The definitive collection of data points, benchmarks, and forecasts shaping AI-driven commerce, sourced, verified, and organized for strategists.
172 data points Last updated November 19, 2026 · Cite this page
Key Takeaways
AI-referred retail traffic grew 393% year over year in Q1 2026.
51% of US consumers used AI to shop in the past month.
Agent-completed orders nearly tripled from Q2 to Q3 2026.
Meta Muse drives 29.7% of agent-completed purchases despite only a 6.9% session share.
Only 1 in 5 merchants have 80%+ machine-readable product data.
Total cost of fraud reached $5.13 per $1 of direct loss.
The agentic commerce market is projected to reach $3.5 trillion by 2031.
No statistics match your search. Try a different keyword or category.
Market Size
How large is agentic commerce actually getting, and how fast? This section tracks the dollar forecasts analysts are putting on AI-mediated shopping, from near-term 2026 estimates to long-range projections out to 2031. The estimates vary widely between firms because few have settled on a shared definition of what counts as an agent-assisted transaction versus a fully autonomous one. Read these as directional signals of scale, not a single consensus number.
$3T – $5T
Projected global agentic commerce market by 2030 for goods alone, roughly a quarter of all retail spending shifting to AI-mediated transactions.
Market Size McKinsey
$900B – $1T
Estimated US B2C agentic commerce opportunity by 2030, representing the domestic share of agent-driven retail.
Market Size McKinsey
$300B – $500B
Alternative US forecast pegging agentic commerce at 15–25% of total domestic e-commerce value by the end of the decade.
Market Size Bain & Company
$190B – $385B
Conservative US estimate suggesting AI agents will handle 10–20% of all online retail transactions by 2030.
Market Size Morgan Stanley
$20.57B
Projected US retail spending through AI agents in 2026 alone, still just 1.5% of total e-commerce, but doubling year over year.
Market Size eMarketer
$15T+
Total value of B2B transactions expected to flow through AI agent exchanges by 2028 as enterprise procurement automates.
Market Size Gartner
$115B
Estimated new revenue AI shopping agents will add to the US e-commerce sector by 2030 as nearly half of online shoppers adopt them.
Market Size IDC
50%
Share of online shoppers expected to use AI agents by 2030, accounting for approximately 25% of their total spending.
Market Size IDC
20%
Share of all monetary transactions expected to be programmable, executed by agents without human button-clicks, by 2030.
Market Size Gartner
90%
Proportion of B2B buying interactions predicted to be conducted through AI agents by 2028, fundamentally reshaping procurement.
Market Size Gartner
$8B
Size of the agentic commerce market in 2026, projected to reach $3.5 trillion by 2031.
Market Size Juniper Research
300M to 1.3B
Growth in agentic commerce users projected between now and 2031.
Market Size Juniper Research
55% / 18%
Share of commerce expected to be agent-assisted (55%) versus fully autonomous (18%), together impacting $1.3 trillion in spend.
Market Size Stripe + BCG
$400B
Projected value of fully autonomous, agent-driven commerce, where the agent completes the purchase without a human step.
Market Size Stripe + BCG
$46.74B to $175.11B
Growth of the retail AI market between 2025 and 2030.
Market Size Wipro
$7B to $93B
Growth of the agentic payment market between 2026 and 2032.
Market Size Galileo Financial Technologies
25%
Share of US online sales expected to be concentrated in recurring, low-risk categories ideal for agent automation by 2030.
Market Size J.P. Morgan
AI Traffic
Before an agent ever completes a purchase, it has to send a human there, or go looking on their behalf. This section tracks how much retail traffic now arrives by way of AI assistants and chatbots, and how that traffic behaves once it lands: how long it stays, how many pages it views, and how often it converts. The growth curves here are some of the steepest in the entire agentic commerce dataset, which is why AI-referred traffic is becoming a line item retailers track separately from organic and paid search.
1,324%
Growth in AI-referred traffic to US retail sites since October 2024, signaling a permanent shift in how shoppers discover products.
AI Traffic Adobe Digital Economy Index
693%
Year-over-year surge in AI-driven traffic to US retail sites during the 2025 holiday season, the steepest climb recorded.
AI Traffic Adobe
138%
Year-over-year growth in AI-referred retail traffic as of May 2026, indicating the trend is sustained well beyond seasonal spikes.
