Data & Statistics March 2026 • 14 min read

AI customer service statistics 2026

50+ verified data points on AI chatbots, automation, costs and ROI, and customer satisfaction in customer service, each linked to its source. Teams that want to act on these numbers typically start with an AI marketing agency or plug it into a broader marketing automation program.

30%

of service cases handled by AI (team estimate)

Source: Salesforce State of Service, 2025

4%

of service organizations got the AI ROI they expected

Source: Gartner, 4 Nov 2026

87%

of customers want an option to reach a human

Source: Gartner, 4 Aug 2026

Ruud ten Have

Ruud ten Have

Updated: 9 November 2026 • Sources: Gartner, Zendesk, Salesforce, Qualtrics, Eurostat, CBS

01 — Adoption Overview

AI customer service adoption

AI has moved to the top of the customer service agenda: budgets for AI grow much faster than service budgets, and leaders feel the pressure. The return is lagging behind, as the cost section shows.

30%

of service cases handled by AI (team estimate)

Source: Salesforce State of Service, 2025

50%

expected share handled by AI in 2027

Source: Salesforce State of Service, 2025

+38%

AI spending by service leaders (total budget +2%)

Source: Gartner, 26 Aug 2026

91%

of service leaders feel executive pressure to implement AI

Source: Gartner, 18 Feb 2026

Service and support leaders invested a median of 12% of their 2025 budget in AI, the highest of the ten business functions Gartner assessed. In Salesforce's survey of 6,500 service professionals in 40 countries, AI rose from #10 to #2 on service leaders' priority list in one year; improving the customer experience is still #1.

The next step is autonomy: 83% of CX leaders believe memory-rich AI agents are essential for personalized customer journeys. The brakes are mostly practical: 51% of service leaders say security concerns delayed or limited their AI initiatives, while 9 in 10 say the obstacles they met were expected.

Note: the 30% and 50% figures are estimates by service teams in a vendor survey (fieldwork April to June 2025), not measured case counts.

Sources: Salesforce State of Service, 2025, Gartner, 26 Aug 2026, Gartner, 8 Jul 2026, Zendesk CX Trends, 2026

02 — Chatbot Data

Chatbot statistics

Company chatbots are the most visible form of AI in customer service, but customers increasingly bring their own AI. The best documented case of a chatbot at scale remains Klarna.

Klarna: share of service chats handled by its AI assistant (first month) 2.3 million chats
2/3

Source: Klarna, 2024 (company figures, February 2024)

GenAI users who let GenAI complete a task for them 58% of users
58%

Source: Gartner, 4 Aug 2026 (3,566 customers)

Same, in B2B 74% of users
74%

Source: Gartner, 4 Aug 2026

Customers who say contact is easier when companies use GenAI 50% of customers
50%

Source: Gartner, 4 Aug 2026

Klarna AI assistant: first month (February 2024)

Metric Result
Conversations 2.3 million
Share of all service chats two-thirds
Work equivalent 700 full-time agents
Time to resolve an errand under 2 minutes (was 11)
Repeat inquiries 25% fewer
Availability 24/7, 23 markets, 35+ languages
Expected profit improvement 2024 $40 million

Klarna reported customer satisfaction "on par" with human agents. These are the company's own figures for a single business.

Meanwhile customers are roughly three times more likely to use third-party GenAI (ChatGPT, Gemini, Copilot) than a company's own chatbot to solve a service issue. Third-party GenAI use nearly doubled in a year, while company chatbot use has been flat since 2022.

The global chatbot market was estimated at $9.6 billion in 2025 and is projected to grow from $11.8 billion in 2026 to $41.2 billion in 2033 (19.6% per year). Customer service is the largest application.

Sources: Klarna, 2024, Gartner, 4 Aug 2026, Gartner, 8 Jul 2026, Grand View Research, 2026

03 — Cost Savings

Cost savings from AI support

The business case for AI in service is promised more often than proven. Recent Gartner surveys show how few organizations see the expected return so far. We found no reliable public benchmark for cost per chatbot interaction, so this page does not show one.

Expected ROI

4%

of service organizations achieved the ROI they expected from AI

Source: Gartner, 4 Nov 2026 (275 leaders)

Positive returns

24%

of service leaders show positive financial returns on AI use cases

Source: Gartner, 8 Jul 2026 (1,303 leaders)

AI budget

12%

median share of the 2025 service budget spent on AI

Source: Gartner, 8 Jul 2026

Foundations

1.8x

more invested in data and knowledge than in AI use cases by AI-forward organizations

Source: Gartner, 4 Nov 2026

Prediction 2029

-30%

operational costs as agentic AI resolves common issues

Source: Gartner, 5 Mar 2025

Single case

$40M

profit improvement Klarna expected from its AI assistant in 2024

Source: Klarna, 2024

AI spending by service leaders rose 38% while total service budgets grew just 2% (199 leaders, April to May 2026).

