McKinsey’s analysis estimates that artificial intelligence could unlock between €280 billion and €700 billion of economic value in Central Europe — roughly 6–15 percent of the region’s total net turnover. The scale of this potential is large, but capturing it requires swift and deliberate action to move beyond pilots and retrofit AI into core operations.
Why the urgency?
Two realities increase the pressure. First, adoption is widespread but impact is not: 88 percent of companies globally have deployed AI in at least one function, yet 94 percent have not achieved a significant EBIT impact (McKinsey, "The state of AI in 2025: Agents, innovation, and transformation", November 2025). Second, Central Europe lags Western Europe by 16 percentage points in enterprise AI adoption, and about 60 percent of the region’s net turnover lies in sectors where scaling AI is particularly difficult.
Where will most value come from?
McKinsey argues that much of the value will come from embedding AI into physical operations — not just layering tools on digital channels. In manufacturing, engineering and construction, consumer goods and retail, AI can improve factory utilization through better production planning, reduce yield loss by refining input materials, and increase sales conversion via faster, more personalized interactions.
The firm highlights two value channels: automation/augmentation/robotization of existing work (which could improve value by 6–9 percent) and new product and business‑model upside from faster innovation and personalization (an additional 3–6 percent). The largest opportunities are in advanced manufacturing, consumer goods and retail, and technology; financial services, engineering and construction, energy, and logistics also contribute materially.
AI capabilities are accelerating
The report quantifies the acceleration: leading large language models improved by roughly two index points per year on the Artificial Analysis Intelligence Index between 2019 and 2022; from 2024 onward gains exceeded 20 points per year, and overall capability now doubles roughly every 12 months. This speed compresses traditional strategic, capital allocation and operating redesign cycles.
Early movers gain practical advantages
Organizations that move early discover where models fail and where they can be trusted, build proprietary data flows for training, redesign human–machine decisioning, and create routines to rapidly update workflows. Delaying adoption risks falling behind in both capability and talent as the technology itself continues to advance.
In white‑collar functions — finance, legal, procurement, customer service — specialized AI agents will handle most structured work while people shift toward orchestration, judgment and exception handling. The authors compare this to radiology circa 2016: computer vision enabled smaller teams to read more scans with higher accuracy and speed. A similar dynamic is emerging across knowledge industries: leaner teams supported by AI achieve higher throughput and fewer errors.
The agentic enterprise
AI systems are evolving from standalone assistants into human‑supervised, specialized agents embedded in workflows. In banking examples, squads of agents collaborate: one handles document ingestion and insight extraction, another crafts credit memos using financial, sector and transaction analysis; additional squads manage checking, contract creation, compliance validation, orchestration and client communications, with human managers overseeing judgment points.
However, adopting tools is not the same as changing how work gets done. In software development, about 90 percent of developers now use AI coding tools, yet only 20–30 percent have altered their way of working accordingly — yielding aggregate productivity improvements of less than 15 percent and indicating implementation shortfalls rather than a lack of technology capability.
From deployment to profit: two necessary moves
Most organizations deploy AI in fragmented ways that do not support end‑to‑end economics. To close the gap between deployment and profit, McKinsey recommends two moves: set the right scope and rewire the organization. Domain‑level starts (sales, customer operations, supply chain, engineering, claims) are large enough to matter financially, coherent enough to redesign end‑to‑end, and can often complete full AI integration within six months.
Rewiring ways of working is essential: redesign processes end‑to‑end, integrate agentic AI into core activities, and align incentives to measurable outcomes. This approach differs materially from launching disconnected pilots.
Quantified potential and sector differences
McKinsey estimates that AI could unlock more than €700 billion of value across the region, with over €280 billion attributable to automation effects. Value will be dispersed across sectors that already form Central Europe’s productive core; the technology sector itself accounts for about 10 percent of total AI value potential, indicating gains extend beyond dominant industries.
Scaling speed varies by sector. Digitally mature sectors (technology, media, telecommunications, healthcare and pharmaceuticals) will move fastest due to richer structured data, higher margins and standardized processes. Asset‑heavy, operationally complex sectors (manufacturing, construction, consumer goods) will scale more gradually — and Central Europe is disproportionately exposed to those slower sectors. About 60 percent of the region’s net turnover is in industries where only 17–18 percent of AI adoption has reached the scaling phase.
Closing this gap requires stronger digital foundations, prioritizing high‑value use cases, and targeted capability building.
High‑value domains and use cases
Based on client experience, the domains where AI can deliver the most value are IT and knowledge management, marketing and sales, service operations, software engineering and product development. These functions have structured workflows and codified data, enabling a clearer link between agent actions and productivity gains.
Top use cases include agentic service desks in IT, deep‑research agents in knowledge management, code generation in software engineering, content and campaign personalization in marketing, and end‑to‑end contact‑center automation in service operations. McKinsey’s client experience suggests software engineering alone can achieve 10–20 percent cost reductions when agents are embedded in workflows rather than layered on top.
How organizations must change
To scale, companies should treat AI integrations like product development: define MVPs with explicit success metrics, build inside real workflows using representative data and users, and iterate rapidly. Governance must establish clear accountability and a single source of truth so decisions are made once and accepted across committees.
The barriers at scale are structural. Many organizations attempt advanced model deployment while keeping legacy governance, siloed data, fragmented tech stacks, and unchanged role definitions — resulting in AI that improves tasks but not overall performance. Sustainable value requires coordinated rewiring across strategy, talent, operating model, technology, data and change management (the six enablers of McKinsey’s Rewired framework).
The three stages to secure value are: strategic alignment (senior leadership sets sequenced priorities and value targets), building internal capabilities (talent, shorter development cycles, modular technology architecture, enterprise‑grade data), and change management and adoption (role redesign, capability programs, KPI tracking and progressive handover to business owners).
Market consequences and the case for action
Market valuations reflect whether firms adapt. The authors cite Duolingo, which lost nearly 80 percent of its market value after investors concluded AI‑native alternatives could replicate its offering (Companies Market Cap, updated April 2026). By contrast, Palantir — which aligned its operating model around AI‑driven decision‑making and data integration — saw its stock rise more than tenfold while revenue grew 70 percent year‑on‑year (Palantir, February 2, 2026). The difference: AI built into the operating model versus AI bolted on too late.
For Central European enterprises in manufacturing, financial services or retail, the threat is slower erosion of competitiveness rather than sudden collapse, but that erosion becomes much harder to reverse once gaps widen.
Conclusion
Central Europe has navigated big structural transitions before and those who moved decisively captured disproportionate value. The AI transition follows a similar pattern but on a compressed timeline. The region’s conditions are favorable — mature solutions, proven delivery models, deeper local talent and a substantial industrial base — but the window to act will not remain open indefinitely. Leaders must prioritize domain‑level bets, embed AI into workflows tied to measurable outcomes, and redesign enterprises to scale if they are to capture the region’s €280–700 billion opportunity.
Authors: Lieven Van der Veken (McKinsey, Lyon), Balázs Czímer (McKinsey, Budapest) and Matyáš Zetek (McKinsey, Prague). The authors thank Dan Svoboda, Eszter Teszárik, Lászlo Horváth, Michal Horáček, Michal Miktus, Nadezhda Kitipova, Rana Hamadeh and Roman Vymazal for their contributions.