AI Traffic Adobe
42%
How much better AI-referred traffic converts compared to non-AI traffic as of March 2026, a reversal from 38% worse just one year earlier.
AI Traffic Adobe
700%
Conversion advantage of AI shopping assistant traffic over social media referrals, making agents the highest-converting discovery channel.
AI Traffic Salesforce
9×
How many more times shoppers arriving from AI search convert compared to those referred from social media platforms.
AI Traffic Salesforce
53%
Additional time AI-referred shoppers spend on retail sites versus direct visitors, browsing 23% more pages per session.
AI Traffic Adobe (May 2026)
37%
More revenue generated per visit by AI-referred shoppers who spend longer on site and browse more deeply before purchasing.
AI Traffic Adobe
13×
Year-over-year increase in orders originating from AI-powered searches across Shopify's merchant base in Q1 2026.
AI Traffic Shopify
8×
Year-over-year growth in AI-driven traffic to Shopify storefronts, with catalog-powered searches converting at double the rate of general AI queries.
AI Traffic Shopify
14%
Higher average order value carried by AI-powered search orders on Shopify compared to orders arriving through organic search.
AI Traffic Shopify
2×
Conversion rate advantage of catalog-powered AI searches over general AI search queries on Shopify, structured data wins.
AI Traffic Shopify
393%
Year-over-year growth in Q1 2026 AI-referred retail traffic.
AI Traffic Adobe, Q1 2026
89%
Year-over-year rise in AI traffic during Prime Day 2026, converting 40% better than non-AI traffic.
AI Traffic Adobe, Jul 2026
$14.2B
Global online sales influenced by AI traffic during Black Friday 2025.
AI Traffic Adobe, Nov 2025
300%
Growth in third-party AI agent traffic during the first half of Black Friday 2025.
AI Traffic Digital Commerce 360, Nov 2025
<0.2%
Share of total e-commerce traffic AI sessions still represent, despite the rapid growth rates recorded elsewhere in this category.
AI Traffic Digiday, 2026
~20%
ChatGPT referral traffic now equals roughly a fifth of Walmart's total site visits.
AI Traffic Digital Commerce 360, 2026
1.6×
Conversion rate of AI-referred visits compared to other retail traffic.
AI Traffic First Page Sage, Oct 2026
4,700%
Year-over-year increase in generative AI traffic to retail sites.
AI Traffic Adobe via BCG
Adoption & Trust
Technology adoption and consumer trust move at different speeds, and nowhere is that gap more visible than in agentic shopping. These stats separate how many people have tried an AI shopping tool from how many are actually willing to hand over budget and payment authority to one. The split matters for anyone building an agent-facing storefront: broad usage doesn't automatically translate into broad delegation.
39%
Share of US consumers who have already used AI tools for shopping as of early 2026, crossing from novelty to mainstream behavior.
Adoption & Trust Adobe
84%
European consumers who report using AI tools in their everyday lives, setting the stage for commerce-specific adoption.
Adoption & Trust McKinsey
51%
US shoppers who say they would let an AI agent handle their entire purchase process, including payment, if given budget and brand controls.
Adoption & Trust Adyen
44%
UK consumers willing to delegate the full shopping journey to an AI agent, from discovery through checkout, within set parameters.
Adoption & Trust Adyen
47%
Consumers who would let an AI agent handle repetitive purchases autonomously, rising to two-thirds among 25–44-year-olds.
Adoption & Trust Checkout.com
57%
Shoppers who would allow an AI agent to switch brands on their behalf if the agent identifies measurably better value.
Adoption & Trust Checkout.com
35%
US consumers now using AI shopping assistants, more than doubling from 12% in the prior year.
Adoption & Trust Adyen
28%
UK AI assistant user share, up from 12% in just one year, matching the growth trajectory seen in the US market.
Adoption & Trust Adyen
85%
AI shoppers who say the experience improved their shopping, while 66% believe the tools provide accurate product results.
Adoption & Trust Adobe
76%
Consumers who express concern about how AI chatbots collect and use their personal data during shopping interactions.
Adoption & Trust Morgan Stanley AlphaWise
60%
Shoppers who do not trust AI chatbots with their payment information, the single biggest barrier to autonomous checkout.
