136 AI customer support statistics for e-commerce 2026
AI, humans and the handoff. Every figure checked against its original source.
Overview
Customers want AI that works, and a person when it does not.
Most e-commerce brands now use AI somewhere in customer support. Customers are open to it, and they are clear about one condition: there has to be an easy way to reach a person. This page collects 136 statistics on that balance, from what customers expect and how teams are adopting AI, to the handoff, the cost of support, the e-commerce moments that drive contacts, messaging channels and the new disclosure rules.
Every number links to its original source, which we opened and checked. We keep each source's own wording for handled, contained and resolved, because they mean different things. Vendor research is labelled, and projections are marked as projections.
Written for e-commerce founders, heads of CX and operations leads, and for the writers and research tools that cover them. Last updated 20 October 2026.
Key takeaways
01 / What customers expect from support
What customers expect from support
Customers expect fast, always-on help that remembers them. A single bad experience is enough for many to leave.
Speed and availability
74% of consumers now expect 24/7 customer service because of AI.
AI raised the bar on availability. Customers now expect help at any hour.
60% of consumers say short wait times are critical when they contact customer service.
Waiting is still one of the main things customers judge support by.
The median first response time for e-commerce brands at $10M GMV on Gorgias is 6.3 hours, ranging from 1.6 hours to 9.1 hours by vertical.
By the vendor's own data, first replies still take hours at many e-commerce brands.
US customer satisfaction with online retailers held stable at 79 out of 100 in 2026.
Online retail satisfaction is steady, which makes service a point of difference.
US customer satisfaction with specialty retailers' call centers rose 4% to 81 out of 100 in 2026.
Phone support in specialty retail is improving, and customers notice.
Context and consistency
81% of consumers want agents to continue a conversation without backtracking, and 74% are frustrated when they have to repeat information.
Context is the expectation, not a bonus. Making people repeat themselves is the fastest way to lose them.
69% of consumers expect consistent interactions across a company's departments.
Customers expect one company, not a set of separate departments.
83.2% of UK customer experiences were right first time, the highest level recorded by the UK Customer Satisfaction Index.
Getting it right the first time is the measure customers reward.
The cost of one bad experience
85% of CX leaders say customers will drop brands that cannot resolve issues on first contact.
Leaders see first-contact resolution as the line between keeping and losing a customer.
34% of consumers reduce their spending with a company after a negative experience, and 13% cut their spending entirely.
A bad experience costs revenue straight away, not just goodwill.
Consumers cut spending after 58% of bad online retail experiences, the second highest rate of any industry after fast food (62%).
Online retail is one of the least forgiving categories after a bad experience.
US organizations risk $973 billion in sales from bad customer experiences, 4.9% of total consumption.
Poor experiences put a large share of US consumer spending at risk.
52% of US consumers say they stopped using or buying from a brand because of a bad experience with its products or services.
Half of US consumers have walked away from a brand after a bad experience.
29% of US consumers stopped using or buying from a brand because of poor customer experience, online or in person.
Poor customer experience alone is enough to lose more than a quarter of customers.
43% of consumers say a poor customer service experience will stop them from making a repeat purchase.
Service quality directly shapes whether customers buy again.
40% of consumers say they will stop doing business with a company after just one bad experience.
For many customers there is no second chance.
02 / AI adoption in service teams
AI adoption in service teams
AI agents went from a minority of service teams to a majority in a year. Most teams are still early in how deeply AI runs their operations.
Figure 2. AI agent adoption and share of cases handled by AI. Source: Salesforce, with State of Service estimates. 2027 is a projection.
Share of cases handled by AI
Service teams estimate that AI handles 30% of cases today and project that share will reach 50% by 2027.
AI already handles a large share of service volume, and teams expect it to handle half within two years.
Gartner predicts that by 2029 agentic AI will autonomously resolve 80% of common customer service issues without human intervention.
A bold forecast. It covers common issues only, and the date is three years out.
Business and technical decision-makers expect 68% of their customer service and support interactions with technology vendors to be handled by agentic AI within three years.
Buyers of technology expect most vendor support to run through agentic AI soon.
Intercom reports that its Fin AI agent resolves over 81% of Intercom's own customer support volume.
A vendor's figure for its own support team, useful as a ceiling, not a benchmark.
Adoption and maturity
Adoption of AI agents in customer service organizations rose from 39% in 2025 to 66% in 2026.
AI agents moved from minority to majority use in service teams within a year.
35% of large B2C contact centers already use agentic AI, and another 6% plan to adopt it in 2026.
Agentic AI is already in a third of large consumer contact centers.
82% of senior support leaders say their teams invested in AI for customer service in the last 12 months.
Investment in AI for support is now the norm, not the exception.
Only 10% of support professionals say their organization has reached mature AI deployment, with AI fully integrated into support operations at scale.
Few teams have AI fully built into how support runs day to day.
92% of organizations have implemented or piloted AI use cases in customer service.
Almost every organization has tried AI in service, though many are still piloting.
98% of contact center managers surveyed say AI is integrated into their contact center operations to some degree.