Gartner predicts that organizations building AI portfolios will achieve twice the business value of those pursuing standalone use cases by 2028.

Sources: Gartner, 4 Nov 2026, Gartner, 8 Jul 2026, Gartner, 26 Aug 2026, Gartner, 5 Mar 2025, Klarna, 2024

04 — Customer Satisfaction

Customer satisfaction & AI

Customers accept AI in service, on one condition: a human must stay within reach. AI in customer service also disappoints more often than AI in other tasks.

87%

say an option to reach a human agent is essential

Source: Gartner, 4 Aug 2026

1 in 5

AI service users saw no benefit, almost 4x the rate of AI in general

Source: Qualtrics XM Institute, 2025

72%

want to know if they are talking to an AI agent

Source: Salesforce Connected Customer, 2026

95%

expect an explanation for decisions made by AI

Source: Zendesk CX Trends, 2026

Consumer worries about AI in service, from Qualtrics' survey of more than 20,000 consumers in 14 countries (Q3 2025):

  • 53% fear misuse of personal data when companies automate interactions with AI, up 8 points in a year
  • 50% worry AI will keep them from reaching a human
  • 47% worry about job losses
  • only 39% trust companies to use personal data responsibly

Other customer signals: 74% find it frustrating to repeat their story to different agents and 63% demand more transparency than a year ago. Only 37% of companies currently give any reasoning behind AI decisions, while 80% of CX leaders say transparency will be a must-have (Zendesk). 61% say AI makes it even more important for companies to be trustworthy, and 64% see companies as careless with customer data (Salesforce).

Comfort depends on the task: 38% are comfortable with agents creating personalized content, but only 17% with an agent making financial decisions for them.

Sources: Gartner, 4 Aug 2026, Qualtrics XM Institute, 2025, Zendesk CX Trends, 2026, Salesforce Connected Customer, 2026

05 — Self-Service

Self-service & knowledge base statistics

Customers expect answers at any hour and faster than last year. Whether AI self-service delivers depends on the knowledge behind it.

74%

expect 24/7 customer service because of AI

Source: Zendesk CX Trends, 2026

88%

expect faster responses than a year ago

Source: Zendesk CX Trends, 2026

76%

would choose a company where text, images and video fit in one thread

Source: Zendesk CX Trends, 2026

85% of CX leaders say customers will leave over unresolved issues, even on first contact, and 82% say ignoring multimodal support will leave them behind.

Knowledge is the bottleneck. 58% of service leaders plan to upskill agents into knowledge management specialists, and AI-forward organizations invest nearly 1.8 times more in data and knowledge foundations than in the AI use cases themselves. A knowledge base that keeps learning is part of any serious AI implementation.

Note the gap with actual use: customers are about three times more likely to turn to third-party GenAI than to a company chatbot for a service issue.

Sources: Zendesk CX Trends, 2026, Gartner, 18 Feb 2026, Gartner, 4 Nov 2026, Gartner, 8 Jul 2026

06 — Agent + AI

Human agent + AI: augmented support

The best measured effect of AI in customer service is on human agents, not instead of them. Roles shift toward complex cases and knowledge work.

+14%

issues resolved per hour with an AI assistant

Source: NBER (Brynjolfsson, Li, Raymond), 2023

+34%

for novice and low-skilled agents

Source: NBER (Brynjolfsson, Li, Raymond), 2023

-20%

time on routine cases for reps using AI (about 4 hours a week)

Source: Salesforce State of Service, 2025

1/4

of the week spent on highly complex issues by reps with agentic AI

Source: Salesforce State of Service, 2025

How AI changes the service role

Finding Share Source
Reps using AI who say it creates growth opportunities 71% Salesforce State of Service, 2025
Reps using AI who developed new skills 86% Salesforce State of Service, 2025
Reps using AI whose role became more specialized 81% Salesforce State of Service, 2025
Organizations expanding human agent responsibilities more than 80% Gartner, 18 Feb 2026
Organizations moving some agents into new roles nearly 80% Gartner, 18 Feb 2026
Leaders adding new skills to the agent role and hiring profiles 84% Gartner, 18 Feb 2026

The productivity figures come from a study by Stanford and MIT economists of 5,179 support agents at one company: the AI assistant raised issues resolved per hour by 14% on average and by 34% for novices.

Service professionals also expect agentic AI to raise upsell revenue by 15%, which is why more companies connect service to AI-powered marketing automation.