Adoption & Trust Morgan Stanley AlphaWise
42%
Consumers whose primary concern with AI shopping agents is losing control over purchasing decisions.
Adoption & Trust Checkout.com
24%
Shoppers who say they will never delegate any purchasing decision to an AI agent, while 27% trust no organization to run one.
Adoption & Trust Checkout.com
£177
Average amount UK consumers would allow an AI agent to spend per transaction before requiring re-authorization.
Adoption & Trust Checkout.com
3×
How much more consumers trust retail-owned AI agents versus third-party agents, brand equity transfers to the agent layer.
Adoption & Trust Bain & Company
44%
AI-search users who now call it their primary search method, overtaking the 31% who still prefer traditional search engines.
Adoption & Trust McKinsey
69%
Consumers who abandon immediately when they receive irrelevant recommendations, explaining why agents are highly selective about what they surface.
Adoption & Trust Industry benchmark
51%
US consumers who used AI to shop in the past month.
Adoption & Trust NIQ via Gladly, Sep 2026
61.4%
US online shoppers who use AI to research purchases.
Adoption & Trust First Page Sage, Oct 2026
6.8%
US online shoppers who have let an agent complete a checkout.
Adoption & Trust First Page Sage, Oct 2026
0.27% to 0.68%
Rise in agent-completed orders as a share of US e-commerce between Q2 and Q3 2026.
Adoption & Trust First Page Sage, Oct 2026
73% / 60%
Consumers familiar with AI shopping tools (73%), and share who expect to use agents within the next 12 months (60%).
Adoption & Trust Kearney
32% / 9%
Consumers who would let an agent decide what to buy (32%), versus just 9% open to fully autonomous shopping.
Adoption & Trust Accenture
56%
Consumers comfortable with AI suggesting options, but who still want the final say on the purchase.
Adoption & Trust McKinsey
23%
US consumers who trust generative AI to handle payments.
Adoption & Trust Visa Trust Index, Sep 2026
50% / 43%
Shoppers aged 18–34 who trust AI recommendations (50%), and share who bought an AI-recommended item in the past 3 months (43%).
Adoption & Trust PartnerCentric, Aug 2026
55%
Shoppers who have knowingly used a retailer's own AI assistant.
Adoption & Trust Wing
64%
Shoppers who would not let an AI agent spend money without their explicit approval.
Adoption & Trust Spreedly
Platform Comparison
Not all AI shopping agents convert at the same rate, and the gap between session share and purchase share is often the real story. This section compares how major AI platforms, including ChatGPT, Meta Muse, Amazon Alexa, Google Gemini, Claude, and Microsoft Copilot, actually perform once a session starts, measured by completed purchases and average order value rather than raw traffic. A platform with a small share of sessions but a disproportionate share of completed purchases is the one merchants should be watching most closely.
41.3% / 9.4%
ChatGPT holds 41.3% of AI shopping sessions but only 9.4% of agent-completed purchases.
Platform Comparison First Page Sage, Oct 2026
6.9% / 29.7%
Meta Muse holds 6.9% of AI shopping sessions but 29.7% of agent-completed purchases, with a $118 average order value.
Platform Comparison First Page Sage, Oct 2026
14.2% / 31.8%
Amazon Alexa for Shopping handles 14.2% of sessions and 31.8% of purchases, with a $52 average order value.
Platform Comparison First Page Sage, Oct 2026
27.6% / 23.1%
Google Gemini holds 27.6% of sessions and 23.1% of purchases, with a $74 average order value.
Platform Comparison First Page Sage, Oct 2026
5.8% / 2.6%
Claude holds 5.8% of sessions and 2.6% of purchases, but the highest average order value of any platform measured, at $141.
Platform Comparison First Page Sage, Oct 2026
4.2% / 3.4%
Microsoft Copilot holds 4.2% of sessions and 3.4% of purchases, with a $93 average order value.
Platform Comparison First Page Sage, Oct 2026
71%
Share of Q3 2026 growth in agent-completed orders traced to Meta Muse alone.
Platform Comparison First Page Sage, Oct 2026
3.4×
Completion rate of the Meta Muse + Shopify partnership integration compared to non-partner storefronts.