AI is now part of nearly every contact center, at very different levels of depth.
Retail and consumer products executives expect AI use in customer service, particularly for personalized responses and follow-ups, to grow by 236% over the next 12 months.
Retail leaders expect AI in customer service to grow sharply within a year.
Early results
70% of organizations with AI service agents say they see measurable value within 60 days of deployment.
Where agents are deployed, most teams report value quickly, by their own measure.
AI-mature contact centers report 85% greater contact center profitability than low-maturity peers.
Maturity tracks with profitability, though the link is not proven to be causal.
03 / How customers judge AI service
How customers judge AI service
Customers will use AI and often find it easier. Most still prefer a person, worry about their data, and increasingly bring their own AI tools.
Ease vs preference
50% of customers say their interactions are easier when companies use generative AI.
Customers see real benefits from AI service when it works.
80% of consumers are willing to use AI-powered customer service, but two-thirds still prefer speaking with a human.
Customers will use AI, but most still want a person available.
59% of consumers in the US, UK and Canada prefer an instant AI chatbot over waiting for a live agent.
Speed wins when the alternative is a long wait.
93.4% of US consumers surveyed prefer interacting with a human over AI for customer service.
In this US survey, almost all consumers preferred a human for service.
80.6% of US consumers surveyed believe AI is used in customer service primarily to save money, not to improve service.
Customers suspect AI is there to cut costs, so service quality has to prove otherwise.
Nearly 1 in 5 consumers who have used AI for customer service saw no benefits from the experience, a failure rate almost 4 times higher than for AI use in general.
AI service disappoints far more often than AI used elsewhere.
Bring-your-own AI
Customers are about 3 times more likely to use thirdparty generative AI tools than company-provided chatbots to resolve customer service issues.
Customers increasingly bring their own AI, which raises the bar for company chatbots.
About half of US adults (49%) now report using AI chatbots, and 44% say they use ChatGPT, up from 34% a year earlier.
Half of US adults now use AI chatbots, so customers arrive with expectations.
39% of US consumers say they have used AI for online shopping, and 85% of those say it improved their experience.
Shoppers who use AI mostly say it improves the experience.
66% of US consumers surveyed believe AI tools provide accurate shopping results.
Most shoppers trust AI shopping results, which raises the cost of errors.
46% of US shoppers always do additional research on retailer sites and apps after receiving an AI assistant's recommendation.
Shoppers still verify AI advice on the retailer's own site.
Trust, data and disclosure
53% of consumers say misuse of personal data is their top concern when companies use AI to automate interactions, up 8 points in a year.
Data misuse is the top worry when companies automate customer interactions.
71% of consumers consider it very or extremely important to know when they are interacting with an AI agent.
Disclosure matters to customers before the law requires it.
76% of consumers want clear rules for when an AI assistant acts, and 71% are concerned about how generative AI tools use their data.
Customers want clear limits on what AI assistants can do for them.
Only 46% of people globally are willing to trust AI systems, even though 66% use AI regularly.
Use of AI runs well ahead of trust in it.
32% of people in the United States say they trust AI, compared with 87% in China.
Trust in AI varies widely by country, and the US sits at the low end.
04 / The AI-to-human handoff
The AI-to-human handoff
The handoff is where AI service succeeds or fails. Customers want an easy way to a person, and they do not want to start over when they get there.
Access to a person
87% of customers say it is essential for companies using generative AI in customer service to provide an option to reach a human agent.
Access to a person is now a basic requirement for AI service.
1 in 2 consumers are concerned that companies' use of AI will prevent them from connecting with a human being.
Fear of losing access to people shapes how customers judge AI service.
61% of US consumers say they prefer a human agent over AI in 2026, with preference for a human up 5% year over year.
Preference for a human is rising, not falling, in this survey.
69% of US consumers who prefer human agents say they would switch to automated self-service if it could fully resolve their issue.
Most customers who prefer people would switch if selfservice could actually resolve the issue.
78% of consumers have tried to bypass AI agents to speak directly to a human.
Most customers have tried to get around an AI agent to reach a person.
78% of consumers have tried to bypass AI agents to speak directly to a human.
Most customers have tried to get around an AI agent to reach a person.
86% of US adults say knowing they can easily switch from automated help to a person increases their trust.
An easy path to a person builds trust in the automation itself.
61.2% of US consumers have yelled at automation to get routed to a human faster.
Frustrated customers will fight automation to reach a person.
Context at the handoff
83% of consumers say they still have to repeat themselves at least sometimes after being transferred from AI to a human agent.
The handoff is where most AI service breaks, because context does not transfer.
48% of organizations do not automatically pass information between virtual and human agents.
Half of organizations lose information between AI and human agents.
95% of consumers expect their information to be remembered across channels.
Customers expect to be known across every channel they use.
59% of US adults say AI-to-human handoffs feel difficult because they have to explain the issue again.
Repeating the problem after a handoff is the most common complaint.
60.1% of US consumers will repeat themselves only once before disengaging from customer service.
Customers tolerate repeating themselves once at most.
Limits of self-service
Only 14% of customer service and support issues are fully resolved in self-service, and only 36% of issues customers describe as very simple.
