INTRODUCTIONThe software-defined vehicle was sold on a revenue story. The revenue didn’t meet expectations - by far.In this episode of Automotive Storytelling, Stéphane Lagresle sits down with two people who watched that gap open from opposite sides. Konstantin Shirokinskiy is a Partner at Roland Berger in Detroit and Global Co-Head of the firm's Software Technology Cluster, with earlier stops at DaimlerChrysler and Rivian. Prashant Gulati is CEO of SDVerse, the B2B automotive software marketplace founded by General Motors, Magna and Wipro, and previously VP of Strategy at Faraday Future.Together they take apart the story the industry told itself about the software-defined vehicle, and describe what is being built in its place.KEY INSIGHTSThe SDV business case was underwritten by revenue that never arrivedNew architectures were expensive, so the industry justified them with projected subscription income. Consumers wanted the features. They did not want another subscription. When the top line failed to appear, platforms and architectures were cancelled and written down.The bottleneck is organizational, not technicalClean-sheet companies solved a problem large OEMs cannot solve the same way. Konstantin compares it to remodelling a castle while living in it. Reporting lines, processes and decades-old supplier relationships all have to break before the model changes.Consolidating the hardware left the software layer with no supplierCollapsing separate ECUs into a central computer was the part the industry understood. The consequence was not. The algorithms inside those ECUs still need to come from somewhere, and a supply base built to sell hardware was not selling them. The Western cost gap is a standardization gap On a comparable mid-size SUV, Konstantin puts the Chinese cost advantage at around 30 percent. Strip out subsidies, labor and materials and a large unexplained bucket remains. It is scale, earned by customizing only where it matters.KEY TAKEAWAYSUnderwrite software investment on cost and time, not on new consumer revenue. Compressed sourcing cycles and engineering hours not spent rebuilding commodity code are measurable now. Subscription forecasts were not. Decide what actually differentiates, and buy the rest. Insourcing a tire pressure monitoring algorithm wins nothing and consumes the capacity you need elsewhere. Take first principles from Tesla, not the playbook. Several companies started at the same moment with the same clean-sheet, vertically integrated approach and did not survive it.Change the story for each audience, never the substance. Engineers, investors, executives and supply chain need different levels of detail. Selling a two-year roadmap as a shipping product is where trust goes.ABOUT THE GUESTSKonstantin Shirokinskiy is a Partner at Roland Berger and Global Co-Head of the firm's Software Technology Cluster, based in Detroit, advising OEMs, suppliers and investors on software strategy and sourcing. He started out at DaimlerChrysler in Stuttgart and was Senior Director for purchasing and strategy at Rivian before rejoining the firm. He is a co-author of Automotive 4.0.Prashant Gulati is CEO of SDVerse, the automotive industry's first B2B software marketplace, founded in 2024 by General Motors, Magna and Wipro. He was previously VP and Head of Strategy at Faraday Future, where he co-led the SPAC IPO process that listed the company on Nasdaq and raised a billion dollars, and led the launch of one of the industry's first in-vehicle generative AI products.Konstantin's wish for the industry: that people stop saying SDV, the way they stopped saying internet-enabled business.#AutomotiveStorytelling #SoftwareDefinedVehicle #AutomotiveSoftware #SDVerse"Under the Hood: Automotive Storytelling" uncovers the human narratives driving innovation in an industry that impacts lives worldwide. Hosted by Stéphane Lagresle, this podcast explores how the power of storytelling ultimately connects technology to humans.

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Transcript

Related