Platform Comparison First Page Sage, Oct 2026
Category Breakdown
Shoppers don't trust AI agents equally across every kind of purchase. This section breaks down how research and checkout behavior differ by product category, from low-stakes groceries to high-consideration travel and electronics, along with the spending limits shoppers set before wanting to step back in. The pattern is consistent: the higher the price and the more personal the decision, the more shoppers use AI to research, but the more they still want to complete the purchase themselves.
44.8% / 11.9%
Groceries: 44.8% use AI to research, 11.9% let an agent complete checkout, with a $112 median spending limit.
Category Breakdown First Page Sage, Oct 2026
36.2% / 10.7%
Event tickets: 36.2% research with AI, 10.7% let an agent check out, with an $85 spending limit.
Category Breakdown First Page Sage, Oct 2026
47.6% / 8.8%
Beauty and personal care: 47.6% research with AI, 8.8% let an agent check out, with a $68 limit.
Category Breakdown First Page Sage, Oct 2026
52.3% / 7.4%
Apparel: 52.3% research with AI, 7.4% let an agent check out, with a $96 limit.
Category Breakdown First Page Sage, Oct 2026
58.4% / 6.3%
Travel: 58.4% research with AI, 6.3% let an agent check out, with a $620 spending limit.
Category Breakdown First Page Sage, Oct 2026
68.1% / 5.2%
Consumer electronics: 68.1% research with AI, 5.2% let an agent check out, with a $240 limit.
Category Breakdown First Page Sage, Oct 2026
Brand Trust
Delegating a purchase to an AI agent means trusting whoever built it with your money, your data, or both. This section ranks which companies consumers currently trust most to operate an AI shopping agent, and surfaces the concerns, around data visibility and loss of control, that are holding broader adoption back. Brand trust here looks less like a single ranking and more like a set of tiers, with payment and retail incumbents currently leading.
42%
PayPal, Amazon, and Visa are the most trusted brands for AI agent shopping.
Brand Trust Visa Earning Trust Report, Sep 2026
28%
Apple, Google, and Mastercard form the second trust tier for AI agent shopping.
Brand Trust Visa Earning Trust Report, Sep 2026
15%
Microsoft, Meta, and American Express sit at the third trust tier for AI agent shopping.
Brand Trust Visa Earning Trust Report, Sep 2026
1 in 5
US consumers who are not comfortable with an AI agent from any listed company.
Brand Trust Visa Earning Trust Report, Sep 2026
85%
Consumers who want visibility into how AI agents collect their data.
Brand Trust Visa Earning Trust Report, Sep 2026
50%
Consumers who would stop using agents entirely if they lost control over purchases.
Brand Trust Visa Earning Trust Report, Sep 2026
Platforms
Agentic commerce runs on a growing stack of protocols, SDKs, and in-platform assistants connecting AI agents to merchant catalogs and payment rails. This section tracks the infrastructure layer, things like the Model Context Protocol, Google's Agent Payments Protocol, and native shopping assistants such as Amazon Rufus, that make agent-to-merchant transactions technically possible at scale. These numbers tend to be leading indicators: infrastructure adoption here tends to show up as transaction volume a few quarters later.
~1M
Shopify merchants anticipated to sell products directly through ChatGPT following the OpenAI–Shopify integration.
Platforms OpenAI
97M
Monthly SDK downloads for Model Context Protocol, the open standard connecting AI agents to external tools and data sources.
Platforms Anthropic
~10,000
Public MCP servers now live, forming the largest open ecosystem for agent-to-tool integration across commerce and beyond.
Platforms Anthropic
60+
Launch partners for Google's Agent Payments Protocol (AP2), enabling secure agent-driven transactions across the web.
Platforms Google
10-to-1
Expected ratio of AI agents to human sellers by 2028, agents will outnumber salespeople across digital commerce.
Platforms Gartner
100%+
Year-over-year growth in Amazon Rufus usage, indicating rapid consumer adoption of in-platform AI shopping assistants.
Platforms Amazon (AWS re:Invent 2025)
$10B+
Projected annual sales Amazon's Buy for Me feature is expected to generate by directing shoppers to third-party sites through AI agents.
Platforms Amazon
Merchant Readiness
Consumers are adopting AI shopping agents faster than most merchants are preparing their systems to handle them. This section measures that readiness gap directly, from how much of a typical product catalog is actually machine-readable to how many retailers have deployed agentic AI versus still experimenting with chatbots. For merchants, these numbers double as a checklist of what agent-facing commerce actually requires before it can work.