Self-service fully resolves only a small share of issues, even simple ones.
84% of consumers will give a virtual agent three attempts or fewer to resolve an issue.
Customers give a virtual agent very few chances before giving up on it.
In a field experiment on Alibaba's Taobao platform, 65.0% of AI-eligible chats handled by an agentic AI under human supervision were escalated to a human.
In real e-commerce traffic, most AI-eligible chats still needed a human.
05 / Agents, jobs and the human team
Agents, jobs and the human team
AI is changing what support agents do more than how many there are. Leaders are adding responsibilities and new roles, and many AI-driven cuts are expected to reverse.
Figure 5. Workforce actions by customer service and support leaders. Source: Gartner, April 2026.
Headcount and hiring
85% of customer service and support leaders are expanding human agent responsibilities, while only 31% have implemented or plan AI-driven frontline layoffs through 1Q27.
Leaders are giving human agents more to do, not removing them.
42% of customer service organizations are hiring specialized roles such as AI strategists, conversational AI designers and automation analysts to deploy and manage AI, and 55% report stable staffing while handling higher customer volumes.
AI creates new roles in service teams while overall staffing holds steady.
Gartner predicts that by 2027, 50% of companies that attributed customer service headcount reduction to AI will rehire staff for similar work under different job titles; only 20% of leaders surveyed had reduced agent staffing due to AI.
Cutting staff for AI often reverses. Few leaders have cut so far.
US employment of customer service representatives is projected to decline 5% from 2,666,000 in 2025 to 2,524,100 in 2035, a loss of 141,800 jobs.
The US expects fewer customer service jobs over the next decade.
Despite the projected decline, about 289,500 openings for US customer service representatives are projected each year on average from 2025 to 2035, all from replacing workers who leave the occupation or retire.
Even with decline, turnover keeps hundreds of thousands of roles open each year.
31% of contact center agents say they are likely to leave their current role within the next six months.
Attrition remains a major cost in contact centers, AI or not.
How agent roles are changing
Customer support agents with access to a generative AI assistant resolved 14% more issues per hour on average, with a 34% improvement for novice and low-skilled workers and minimal impact on experienced agents.
Gen AI lifted agent productivity most for newer agents, closing the experience gap.
86% of service reps say they have developed new skills and 81% say their role has become more specialized as AI takes on routine work.
Agents report broader skills and more specialized roles as AI takes routine work.
Service reps using AI spend 20% less time on routine cases, freeing an estimated four hours per week for more complex work.
AI frees agent time for the harder cases customers care about most.
94% of contact center agents expect AI to change their roles within three years, and 61% expect to handle more complex and technical work as a result.
Agents expect their jobs to change, and to become more complex.
The median pay for US customer service representatives was $44,770 a year, or $21.53 an hour, in May 2025.
A baseline for comparing the cost of human support with AI.
The wider labor market
The Philippine IT-BPM industry employed 1.89 million full-time equivalent workers and generated $40.3 billion in revenue in 2025, with a 2026 forecast of 1.94 million FTEs.
The Philippines remains a major source of customer support talent.
The Philippines' contact center and business process services segment employed 1.68 million FTEs and generated $33.9 billion in revenue in 2025.
Contact centers are the bulk of the Philippine IT-BPM industry.
Globally, 1 in 4 workers is in an occupation with some generative AI exposure, and clerical occupations have the highest exposure levels.
Clerical and support work is among the most exposed to generative AI.
Employers expect 170 million jobs to be created and 92 million displaced worldwide from 2025 to 2030, a net gain of 78 million, with clerical and secretarial workers seeing the largest absolute decline.
Employers expect net job growth overall, with clerical roles declining most.
Employment of US workers aged 22 to 25 in AI-exposed occupations stands 19% below where it would be had it kept pace with less-exposed peers, as of June 2026 payroll data.
Entry-level roles in AI-exposed jobs are already growing more slowly.
06 / Cost and ROI of AI support
Cost and ROI of AI support
AI budgets in service are high and measured returns are still rare. The cost of the contacts that remain with people is rising.
Figure 6. Cost per contact by channel and over time. Source: ContactBabel, with Deloitte Digital and Gartner data. The 2030 figure is a prediction.
Investment and returns
Only 24% of service and support leaders report positive financial returns across their AI use cases, despite investing a median 12% of their 2025 budget in AI.
Investment is high and measured returns are rare, which is the core ROI problem in service.
68% of US contact centers rank AI as a top 5 investment priority.
AI is a top investment priority for most US contact centers.
87% of customer service leaders surveyed by BCG say they will boost AI and generative AI investment over the next one to two years.
Leaders plan to keep increasing AI spend in service.
Only 28% of customer service leaders surveyed by BCG say generative AI has met or exceeded their expectations.
Most leaders say generative AI has not yet met expectations in service.
A Forrester study commissioned by Zendesk modeled a 301% three-year ROI for a composite company using Zendesk's AI-powered service tools.
A vendor-commissioned model. Treat it as a scenario, not an outcome.
