumer research finds 58% of consumers have replaced traditional search engines with generative-AI tools for product recommendations. The exposure to decision invisibility is growing with agent adoption.

7. How do I detect decision invisibility? Through a structured-data audit of the brand’s top SKUs against what an agent reads, plus controlled agent query simulations across three to five agent interfaces. The BVAC Framework’s Layer 3 assessment methodology specifies the procedure. Detection is deliberate, not passive.

8. How do I address it? Through the BVAC Framework’s dependency-ordered remediation sequence: prerequisites first (Identity Legibility, Attribute Completeness), Trust Signal Density floor second, strategic dimensions third. Most prerequisite remediation is 90-day schema and markup work. The cost is low; the deferred cost of leaving it undone compounds.

9. Why is this a structural problem rather than a creative one? Agents filter on the attributes they can read. A brand with rich brand storytelling and weak structured data fails the filter the same way as a brand with weak storytelling and weak structured data. The remedy isn’t better creative; it’s structured, machine-readable data on the surfaces agents query.

10. What if my analytics still look healthy? That’s the diagnostic signature of decision invisibility, not reassurance against it. Under agentic discovery, healthy traffic and stable conversion are consistent with a brand losing increasing share of agent-mediated consideration silently. Detection has to be deliberate because the dashboards don’t surface the problem.

Related Terms

1. Brand Visibility for Agentic Commerce (BVAC)
2. Agentic Commerce
3. Trust Signal Density Floor Mechanism
4. Zero-Click Searches
5. Filter Bubble
6. Algorithmic Aversion
7. Zero Moment of Truth (ZMOT)
8. Generative Engine Optimization (GEO)
9. Answer Engine Optimization (AEO)
10. Business to Agent to Consumer (B2A2C)
11. Brand Awareness
12. Share of Market (SOM)

Sources

• Adobe Analytics. “AI-Driven Traffic Surges Across Industries, Retail Sees Biggest Gains” (2025 holiday shopping recap). Adobe, 2026. https://business.adobe.com/products/adobe-analytics/customer-journey-analytics/agentic-and-ai-driven-insights.html
• Capgemini Research Institute. “What Matters to Today’s Consumer” (4th ed.). Capgemini, 2025. https://www.capgemini.com/insights/research-library/what-matters-to-todays-consumer-2025/
• Core dna team. “Agentic Commerce Is Here: How to Make Your Products Discoverable by AI.” Core dna, 2026.
• Hanna, C. “Attribution Gap in Agentic Search: How to Close It.” Semrush, 2026. https://www.semrush.com/blog/attribution-gap-in-agentic-search/
• 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
• MetaRouter. “Agentic Commerce Trends and Statistics for 2026” (conversion figure attributed to McKinsey). MetaRouter, 2026. https://www.metarouter.io/post/agentic-commerce-trends-statistics
• Mirakl. “Marketplaces and the Future of Agentic Commerce.” Mirakl, 2026.
• Molino Sánchez, M. “7 Signs Your Brand Is Losing Ground in Agentic Commerce.” Stibo Systems, 2026. https://www.stibosystems.com/blog/7-signs-your-brand-is-losing-ground-in-agentic-commerce
• Sabbah, J. and Acar, O. A. “Research: Traditional Marketing Doesn’t Work on AI Shopping Agents.” Harvard Business Review, May 12, 2026. https://hbr.org/2026/05/research-traditional-marketing-doesnt-work-on-ai-shopping-agents