70%
Brands, retailers, and agencies already experimenting with or actively deploying agentic storefronts in 2026.
Merchant Readiness Swap/Glossy Survey
40%
Companies that have moved past experimentation into active agentic storefront deployment, the largest single segment.
Merchant Readiness Swap/Glossy Survey
Only 7%
Businesses with no current plans to adopt agentic commerce, the smallest segment in the 2026 industry survey.
Merchant Readiness Swap/Glossy Survey
88%
Current agentic commerce adopters who expect to increase their investment over the next 12 months.
Merchant Readiness Swap/Glossy Survey
89%
Merchants actively preparing their infrastructure for AI agent transactions, even as most acknowledge consumers are adopting faster.
Merchant Readiness Checkout.com
72%
Merchants who agree that consumer adoption of AI shopping agents is outpacing their own readiness to support those transactions.
Merchant Readiness Checkout.com
Only 15%
Top 100 UK retailers who say their payment systems are prepared to handle agent-initiated transactions today.
Merchant Readiness TLT UK Retail Report
3%
Current share of transactions that actually involve AI agents, a vast gap between merchant preparation and real transaction volume.
Merchant Readiness Payments Association (UK)
6 months
Time required to build custom ACP endpoints per AI agent without a managed integration layer like Stripe's Agentic Commerce Suite.
Merchant Readiness Stripe
62%
Companies citing improved brand discovery by AI agents as their leading goal for adopting agentic commerce.
Merchant Readiness Swap/Glossy Survey
57%
Top obstacles to agentic adoption: data requirements and ROI measurement, tied as the most-cited challenges by merchant teams.
Merchant Readiness Swap/Glossy Survey
75%
Merchants who say real-time permission revocation, letting consumers instantly cut off agent access, is critical to mass adoption.
Merchant Readiness Checkout.com
Only 1 in 5
US merchants with 80% or more of their catalog data in a machine-readable format.
Merchant Readiness PayPal
40% / 33%
E-commerce businesses still standardizing product pages for AI (40%), and share who have not started at all (33%).
Merchant Readiness Mirakl
51% / 15%
Retailers who name chatbots as their leading AI initiative (51%), versus only 15% using multi-agent systems.
Merchant Readiness TCS
68%
Retail executives who expect to deploy agentic AI within the next 12 to 24 months.
Merchant Readiness Deloitte
45% / 36%
Retailers who face data quality issues sometimes (45%) or often (36%).
Merchant Readiness Rithum
Personalization
When an AI agent gets the recommendation right, it tends to change buying behavior more than a traditional personalization engine would. This section covers the measurable lift from agent-driven personalization, higher order values, faster-converting onboarding flows, and lower return rates when shoppers can try before they buy. The upside is real, but conditional: these gains only show up when the agent is actually solving the shopper's problem, not just upselling.
15–25%
Average order value uplift reported by early agentic commerce adopters, when personalization genuinely solves the right problem, users spend more.
Personalization Industry benchmarks
32%
Faster sales growth for retailers operating their own branded AI agents compared to those relying solely on third-party discovery.
Personalization Industry data
836%
ROI increase achieved by Grubhub's agentic onboarding system, which personalizes multi-stage journeys based on real-time user actions.
Personalization Braze / Grubhub
188%
Rise in student signups after Grubhub deployed agent-driven personalized onboarding sequences through Braze.
Personalization Braze / Grubhub
49%
Consumers who now receive beauty product recommendations from generative AI, making it one of the most agent-penetrated retail categories.
Personalization Nielsen
58%
Virtual try-on sessions that end with the shopper adding the item to their cart, agents remove the uncertainty barrier.
Personalization Swap Platform Data
20%
Virtual try-on sessions that convert all the way to a completed purchase, significantly above typical browse-to-buy rates.
Personalization Swap Platform Data
20%
Reduction in product return rates when agentic storefronts with virtual try-on and guided discovery replace standard product pages.
Personalization Swap Platform Data
Agent Performance
An AI agent being available to shop for you is not the same as it being good at it. This section looks at how well current AI shopping agents actually perform on real tasks, from benchmark completion rates against human shoppers to how reliably they pass basic transaction safety checks. The gap between agent capability and agent reliability is still wide, which is the main reason most deployments today keep a human in the loop for anything beyond routine, low-risk purchases.