Cost per contact
The average cost per assisted contact rose from $6.70 in 2024 to $7.80 in 2026, according to B2C contact center leaders.
Assisted contacts are getting more expensive as the simple ones move to AI.
The average cost of an inbound call in US contact centers is $7.20, 47% more than an email and 23% more than a web chat.
Channel choice changes the cost of every contact.
Gartner predicts that by 2030 the cost per resolution for generative AI in customer service will exceed $3, more than many B2C offshore human agents.
AI resolution is not automatically cheaper than people, and may cost more by 2030.
Customer service costs per transaction at Klarna fell 40% between Q1 2023 and Q1 2025, after the rollout of its AI assistant.
One company's reported result, with a large cost reduction over two years.
Efficiency gains
Contact center leaders believe AI will reduce their cost base by 28% on average over the next three years.
Leaders expect large savings, which still need to be proven.
39% of service leaders surveyed in 2026 report lower cost per contact as a result of AI.
Fewer than half of leaders see lower cost per contact so far.
64% of service leaders surveyed in 2026 report higher agent productivity from AI.
Productivity gains are more common than cost savings.
07 / The e-commerce moments that drive contacts
The e-commerce moments that drive contacts
Most e-commerce support starts at three moments: checkout, delivery and returns. Each is growing with online sales.
Figure 7. Reasons for checkout abandonment. Source: Baymard Institute, September 2025.
Checkout and cart abandonment
70.22% of online shopping carts are abandoned, based on the average of 50 cart abandonment studies.
Most carts are abandoned, which makes pre-purchase questions a support moment.
40% of US online shoppers who abandoned a checkout did so because extra costs such as shipping, tax and fees were too high.
Unexpected costs are the top reason shoppers leave at checkout.
20% of US online shoppers who abandoned a checkout did so because delivery was too slow.
Delivery speed questions start before the order is placed.
E-commerce accounted for 17.1% of total US retail sales in the second quarter of 2026.
Online is now a sixth of US retail, and its support load grows with it.
US e-commerce sales reached $340.2 billion in the second quarter of 2026, up 12.2% from a year earlier, while total retail sales rose 6.7%.
Online sales are growing almost twice as fast as total retail.
US consumers spent $257.8 billion online from November 1 to December 31, 2025, up 6.8% year over year.
Peak season sets the support volume teams must plan for.
Delivery and order tracking
86% of US online shoppers encountered at least one delivery issue in the past year, and 74% experienced a late delivery.
Delivery problems are close to universal, which makes order status a core support job.
45% of US online shoppers want real-time updates when a delivery problem occurs, before they have to ask.
Proactive delivery updates head off contacts before they happen.
68% of UK parcel recipients experienced a delivery issue in the past six months in 2025.
Delivery issues are common in the UK as well.
45% of UK consumers who contacted a delivery company about an issue were satisfied with the experience in 2025.
Contacting a carrier leaves most UK customers unsatisfied.
Among UK consumers dissatisfied with contacting a delivery company in 2025, 53% said it was because there was no way to contact the company by telephone.
When customers cannot call, satisfaction drops.
US parcel volume reached 23.1 billion shipments in 2025, a 3.3% increase from 22.4 billion in 2024.
Parcel volume keeps rising, and so does the number of delivery questions.
Returns
US retailers estimated that 15.8% of their annual sales would be returned in 2025, totaling $849.9 billion.
Returns are a large, recurring support workload for US retailers.
An estimated 19.3% of US online sales were expected to be returned in 2025.
Online orders come back far more often than in-store purchases.
71% of US consumers who returned an online purchase say they are less likely to shop with a retailer again after a poor returns experience, up from 67% in 2024.
A poor return experience costs the next sale.
08 / Channels: chat, voice and messaging
Channels: chat, voice and messaging
Customers want to show the problem, talk to it, or message it. Messaging apps are now a major business channel, with reach that varies widely by market.
Figure 8. Messaging reach by market. Source: Kantar for WhatsApp Business, with Ofcom, Zbooni and YouGov and Pew data.
Multimodal support
76% of consumers say they would choose a company that allows text, images and video in the same support thread without restarting.
Customers want to show the problem, not just describe it.
79% of CX leaders say customers now expect the option to use video or visual sharing during support interactions
Visual support is becoming an expected option.
Gartner predicts that by 2028, 30% of Fortune 500 companies will offer service only through a single, AIenabled channel that supports text, image and sound.
Gartner expects some large companies to move service to a single AI channel.
Voice AI
83% of respondents in Zendesk's CX Trends 2026 research believe voice AI has reached the point where it can significantly evolve the customer experience.
Voice is the next channel where AI is expected to change service.
84% of business leaders surveyed plan to increase their voice AI budgets in the next 12 months, and 67% consider voice AI core to their product and business strategy.
Voice AI budgets are set to grow, by this vendor's survey.
Phone remains the most preferred customer service channel overall in the US, UK and Germany, preferred by 23% of Gen Z consumers, 33% of Gen X, 47% of Baby Boomers and 66% of the Silent Generation.
Phone is still the preferred channel for many customers, especially older ones.
