A new revenue line is forming. Mastercard equips merchants to win it.

As agents start shopping on behalf of consumers, every merchant faces the same question: are we ready to be chosen, trusted, and transacted with? The outcomes below define the prize. The foundations after them are how Mastercard puts merchants in position to claim it.

The outcomes at stake
What winning looks like on the P&L.
A new question is landing in boardrooms: how big is our agentic-commerce business? Most merchants can't yet answer with defensible numbers. These are the four outcomes that frame winning.
Top line
Agent-attributed GMV
Revenue from orders placed by an autonomous agent acting on behalf of a user.
intents × win rate × AOV
Efficiency
Agent CAC
Cost to acquire a customer through agent channels. Lower than paid search at scale.
infra spend ÷ agent buyers
Retention
Agent LTV
Lifetime value of customers who originated via agents. Often higher: lower friction, higher repeat.
AOV × frequency × lifespan
Moat
Share of agent voice
% of category prompts where you surface in the top-3 agent shortlist. Compounds over time.
your citations ÷ total
Produced by
Three foundations
The capabilities Mastercard brings to each.

Each P&L outcome above is produced by three foundational capabilities working together. Mastercard's network, data, and trust infrastructure shorten the build for merchants on every one.

1 Foundation 01
Brand recall in AI reasoning
A brand strong enough to be remembered by models, not only by consumers. Powers share of agent voice and feeds top-of-funnel demand.
Mastercard enablement
Merchant insights SpendingPulse data Category benchmarks Loyalty consortium SoAV index
2 Foundation 02
Technical legibility
Systems that make merchant data legible and actionable to agents. Lifts win rate and parse success.
Mastercard enablement
Click to Pay Token Service Merchant directory Product catalogue APIs
3 Foundation 03
Trust infrastructure
Verified intent, consented data, safe transactions. Drives checkout rate, AOV, and LTV.
Mastercard enablement
Agent Pay Delegated auth 3DS for agents Chargeback rails Identity verification Consent ledger
The funnel underneath
Seven gates from intent to fulfilment.

The three foundations play out across seven operational gates. Each gate has its own levers and its own success metrics, mapped to the P&L outcomes above.

Stage 01 Brand recall
Intent forms
A user asks an agent to shop on their behalf.
Category demand
Agent adoption
Vertical maturity
Metrics
Agent-initiated intentsper month
Category prompt volumeshare
Stage 02 Brand recall
Model awareness
The LLM knows you exist before any retrieval.
Training presence
Wikipedia
Reddit sentiment
Press mentions
Metrics
Brand recall rate% no-retrieval
Share of model voicevs. rivals
Citation countper LLM
Stage 03 Brand recall
Retrieval
The agent fetches and can read your site.
llms.txt
llms-full.txt
AI bot allowlist
SSR pages
Metrics
Agent fetch success% completed
AI bot visitsper day
llms.txt richness0–5 score
Stage 04 Legibility
Parse
The agent understands price, stock, and variants.
schema.org/Product
JSON-LD
Real-time stock
All-in price
Metrics
Parse success rate% SKUs legible
Schema coverage% catalogue
Price accuracy% all-in visible
Stage 05 Legibility
Comparison
The agent chooses you over rivals.
Verified reviews
Aggregate rating
Return policy
Warranty
Metrics
Top-1 selection rate% as first pick
Shortlist inclusion% top-3
Recommendation strengthsentiment score
Stage 06 Trust
Cart & checkout
The agent transacts on the user's behalf.
ACP endpoint
MCP tools
Delegated auth
No CAPTCHAs
Metrics
Agent checkout rate% completed
Agent AOV$ per order
Basket attach rate% multi-SKU
Payment auth rate% first-try
Stage 07 Trust
Fulfilment
Order completes and post-purchase updates flow back.
Delivery promise
Tracking API
Returns flow
Receipts
Metrics
Fulfilment NPSagent-reported
Return rate% agent orders
Repeat agent purchase% within 90d