17.76%
Best score achieved by any AI model on a comprehensive shopping benchmark, versus 30% for human expert shoppers, agents are improving but not yet matching humans.
Agent Performance Research benchmark
35.42%
Rate at which current AI shopping agents pass safety checks, meaning nearly two-thirds of the time, agents fail to verify critical transaction safeguards.
Agent Performance Research benchmark
<65%
Completion rate for the top AI agent on harder multi-step shopping tasks in the WebMall evaluation framework.
Agent Performance WebMall benchmark
~25%
Success rate of GPT-4 agents when required to complete 8 consecutive shopping runs, dropping from ~60% on a single attempt.
Agent Performance Research benchmark
<5%
Failure rate threshold enterprises require before deploying AI agents in production purchasing workflows.
Agent Performance Enterprise research
7 months
Doubling rate for AI agent task-completion capability, roughly how long it takes agents to handle tasks that are twice as complex.
Agent Performance METR
<40%
Sales professionals who report that AI agents have actually improved their day-to-day productivity, despite broad deployment.
Agent Performance Gartner
69%
How much less likely a shopper is to return an item bought with AI assistance.
Agent Performance Adobe, Q2 2026
77%
AI-assisted shoppers who feel more confident in their purchase.
Agent Performance Adobe, Q2 2026
350M+
Shoppers served by Amazon Alexa for Shopping, with interactions up 5x and active users nearly doubling.
Agent Performance Amazon Q2 2026 Earnings
5×
Increase in shopping-intent queries Perplexity saw after expanding shopping access.
Agent Performance Perplexity
Security & Fraud
Autonomous purchasing creates a new attack surface, and fraud losses are already climbing alongside agent adoption. This section tracks AI-facilitated fraud trends, the regulatory response taking shape around consumer data and consent, and the steering and disclosure issues that arise when an agent's recommendations can be commercially influenced. Security and trust are tightly linked here: most of the hesitation consumers report about agentic shopping traces back to exactly these risks.
£423.5M
Remote purchase fraud losses in the UK during 2025, the baseline agentic payment systems must reduce, not increase.
Security & Fraud UK Finance
$16B
Total fraud losses in the US during 2025, up 25% year over year, setting the security context for autonomous agent transactions.
Security & Fraud FTC
978%
Growth in US generative-AI-related lawsuits between 2021 and 2025, signaling that liability frameworks are still catching up to the technology.
Security & Fraud Gallagher Re
10%
Maximum fine as a proportion of worldwide turnover that the UK CMA can impose for consumer law breaches involving AI systems.
Security & Fraud UK CMA
90 days
Maximum interval between consent refreshes required for third-party data access under UK open banking rules, a model for agent authorization cycles.
Security & Fraud FCA (UK)
61.2%
Selection rate for sponsored products when a conversational agent steers recommendations, versus 22.4% when users browse independently.
Security & Fraud Princeton University
<10%
Detection rate: how often consumers spotted that an AI agent was steering them toward a sponsored product when the agent's commercial intent was hidden.
Security & Fraud Princeton University
97%
Organizations that saw more AI-facilitated attacks in the past year, with an average annual loss of $4.5 million.
Security & Fraud Darwinium
25% / 40%
Rise in malicious bot transactions over 6 months (25% globally), with a 40% surge in the US alone.
Security & Fraud Visa
$5.13
Total cost of fraud per $1 of direct loss, up from $4.61.
Security & Fraud LexisNexis
78%
Financial institutions that expect more AI-related fraud going forward.
Security & Fraud Accenture
87%
CTOs and payments heads who see trust as the biggest barrier to agentic payments.
Security & Fraud Accenture
B2B & SaaS
Agentic AI's biggest near-term impact may not be in consumer shopping carts but in B2B sales and procurement workflows. This section covers enterprise adoption of AI agents, for lead scoring, customer interactions, and procurement, along with the ROI organizations are reporting once agents move from pilot to production. Enterprise adoption curves here are running well ahead of consumer-facing agentic commerce, which makes B2B the sector to watch for what's coming next.
70%
Reduction in sales cycle time for organizations processing 10,000+ daily inbound leads using agentic AI for scoring and routing.