90% of consumers failed to correctly identify AIgenerated voice clips, even though 72% say they can immediately identify an AI agent in voice interactions.
AI voices are now hard to tell apart from people, which is why disclosure rules matter.
WhatsApp and messaging, including the GCC
More than 1 million businesses were using Meta Business Agents on WhatsApp and Messenger every week to talk to customers or complete sales, as of Q2 2026.
AI agents on messaging apps are now used by more than a million businesses each week.
More than 10 million conversations each week were being facilitated through Meta's Business AIs in Q1 2026, with weekly conversations up 10x since the start of 2026.
Business AI conversations on Meta's apps grew tenfold in a single quarter.
Paid messaging on WhatsApp crossed a $2 billion annual revenue run-rate in Q4 2025, and US click-to-message ad revenue grew more than 50% year over year.
Messaging is now a large commercial channel, not just a support inbox.
73.3% of consumers across 22 markets, including Saudi Arabia and the UAE, prefer messaging when communicating with a business.
In Meta's research, most consumers prefer messaging a business over other channels.
67.7% of consumers across 22 markets agree that getting a response from an AI chatbot is helpful.
In Meta's research, most consumers find AI chatbot replies helpful on messaging apps.
WhatsApp reached 90% of UK online adults in May 2025, and 74% used it every day, up from 64% in May 2024.
In the UK, WhatsApp is near universal and used daily.
33% of US adults say they ever use WhatsApp, according to Pew Research Center's 2026 survey.
WhatsApp reach is much lower in the US than in the UK or the Gulf.
65% of UAE residents used WhatsApp to contact a business about a product or service in the past year, ahead of call centres (55%) and email (48%), and 59% picked WhatsApp as their preferred customer support channel.
In the UAE, WhatsApp is already the leading way to contact a business.
09 / Regulation and disclosure
Regulation and disclosure
Telling customers they are talking to AI is now a legal requirement in the EU and, in specific cases, in several US states.
Figure 9. Selected AI disclosure rules by date. Source: European Commission, with California, Utah and UK CMA texts
EU AI Act
Since 2 August 2026, the EU AI Act (Article 50) requires providers to ensure people are informed when they interact with an AI system, unless obvious, with fines of up to 15 million euros or 3% of worldwide turnover.
AI disclosure is now a legal requirement for customer-facing AI in the EU.
The EU AI Act entered into force on 1 August 2024, with bans on prohibited practices applying from 2 February 2025 and most of the Act applying from 2 August 2026.
Most of the EU AI Act now applies.
Under the AI Omnibus changes, EU AI Act rules for highrisk systems in areas such as biometrics and education apply from 2 December 2027, and from 2 August 2028 for AI built into products.
High-risk rules were delayed. The chatbot transparency duty was not.
US and UK rules
California has required bots to disclose they are bots since 1 July 2019 when used online to mislead people about their artificial identity to sell goods or services or influence a vote; the platform rules apply to sites with 10,000,000 or more unique monthly US visitors.
The first US bot disclosure law has been in force since 2019.
California's companion chatbot law (SB 243), signed 13 October 2025, requires operators to give a clear and conspicuous notice that the chatbot is artificially generated and not human where a reasonable person could be misled, and excludes bots used only for customer service.
California's newer chatbot law covers companions, not customer service bots.
California's SB 243 requires companion chatbots to remind users known to be minors at least every three hours that the chatbot is not human, and lets injured users recover the greater of actual damages or $1,000 per violation.
Penalties under California's companion chatbot law are per violation.
Since 7 May 2025, Utah law requires a business using generative AI with consumers to disclose that the person is interacting with AI and not a human when the person clearly asks.
In Utah, a business must say it is AI when a customer asks.
Utah can fine businesses up to $2,500 per violation of its generative AI disclosure law, and up to $5,000 per violation of an existing order.
Utah's disclosure law carries per-violation fines.
Since 7 May 2025, Utah requires licensed professionals to prominently disclose generative AI use in high-risk interactions, verbally at the start of a spoken interaction or in writing before a written one.
Regulated professions in Utah must disclose AI use up front in high-risk interactions.
In September 2024 the FTC announced five law enforcement actions against deceptive AI claims under Operation AI Comply, including a proposed $193,000 payment by DoNotPay.
US regulators enforce against misleading AI claims under existing law.
In September 2025 the FTC ordered 7 companies offering AI chatbots that act as companions to report how they measure and monitor harms to children and teens.
US regulators are studying chatbot harms to young users.
About 98 million US users, roughly 37% of the population, engaged with a bank's chatbot in 2022, and the CFPB states these chatbots must comply with all applicable federal consumer financial laws.
US financial chatbots must follow existing consumer finance law.
The UK CMA's March 2026 guidance on AI agents says businesses should tell people when dealing with AI might affect their decisions, and consumer law breaches can bring fines of up to 10% of worldwide turnover.
UK guidance asks businesses to be open about AI agents, with large fines for breaches.
What customers want disclosed
95% of consumers expect an explanation for AI-made decisions.
Customers expect AI decisions to be explained.
80% of CX leaders agree AI transparency will be required for any customer-facing AI within two years.