Sources: NBER (Brynjolfsson, Li, Raymond), 2023, Salesforce State of Service, 2025, Gartner, 18 Feb 2026

07 — By Channel

AI customer service by channel

Which channel customers use for AI service is shifting: away from the company chatbot, toward their own AI assistants and toward conversations that mix text, images and video.

Third-party GenAI

~3x

more likely than a company chatbot in the latest service interaction

Source: Gartner, 8 Jul 2026

Company chatbots

flat

usage statistically unchanged since 2022

Source: Gartner, 8 Jul 2026

Multimodal

76%

want text, images and video in one thread

Source: Zendesk CX Trends, 2026

Chat at scale

35+

languages served 24/7 by Klarna's AI assistant in 23 markets

Source: Klarna, 2024

B2B buyers

46%

would work with an AI agent for faster service

Source: Salesforce Connected Customer, 2026

Human channel

87%

say a route to a human agent is essential

Source: Gartner, 4 Aug 2026

Sources: Gartner, 8 Jul 2026, Zendesk CX Trends, 2026, Klarna, 2024, Salesforce Connected Customer, 2026, Gartner, 4 Aug 2026

08 — Regional Breakdown

AI customer service by region

Few statistics offices measure AI in customer service specifically. What they do measure: how many companies use AI at all and how many use the language technologies behind chatbots. Europe and the Netherlands are best covered; vendor surveys add a view from Asia.

33.2%

of Dutch companies (10+ staff) used AI in 2025; EU 20.0%

Source: Eurostat, 2025

17%

of Dutch companies (2+ staff) used AI in 2025 (2023: 8%)

Source: CBS, 2025

22.9%

of Dutch companies analyse written language with AI; EU 11.8%

Source: Eurostat, 2025

57%

of Dutch adults have used a chatbot (18-25: 90.4%)

Source: CBS AI in society, 2026

Companies using AI, 2025 (10+ employees)

Country 2024 2025
Denmark 27.58% 42.03%
Belgium 24.71% 34.54%
Netherlands 23.06% 33.21%
Germany 19.75% 25.97%
EU-27 13.48% 19.95%

The Netherlands is well above the EU average for the AI techniques behind service chatbots: speech recognition 11.61% of companies (EU 7.22%) and natural language generation 16.34% (EU 8.76%). The Dutch 2025 figure carries a break in the time series.

Statistics Netherlands (CBS) counts companies with 2 or more staff and finds 17% using AI in 2025, rising with size from 14% (2-10 staff) to 66% (250+). Among AI users, 18% apply it to production or service processes. Companies that considered AI but did not adopt it cite lack of experience (73%), privacy (49%) and legal uncertainty (42%).

Outside Europe, service teams in Singapore estimate AI handles 30% of cases today and expect 41% by 2027. North America held the largest share of the chatbot market in 2025 (31.3%).

In the EU, Article 50 of the AI Act applies from 2 August 2026: providers must ensure that people are informed when they interact with an AI system, unless that is obvious from the context.

Sources: Eurostat, 2025, CBS, 2025, CBS AI in society, 2026, Salesforce Singapore, 2026, Grand View Research, 2026, EU AI Act, art. 50

10 — Key Takeaways

Key takeaways

The 50+ verified statistics on this page point to five takeaways for companies looking to deploy AI in their customer service.

1. Hybrid is the model customers ask for

87% of customers say a route to a human agent is essential, and nearly 1 in 5 who used AI for service got no benefit. Use AI for routine questions and keep the hand-off to people easy.

2. ROI is not automatic

Only 4% of service organizations achieved the ROI they expected, and 24% show positive financial returns, while AI spending rose 38%. Start with a measurable use case and a baseline.

3. Knowledge is the unlock

AI-forward organizations invest nearly 1.8 times more in data and knowledge than in AI use cases, and 58% of leaders plan to retrain agents as knowledge managers. Fix the knowledge base before the chatbot.

4. The Netherlands is ahead of Europe

33.2% of Dutch companies with 10+ staff used AI in 2025, against 20.0% in the EU, and Dutch use of text analysis and language generation is about double the EU level.

5. Agents augment people first

The best measured gain is 14% more issues resolved per hour (34% for novices). Gartner expects agentic AI to resolve 80% of common issues by 2029; until then the payoff sits with augmented agents.