B2B & SaaS Deloitte (2026)
2–3×
How much more predictive product-qualified signals (feature activation, invite events) are than email or firmographic signals for pipeline conversion.
B2B & SaaS SaaS industry data
90%
Companies in the Swap/Glossy survey already using AI for customer experience workflows, even if not yet for full agentic checkout.
B2B & SaaS Swap/Glossy Survey
37%
Businesses reporting declines in upper-funnel search traffic as AI agents intercept discovery queries before users reach traditional search results.
B2B & SaaS Industry survey
21%
Businesses reporting declines in lower-funnel search traffic, suggesting AI agents are capturing purchase-intent queries, not just browsing.
B2B & SaaS Industry survey
96%
Enterprises that are expanding their use of AI agents.
B2B & SaaS market.us, 2025
52%
Enterprises projected to have AI agents in production by the end of 2025.
B2B & SaaS Google Cloud, 2025
920%
Growth in agentic framework usage across developer repositories.
B2B & SaaS GitHub via BCG
88% / 74%
Agentic AI adopters reporting positive ROI (88%), versus 74% for generative AI broadly.
B2B & SaaS BCG
<5% to 40%
Share of enterprise applications expected to embed AI agents by 2026, up from under 5% in 2025.
B2B & SaaS Gartner
96%
Retailers exploring or implementing AI agents.
B2B & SaaS BCG
68%
Retailers who expect agents to handle most customer interactions within 5 years.
B2B & SaaS BCG
63%
Retailers who believe non-adopters will fall behind within 2 years.
B2B & SaaS BCG
83%
Executives who see agentic AI investment as essential for competitive advantage.
B2B & SaaS market.us, 2025
Holiday Commerce
Peak shopping seasons are where agentic commerce gets its most meaningful stress test: concentrated demand, high transaction volume, and no room for downtime. This section tracks how AI agents performed during the 2025 holiday season and major shopping events like Cyber Week, both in dollars influenced and in the operational load placed on the systems handling them. The results so far suggest agentic infrastructure can handle peak-scale demand, and that retailers using it saw measurably faster growth than those that didn't.
$262B
Global online sales driven by AI and agents during the 2025 holiday season, establishing agentic commerce as a material revenue channel.
Holiday Commerce Salesforce
$336.6B
Total global Cyber Week sales (Nov 25 to Dec 1, 2025), with AI agents influencing $67 billion, one-fifth of all purchases.
Holiday Commerce Salesforce
$67B
Value of Cyber Week purchases directly influenced by AI agents, representing 20% of total global sales during the peak shopping period.
Holiday Commerce Salesforce
59%
Higher revenue growth for companies with deployed AI agents during the 2025 holidays: 6.2% growth versus 3.9% for those without.
Holiday Commerce Salesforce
47%
Consumers who planned to use an AI shopping agent for their Christmas 2025 purchases, with a third relying on agents over Black Friday.
Holiday Commerce Checkout.com
142%
Increase in tasks handled by AI agents in December 2025 compared to the preceding two months, driven by holiday shopping demand.
Holiday Commerce Salesforce
61M
Orders powered by Salesforce's AI infrastructure through Cyber Week with 100% uptime, proving agent systems can handle peak scale.
Holiday Commerce Salesforce
66%
Spike in agentic AI service conversations in December 2025 versus November, as post-purchase queries flooded agent-enabled support channels.
Holiday Commerce Salesforce
Data Quality
An AI agent can only recommend or cite what it can actually parse, which makes structured, machine-readable product data a prerequisite for being visible to agents at all. This section covers how well current product pages are set up for machine readability, and which data types, core attributes, reviews, inventory feeds, matter most for being selected by an agent rather than just being technically discoverable. This is the technical SEO of the agentic era: invisible to shoppers, decisive for whether an agent ever surfaces your product.
66%
Average machine-readability score of US product pages, meaning 34% of typical product data is unparseable by AI shopping agents.
Data Quality Adobe
56%
Companies that say core product attributes are the most critical data type for feeding AI shopping agents effectively.
Data Quality Swap/Glossy Survey
54%
Companies prioritizing review signals as essential agent-facing data, social proof is as important for machines as for humans.
Data Quality Swap/Glossy Survey
43%
Businesses already feeding AI agents their brand voice and tone guidelines to maintain consistency across automated interactions.