CX leaders expect transparency to become mandatory soon.
76% of US adults say it is extremely or very important to be able to tell whether pictures, videos and text were made by AI or by people.
Most Americans want to know when content is made by AI.
Nearly 75% of consumers want to know if they are communicating with an AI agent.
Most customers want to know when they are talking to an AI agent.
Our view
What these statistics mean
We run AI agents and human support teams together for e-commerce clients. Read together, these numbers point to four practical lessons.
If you are planning AI support with a human team behind it, talk to us. project@incrediblevisibility.com
Method
Methodology
We reviewed more than 200 candidate figures and kept 136. Each kept figure was checked against its original publication on 20 October 2026: the page had to load, show the exact number, and match the population and date we describe.
We used government and statistical agencies, regulators, legal texts and peer-reviewed research first, then original research with a stated method. Customer service research is dominated by help desk, contact center and chatbot vendors, so many sources here are vendor surveys. We include them only where the method and sample are stated, label them, and exclude vendor marketing claims. Company-reported results are labelled as vendor data.
Dates are the original publication or data release. Where a report page states no publication month, we cite the year only. Projections and predictions are labelled. Surveys measure different populations and use different terms for handled and resolved, so figures from different sources should not be averaged.
We refresh this page monthly. The US Census e-commerce figures update on 19 November 2026. Spotted an error or a newer edition? Email project@incrediblevisibility.com.
Cite this page
Incredible Visibility. "136 AI Customer Support Statistics for E-commerce 2026: AI, Humans and the Handoff." Last updated October 20, 2026. https://incrediblevisibility.com/research/ai-customer-support-statistics
Frequently asked questions
Service teams surveyed by Salesforce estimate AI handles 30% of cases today and project 50% by 2027. Gartner predicts agentic AI will autonomously resolve 80% of common customer service issues by 2029. Both are estimates, and handled is not the same as resolved.
Both, depending on the moment. Gartner found 50% of customers say service is easier with generative AI, and 87% say companies must offer access to a human. A Kinsta survey found 93.4% of US consumers prefer a human, while Five9 found 80% are willing to use AI-powered service.
Not automatically. Only 24% of service leaders report positive financial returns across their AI use cases, according to Gartner, which also predicts generative AI cost per resolution will exceed $3 by 2030. A US inbound call costs $7.20 on average, according to ContactBabel.
In the EU, yes, since 2 August 2026 under Article 50 of the AI Act, unless it is obvious. In the US, California has required bot disclosure in specific sales and election contexts since 2019, and Utah requires disclosure when a customer asks. This is not legal advice.
Sources
- Zendesk (November 2025). https://www.zendesk.com/newsroom/press-releases/contextual-intelligence-becomes-the-new-standard-forexceptional-customer-experience-in-2026/
- Five9 (March 2025). https://five9.com/news/news-releases/new-five9-research-reveals-balancing-ai-human-connection-key-winningcustomer
- Gorgias vendor data (April 2026). https://www.gorgias.com/research/stop-benchmarking-against-the-average
- ACSI (January 2026). https://theacsi.com/news-and-resources/press-releases/2026/01/27/press-release-retail-and-consumer-shippingstudy-2026/
- Salesforce (October 2024). https://www.salesforce.com/news/stories/ai-customer-research/
- Institute of Customer Service (January 2026). https://www.instituteofcustomerservice.com/?p=210803
- Qualtrics XM Institute (November 2025). https://www.qualtrics.com/articles/news/businesses-risk-3-trillion-sales-poor-customer-experiencesconsumers-cut-spending/
- Qualtrics XM Institute (November 2025). https://qualtrics.com/articles/customer-experience/3-trillion-risk-due-bad-customerexperiences-2026
- PwC (September 2025). https://www.pwc.com/us/en/services/consulting/business-transformation/library/2025-customer-experiencesurvey.html
- Salesforce (November 2025). https://www.salesforce.com/news/?p=90947
- Gartner (March 2025). https://www.gartner.com/en/newsroom/press-releases/2025-03-05-gartner-predicts-agentic-ai-will-autonomouslyresolve-80-percent-of-common-customer-service-issues-without-human-intervention-by-20290
- Cisco (May 2025). https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2025/m05/agentic-ai-poised-to-handle-68-of-customer-service-andsupport-interactions-by-2028.html