Sources & Methodology

Our sources

Gartner: Customer service AI ROI survey (Nov 2026)
Gartner: Service AI spending survey (Aug 2026)
Gartner: Customer survey on GenAI in service (Aug 2026)
Gartner: Third-party GenAI vs chatbots (Jul 2026)
Gartner: Service leaders under pressure to implement AI (Feb 2026)
Gartner: Agentic AI prediction 2029 (Mar 2025)
Zendesk: CX Trends (2026)
Salesforce: State of Service, 7th edition (2025)
Salesforce: State of Service, Singapore (2026)
Salesforce: State of the AI Connected Customer (2026)
Qualtrics XM Institute: Consumer Experience Trends (2025)
NBER: Generative AI at Work, w31161 (2023)
Klarna: AI assistant press release (2024)
Grand View Research: Chatbot market (2026)
Eurostat: AI use in enterprises, isoc_eb_ai (2025)
CBS: Companies using AI (2025)
CBS: AI in society (2026)
EU AI Act: Article 50 (applies from 2 Aug 2026)

Every figure on this page links to the source it was taken from; we checked each source on 9 November 2026. Survey figures from vendors (Salesforce, Zendesk, Qualtrics) reflect their own respondents; forecasts are labeled as predictions. International figures are not Dutch data unless stated.

Related Reading

Further reading

Technology

What are AI agents?

The next wave in AI-powered support automation

Strategy

AI implementation in 5 steps

A proven roadmap for deploying AI in your organization

Data

AI & business statistics 2026

Implementation, ROI, and adoption across industries

Automation

Automate marketing with AI

Complete guide to AI-driven marketing automation

Service

AI marketing agency

Searchlab: your partner for AI-powered marketing

Data

AI marketing statistics 2026

All the numbers on AI in marketing

Ready to deploy AI in your customer service?

At Searchlab, we help businesses implement AI successfully — from strategy to execution. Discover what AI can do for your support operations, or browse our full online marketing services.

Explore our AI services Get in touch
Ruud ten Have

Ruud ten Have

Marketing & AI Strategy at Searchlab

Ruud is a digital marketer with 10+ years of experience in online advertising and AI implementation. At Searchlab, he combines strategic thinking with hands-on AI tooling to deliver measurable results for businesses.

Frequently Asked Questions

FAQ: AI customer service statistics

What percentage of companies use AI in customer service in 2026?
Gartner surveys show how central AI has become: service leaders invested a median of 12% of their 2025 budget in AI, the highest of ten business functions, and AI spending rose 38% while total service budgets grew 2%. In Salesforce's survey of 6,500 service professionals, teams estimate AI handles 30% of cases today and expect 50% by 2027. In the Netherlands, 33.2% of companies with 10+ employees used AI in 2025 (EU: 20.0%, Eurostat).
How much does an AI chatbot cost compared to a human agent?
There is no reliable public benchmark for cost per chatbot interaction versus a human agent. What is measured: only 4% of customer service organizations achieved the ROI they expected from AI and 24% show positive financial returns (Gartner, 2026). Gartner predicts agentic AI will cut operational costs by 30% by 2029. Klarna expected a $40 million profit improvement in 2024 from its AI assistant, its own figure for one company.
What is the average customer satisfaction score for AI-powered support?
There is no reliable industry-wide CSAT benchmark for AI support. Klarna reported satisfaction on par with human agents. At the same time, nearly 1 in 5 consumers who used AI for customer service saw no benefit, almost four times the rate for AI in general (Qualtrics, 2025), and 87% of customers say an option to reach a human agent is essential (Gartner, 2026).
What percentage of customer queries can an AI chatbot resolve on its own?
There is no reliable industry average. The best documented case is Klarna: in its first month its AI assistant handled 2.3 million conversations, two-thirds of all service chats, the work of 700 full-time agents, with errands resolved in under 2 minutes instead of 11. Gartner predicts agentic AI will autonomously resolve 80% of common customer service issues by 2029.
How does AI affect customer service agent productivity?
A study by Stanford and MIT economists of 5,179 support agents found an AI assistant raised issues resolved per hour by 14% on average and by 34% for novice and low-skilled workers (NBER, 2023). In Salesforce's 2025 survey, reps using AI spend 20% less time on routine cases, about four hours a week, and 86% developed new skills.

Related guides

Put these stats into action with our small-business marketing guides:

AI Marketing for Small Business: Complete Guide What AI marketing really means for a small business in 2026. AI Website Copy That Converts Write hero, services, About and CTAs with AI.

About Searchlab

We don’t just collect this data, we apply it daily as an AI-first marketing agency in Amsterdam. Dutch businesses can explore our AI marketing bureau services.

[ Free growth analysis ]

See where the
biggest gains are.

  1. TodayName, email and website. That is all we need.
  2. Within 5 working daysA document you can forward to your director or colleagues.
  3. ThenHalf an hour with the specialist who does the work.
You talk to the specialistNot a salesperson

Free, no obligation. We only use your details for the growth analysis.