Integration and legacy
In large organisations, AI meets existing telephony, CRM, collection and policy systems. Most barriers reported by enterprises are about connecting to those systems safely.
Read the insightTa strona nie jest jeszcze dostępna w języku polskim. Wyświetlana jest wersja angielska.
Market insight
Spending on AI in customer service is rising, but few organisations report the return they expected. The gap sits between deploying a model and running a process that delivers a measurable result.
Surveys published between December 2025 and November 2026 consistently find that only a minority of organisations report the return they expected from AI in customer service.
Organisations that report results differ less in the technology they use than in how they run it: governance, measurement and continued optimisation after go-live.
Adoption is broad but shallow: many organisations use AI tools without having put measurable use cases into production.
28% of 180 customer-service leaders reported measurable business value from generative AI.
BCG, Agentic AI Is the New Frontier in Customer Service Transformation, 3 December 2025. Survey of leaders; self-reported.
4% of 275 customer-service and support leaders said they had achieved the expected return on their AI initiatives.
Gartner, press release, 4 November 2026. Survey July–August 2026; self-reported. The same release forecasts that by 2028 portfolios of connected use cases could deliver twice the value; that part is a forecast.
37% of respondents reported a positive impact of AI on EBIT; 6% qualified as “AI high performers”.
McKinsey, The State of AI (1,719 respondents, 97 countries), as reported by TechTarget, 25 September 2026. Secondary report; covers AI across all business functions, not customer service alone.
Among 440 customer-care leaders, 40% of leaders (top 10%) had significantly improved their CX scores in the past 12 months, against 12% of laggards (bottom 30%); 37% of laggards were not at all comfortable with AI handling end-to-end interactions, against 4% of leaders.
McKinsey, Building trust: How customer care leaders pull ahead with AI, 23 February 2026. Self-reported; comparison of maturity groups, not a causal effect.
EY classified respondents as Achievers (23%), Challengers (33%) and Aspirants (44%). More than 40% of the two less advanced groups cited governance, compliance or change management as obstacles; 91% of Achievers had voice AI deployed at scale.
EY, Demystifying AI in customer service: Navigating AI-led transformation to value realization, 4 November 2026. More than 100 survey responses and executive interviews combined, Q4 2025–Q1 2026; country mix not disclosed; reported effects, not audited.
Contact centres classed as AI-mature reported 85% higher profitability than less mature peers.
Deloitte Digital, 2026 Global Contact Center Survey, 4 June 2026. Correlation between subgroups, not a causal effect.
99% of organisations used some form of automation, but only 23% of all respondents said they were both highly automated and consistently optimised using CX insights.
CallMiner, CX Landscape Report 2026. Vendor-commissioned survey of 700 CX leaders in six markets.
In 2025, 20% of EU enterprises with ten or more employees reported using AI technologies; among large enterprises the share was 55%.
Eurostat, European business – AI adoption 2025 (KS-01-26-010), 2026. All uses of AI across the economy; not a measure of AI in customer contact.
In a survey of 73 banks, 42% had given employees AI tools, but only 35% had implemented measurable use cases.
D.A. Davidson bank survey, as reported by Banking Dive, 10 July 2026. Secondary report; US banks; small sample.
Vee's interpretation. The analysis below is our own reading of the sources above, not a finding of the organisations cited.
The surveys above measure different populations with different definitions of value, so their percentages should not be compared or combined. Read together, they point in one direction: putting a model into service is now quick, while producing a measured business result takes longer and depends on how the service is run.
In customer-contact operations the gap usually opens after launch. An application can hold conversations competently and still leave the business outcome – a payment, a kept appointment, a renewed policy – unchanged if scripts, contact timing, routing and escalation rules are not adjusted against results.
The maturity comparisons published by McKinsey, EY and Deloitte Digital suggest that the organisations reporting results treat AI as an operating discipline with an owner, defined measures and a cycle of improvement, rather than as a project that ends at go-live. These are correlations reported by respondents, not proof of cause.
In banking and consumer finance the outcome of a contact process is often directly measurable – a payment made, a promise kept, an application completed – so a shortfall between activity and result becomes visible quickly. The bank survey reported by Banking Dive indicates that access to AI tools is more common than measurable use cases.
In insurance and telecommunications, contact is high in volume and repetitive (renewals, claim status, retention, billing). The value of automation depends on linking the conversation to the systems that hold the policy or the account, which is why integration is treated as a separate theme in these insights.
For healthcare and energy, the research we hold does not measure AI returns directly. The healthcare evidence concerns access and workload and is set out under Service economics; we hold no comparable sector survey for energy and utilities.
Several of the sources are vendor-sponsored or rely on self-assessment. The 4% (Gartner) and 28% (BCG) figures measure different things: achieving the return that was expected, and reporting any measurable value. Neither describes the market as a whole, and neither is evidence about any individual supplier.
Broad adoption statistics, such as Eurostat’s, count any use of AI across the economy. They show how widely the technology has spread, not how much of it has reached production in customer-contact processes or what it has achieved there.
Forrester’s Predictions 2027 (21 October 2026) expect one in five companies to deploy fully agentic orchestration of customer journeys in 2027, with greater pressure to measure customer experience. Forecast.
Forrester also predicts that rising AI costs will lead B2C leaders to scale back about a third of their AI initiatives in 2027 (marketing, CX and digital). Forecast.
Gartner’s forecast that by 2028 portfolios of connected AI use cases could deliver twice the value of isolated ones (4 November 2026). Forecast.
Vee’s Q4 2026 report, which the Q3 2026 report schedules for February 2027.
In its Q3 2026 report (letter to shareholders, p. 7) Vee cites the BCG and Gartner findings and states that, in large enterprises, deploying technology is part of a much wider process covering work organisation, security, integrations, quality control, data analysis and the systematic improvement of effectiveness. The same letter (p. 8) describes Vee’s model as a managed AI Call Center service covering the organisation of contact, running the process, quality control, data analysis, measuring effectiveness and continuous improvement of results. The external findings describe the market; they are not evidence of Vee’s own results.
Market data on this page come from third-party publications and are attributed to their authors. They describe the market, not Vee’s results. Forecasts are the publishers’ own and are not forecasts by Vee S.A. Figures were checked against the publishers’ pages on the date shown.
In large organisations, AI meets existing telephony, CRM, collection and policy systems. Most barriers reported by enterprises are about connecting to those systems safely.
Read the insightAutomation is accepted when it solves the issue and when a person is reachable if it does not. Enterprises are redesigning human roles rather than removing them.
Read the insightOrganisations record almost every call, yet analyse only a small fraction. Process data is where measurable improvement comes from.
Read the insightIn the European Union, transparency in interactions with AI is now a legal obligation. Regulated sectors expect controls, documentation and clear disclosure that a customer is speaking with an AI system.
Read the insightBudgets are moving towards AI, but the cost of automated resolution is not falling to zero. Buyers increasingly ask what a process achieves, not how many minutes it uses.
Read the insight