- Intercom vendor data (2026). https://www.intercom.com/blog/automate-customer-service-while-improving-customer-experience/
- Salesforce (May 2026). https://www.salesforce.com/news/stories/ai-service-agents-improve-customer-satisfaction/
- Deloitte Digital (June 2026). https://www.deloittedigital.com/content/dam/digital/global/documents/insights-20260603-global-contact-centersurvey.pdf
- Intercom (2026). https://www.intercom.com/customer-transformation-report
- Five9 (June 2026). https://www.five9.com/news/news-releases/new-five9-research-ai-adoption-cx-hits-92-consumer-trust-still-dependshuman
- Calabrio (March 2025). https://www.businesswire.com/news/home/20250325095245/en
- IBM Institute for Business Value (January 2025). https://newsroom.ibm.com/2025-01-07-ibm-study-ai-spending-expected-to-surge-52-
- beyond-it-budgets-as-retail-brands-embrace-enterprise-wide-innovation
- Gartner (August 2026). https://www.gartner.com/en/newsroom/press-releases/2026-08-04-gartner-survey-finds-87-percent-of-customerssay-companies-using-genai-for-customer-service-must-provide-access-to-a-human-agent0
- Kinsta (April 2025). https://kinsta.com/blog/ai-vs-human-customer-service/
- Qualtrics XM Institute (October 2025). https://www.qualtrics.com/news/ai-powered-customer-service-fails-at-four-times-the-rate-of-othertasks/
- Gartner (July 2026). https://www.gartner.com/en/newsroom/press-releases/2026-07-08-gartner-survey-finds-customers-are-three-timesmore-likely-to-use-third-party-genai-than-company-provided-chatbots-for-customer-service
- Pew Research Center (June 2026). https://www.pewresearch.org/internet/2026/06/17/americans-and-ai-2026-chatbots-smart-devices-andviews-on-impact/
- Adobe (April 2026). https://business.adobe.com/blog/ai-traffic-surge-retail-sites-not-machine-readable
- Profitero vendor data (July 2026). https://profitero.com/blog/5-shopper-behavior-trends-brands-need-to-prepare-for-in-the-age-of-aiassisted-commerce
- Capgemini Research Institute (January 2026). https://www.capgemini.com/insights/research-library/what-matters-to-todays-consumer-2026/
- KPMG and University of Melbourne (May 2025). https://kpmg.com/za/en/newsroom/press-releases/2025/05/global-study-reveals-trust-of-airemains-a-critical-challenge-reflecting-tension-between-benefits-and-risks.html
- Edelman (November 2025). https://www.edelman.com/insights/familiarity-breeds-trust-in-ai
- Verint (June 2026). https://www.verint.com/blog/ai-vs-human-customer-service-paradox/
- Twilio (2026). https://www.twilio.com/en-us/report/scale-automation-with-losing-customer-confidence
- Liveops vendor data (June 2026). https://liveops.com/blog/liveops-2026-resolution-gap-report-why-ai-powered-cx-is-faster-than-ever-butcustomer-effort-remains-high/
- Parloa (2026). https://go.parloa.com/consumer-patience-index
- Genesys (July 2026). https://www.genesys.com/company/newsroom/announcements/more-than-half-of-consumers-would-rather-doanything-else-than-contact-customer-service-genesys-research-finds
- Gartner (August 2024). https://www.gartner.com/en/newsroom/press-releases/2024-08-19-gartner-survey-finds-only-14-percent-ofcustomer-service-issues-are-fully-resolved-in-self-service
- Wang et al., arXiv (Alibaba field experiment) (May 2026). https://arxiv.org/abs/2605.14830
- Gartner (April 2026). https://www.gartner.com/en/newsroom/press-releases/2026-04-28-gartner-survey-finds-eighty-five-percent-ofservice-and-support-leaders-are-expanding-human-agent-responsibilities-despite-expectations-of-mass-ai-layoffs
- Gartner (December 2025). https://www.gartner.com/en/newsroom/press-releases/2025-12-02-gartner-survey-finds-only-20-percent-ofcustomer-service-leaders-report-ai-driven-headcount-reduction
- Gartner (February 2026). https://www.gartner.com/en/newsroom/press-releases/2026-02-03-gartner-predicts-half-of-companies-that-cutcustomer-service-staff-due-to-ai-will-rehire-by-2027
- US Bureau of Labor Statistics (August 2026). https://www.bls.gov/ooh/office-and-administrative-support/customer-servicerepresentatives.htm
- Verint vendor data (April 2026). https://www.verint.com/press-room/2026-press-releases/nearly-one-third-of-contact-center-agents-plan-toquit-as-agent-experience-falls-short/
- NBER (Brynjolfsson, Li, Raymond) (November 2023). https://www.nber.org/papers/w31161
- Salesforce (November 2025). https://www.salesforce.com/news/stories/state-of-service-report-announcement-2025/
- IBPAP (January 2026). https://admin.ibpap.org/storage/hub-resources/7Hv9U2uoLJVx2kxyRMObNm6W6L0MR57l9dZfrez7.pdf
- International Labour Organization (May 2025). https://www.ilo.org/publications/generative-ai-and-jobs-refined-global-index-occupationalexposure
- World Economic Forum (January 2025). https://www.weforum.org/publications/the-future-of-jobs-report-2025/digest/
- Stanford Digital Economy Lab (August 2026). https://digitaleconomy.stanford.edu/publications/canaries-in-the-coal-mine/
- ContactBabel (March 2026). https://www.contactbabel.com/the-us-contact-center-decision-makers-guide/
