Summary

BMW Group has outlined plans to reshape its vehicle range by region, use AI more widely and reduce management structures. It is targeting an Automotive Segment EBIT margin of 3–5% in 2028, with a return to 8–10% by the start of the next decade.

BMW Group has set out plans to tailor its vehicle range more closely to regional markets, expand the use of artificial intelligence across the company and streamline management. At its 2026 Capital Market Day, the company said the measures are intended to improve resilience and profitability as it faces more challenging market conditions.

BMW is targeting an EBIT margin—the share of revenue represented by earnings before interest and taxes—of 3–5% in its Automotive Segment in 2028. It aims to return to its long-term target range of 8–10% by the start of the next decade. The company also expects Automotive Segment free cash flow to reach at least €7 billion by then. These are company targets, not reported results.

Contents

Profitability targets and a regionalised model range

The product strategy centres on offering models and powertrains suited to local customer preferences, while reducing the number of variants across the portfolio. BMW said some models, including the BMW 2 Series Active Tourer, will not receive successors. It plans to bring the first model under the BMW ALPINA brand to market in 2027, positioning the range between BMW and Rolls-Royce.

Regional plans differ. In China, BMW intends to expand local production in high-volume segments alongside the Neue Klasse and limit imports to higher-margin models. It aims for at least 95% of locally manufactured vehicles there to be developed specifically for Chinese customer preferences by 2030. The company is also considering exporting vehicles made in China to Southeast Asian markets.

For Europe, BMW plans to add a compact fully electric Neue Klasse model in 2028. In the United States, it is considering an additional Sports Activity Vehicle above the current BMW X7. BMW described that model as a possibility, rather than a confirmed launch.

AI moves further into development and production

BMW says AI is already used in many stages of production, including digital factory models, quality inspections and the movement of materials within plants. It plans to extend its use across development, purchasing, sales and aftersales as well.

In vehicle development, BMW expects specialised digital agents to help with processes from technical requirements through testing and release. The company says developers will monitor and review results and make final approvals. One example it cited is work with Mistral on crash simulations, using BMW engineering data to create specialised AI models. BMW says it plans to expand this approach using systems trained on industry-specific engineering and simulation data, which it calls Large Industry Models.

The company also describes the Neue Klasse’s vehicle electronics and software as a basis for an “AI-defined” vehicle, in which functions such as driver assistance, infotainment and comfort systems work together. For 2027, it plans to introduce a driver-assistance system developed with Momenta in China and navigation-guided assistance developed with Qualcomm in Germany. BMW says other countries, including the United States, are expected to follow in stages. Its Highway Assistant remains an SAE Level 2 system, a driver-assistance classification.

Management changes and decisions ahead

BMW said it will reduce the number of divisions and associated management roles by 20% by the middle of 2027, with a comparable reduction at organisational levels below. The company linked the changes to a broader effort to simplify processes and reduce costs; it also said a voluntary severance programme forms part of its personnel restructuring.

The company has begun implementing some measures, while other topics remain under evaluation, with decisions expected by spring 2027. The strategy therefore combines specific planned product and organisational changes with financial goals that BMW intends those measures to support.

Sources