Data Quality Swap/Glossy Survey
39%
Companies identifying real-time inventory feeds as a critical requirement for agent-driven commerce, agents cannot sell what isn't in stock.
Data Quality Swap/Glossy Survey
71%
E-commerce sites that are currently discoverable by AI agents, but discoverability alone doesn't guarantee selection.
Data Quality Swap/Glossy Survey
71%
Pages cited by ChatGPT that include structured data.
Data Quality Alhena
65%
Pages cited by Google AI Mode that include structured data.
Data Quality Alhena
3.1×
How much more often pages with structured data are cited by Google AI Overviews.
Data Quality SE Ranking
14%
Share of shopping queries where Google AI Overviews now appear.
Data Quality Search Engine Land
Frequently Asked Questions
How big is the agentic commerce market?
Estimates vary by analyst and by what counts as "agentic," but near-term figures put the market around $8 billion in 2026, scaling sharply over the next five years. Longer-range forecasts range from roughly $300 billion to over $5 trillion in global agent-influenced retail spend by 2030, with one estimate projecting as much as $3.5 trillion by 2031. The spread reflects differing definitions, from any AI-assisted purchase to fully autonomous, agent-initiated transactions.
What percentage of consumers use AI for shopping?
Roughly half of US consumers, around 51%, say they've used AI to shop in the past month, and about 61% use AI to research a purchase at some point in the process. Far fewer have gone further and let an agent complete a checkout on its own, under 7% by most current measures, showing adoption is concentrated in research and discovery rather than full delegation.
Which AI platform has the highest purchase completion rate?
Based on session-to-purchase conversion, Meta Muse currently converts disproportionately well, holding about 6.9% of AI shopping sessions but nearly 30% of agent-completed purchases, with a $118 average order value. Claude has the highest average order value of any platform measured, at $141, despite a smaller session share, suggesting it's used more for higher-consideration purchases than high-volume ones.
What is the difference between agent-assisted and fully autonomous commerce?
Agent-assisted commerce is when an AI tool helps a person research, compare, or narrow down options, but a human still makes the final decision and completes checkout. Fully autonomous commerce is when the agent completes the transaction itself, within limits the shopper pre-approved, such as a spending cap or product category. Current estimates suggest the large majority of agentic commerce today is still assisted rather than autonomous, with roughly 55% of commerce expected to be agent-assisted versus 18% fully autonomous in the years ahead.
How does structured data affect AI agent visibility?
Product pages with structured data are cited substantially more often by AI systems: roughly 71% of pages cited by ChatGPT and 65% of pages cited by Google AI Mode include structured data, and pages with it are cited about 3.1 times more often in Google AI Overviews than pages without it. For merchants, this makes structured markup less of a nice-to-have and more of a prerequisite for being surfaced by an AI shopping agent at all.
What are the biggest security risks in agentic commerce?
The two biggest concerns are fraud and loss of control. AI-facilitated fraud losses are already significant, with the total cost of fraud now estimated at $5.13 for every $1 of direct loss, and most financial institutions expect that to keep rising. On the consumer side, many shoppers worry about agents being commercially steered toward sponsored products, and a large share say they would stop using an agent entirely if they lost control over how it spends their money.
Which companies do consumers trust most with AI shopping agents?
PayPal, Amazon, and Visa currently form the most-trusted tier, each cited by around 42% of consumers as brands they'd trust to operate an AI shopping agent. Apple, Google, and Mastercard form a second tier around 28%, with Microsoft, Meta, and American Express trailing at roughly 15%. About one in five consumers say they aren't comfortable with an AI agent from any company on the list, a meaningful ceiling on adoption across the category.
Are merchants ready for agentic commerce?
Not yet, by most measures. Only about one in five US merchants have 80% or more of their catalog data in a machine-readable format, and a large share are still standardizing product pages for AI discovery or haven't started. Most retail executives say they expect to deploy agentic AI within the next one to two years, so the readiness gap is something merchants are actively racing to close rather than a permanent condition.
How to cite this page: WebAshes, “Agentic Commerce Statistics 2026.” Individual data points are sourced inline. Statistics are compiled from public research, earnings reports, and industry surveys, and this page is updated as new data becomes available. Last updated November 19, 2026.
Stay ahead of how AI is changing the web
Practical guides and insights for site owners navigating the AI era. No fluff. Unsubscribe any time.