- BCG (September 2025). https://www.bcg.com/assets/2025/executive-perspectives-unlocking-impact-from-agentic-ai-in-customerservice-23-september.pdf
- Forrester (commissioned by Zendesk) vendor data (July 2025). https://tei.forrester.com/go/zendesk/service/
- Gartner (January 2026). https://www.gartner.com/en/newsroom/press-releases/2026-01-26-gartner-predicts-genai-cost-per-resolution-forcustomer-service-will-exceed-offshore-human-agent-costs-by-2030
- Klarna vendor data (May 2025). https://s205.q4cdn.com/644747736/files/doc_financials/2025/q1/Q1-25-Klarna-Earnings-Release.pdf
- Deloitte (February 2026). https://www.deloitte.com/us/en/about/press-room/the-future-of-service.html
- Baymard Institute (September 2025). https://baymard.com/lists/cart-abandonment-rate
- US Census Bureau (August 2026). https://www.census.gov/retail/ecommerce.html
- Adobe (January 2026). https://news.adobe.com/news/2026/01/adobe-holiday-shopping-season
- Narvar vendor data (November 2025). https://corp.narvar.com/press/new-narvar-state-of-post-purchase-report
- Ofcom (October 2025). https://www.ofcom.org.uk/post/deliveries-and-charges/best-and-worst-parcel-firms-for-customer-satisfactionrevealed-2025
- Ofcom (October 2025). https://www.ofcom.org.uk/siteassets/resources/documents/postal-services/monitoring-reports/2024-25/measuringuser-experience-of-parcel-delivery-to-residential-addresses-2025.pdf
- Pitney Bowes (2026). https://www.pitneybowes.com/us/shipping-index.html
- National Retail Federation and Happy Returns (October 2025). https://nrf.com/media-center/press-releases/consumers-expected-to-returnnearly-850-billion-in-merchandise-in-2025
- Gartner (December 2024). https://www.gartner.com/en/newsroom/press-releases/2024-12-11-gartner-predicts-that-30-percent-offortune-500-companies-will-offer-service-through-only-a-single-ai-enabled-channel-by-2028
- Deepgram vendor data (2025). https://deepgram.com/2025-state-of-voice-ai-report
- Five9 (June 2026). https://www.businesswire.com/news/home/20260624086700/en
- Twilio (November 2025). https://www.twilio.com/en-us/press/releases/twilio-s-latest-report-highlights-conversational-ai-s-rapid-adop
- Meta (July 2026). https://s21.q4cdn.com/399680738/files/doc_financials/2026/q2/META-Q2-2026-Prepared-Remarks.pdf
- Meta (April 2026). https://s21.q4cdn.com/399680738/files/doc_financials/2026/q1/META-Q1-2026-Prepared-Remarks.pdf
- Meta (January 2026). https://s21.q4cdn.com/399680738/files/doc_financials/2025/q4/META-Q4-2025-Prepared-Remarks.pdf
- WhatsApp Business (Meta) / Kantar (2026). https://whatsappbusiness.com/resources/resource-library/state-of-business-messaging/
- Ofcom (December 2025). https://www.ofcom.org.uk/siteassets/resources/documents/research-and-data/online-research/online-nation/2025/
- online-nations-report-2025.pdf
- Pew Research Center (June 2026). https://www.pewresearch.org/internet/fact-sheet/social-media/
- Zbooni / YouGov vendor data (February 2024). https://media.zbooni.shop/marketing/reports/the-mena-ccommerce-report-2024.pdf
- European Commission (July 2026). https://digital-strategy.ec.europa.eu/en/faqs/transparency-obligations-under-article-50-ai-act
- European Commission (October 2026). https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai
- California Legislature (SB 1001, Bus. and Prof. Code 17940-17943) (September 2018). https://legiscan.com/CA/text/SB1001/2017
- California Legislature (SB 243) (October 2025). https://legiscan.com/CA/text/SB243/2025
- California State Senate Judiciary Committee (January 2026). https://sjud.senate.ca.gov/media/2016
- Utah Legislature (Utah Code 13-77) (May 2025). https://le.utah.gov/xcode/Title13/Chapter77/C13-77_2025050820250508.pdf
- Utah Legislature (SB 226, 2025) (March 2025). https://le.utah.gov/Session/2025/bills/enrolled/SB0226.pdf
- US Federal Trade Commission (September 2024). https://www.ftc.gov/news-events/news/press-releases/2024/09/ftc-announces-crackdowndeceptive-ai-claims-schemes
- US Federal Trade Commission (September 2025). https://www.ftc.gov/news-events/news/press-releases/2025/09/ftc-launches-inquiry-aichatbots-acting-companions
- US Consumer Financial Protection Bureau (June 2023). https://www.consumerfinance.gov/data-research/research-reports/chatbots-inconsumer-finance/chatbots-in-consumer-finance/
- UK Competition and Markets Authority (March 2026). https://www.gov.uk/government/publications/complying-with-consumer-law-whenusing-ai-agents/complying-with-consumer-law-when-using-ai-agents
- Pew Research Center (September 2025). https://www.pewresearch.org/science/2025/09/17/how-americans-view-ai-and-its-impact-onpeople-and-society/
Let's build something that lasts.
Whether you're launching a new product, modernizing a platform, or scaling your engineering organization — we bring the experience, depth, and clarity to help you move forward confidently.