A meeting that could reshape the automotive aftermarket quietly concluded recently. On September 22, the "First Automotive Data Alliance Conference," themed "Digital Intelligence Empowering the Aftermarket, Ecological Symbiosis for Win-Win Outcomes," was successfully held in Yangshuo. The event was hosted by the Automotive Data Alliance Special Working Group of the Shenzhen Blockchain Technology Application Association, co-organized by Shenzhen Mingrui Data Technology Co., Ltd., and supported as a core unit by Launch Tech Co., Ltd. (02488), a company listed on the Hong Kong Stock Exchange. The conference brought together 25 leading enterprises across the industry chain, including Mengshi, Wanji, TecAlliance, F6, Chaboshi, Ningmengcha, Digital Auto Cloud, CATARC, and Shenzhen Longgang District Data Group Co., Ltd. These companies span core sectors such as OEMs, diagnostic equipment, repair chains, parts data, used-car inspection, insurance risk control, and government data groups—covering virtually every key node in the automotive aftermarket data value chain. These 25 enterprises are the first founding members of the Automotive Data Alliance.
As a representative of the association's chairman unit, Liu Yizhi, Chairman of Launch Tech, welcomed industry chain peers from across the country in his speech and highlighted the core proposition of the meeting: breaking down silos through data interoperability and building the industry's future through alliance co-construction. He stated bluntly that the industry's only "path to breakthrough" is to develop collectively and embrace open symbiosis. Liu Yizhi said that the integration of automotive data and communication technology will realize commercial value through blockchain models—this is precisely the original intention behind Launch Tech and the association jointly initiating the Automotive Data Alliance: using data interoperability to break down silos, co-building standards to regulate circulation, using compliant evidence storage to safeguard applications, and leveraging AI capabilities to unlock data value. "Data is the fuel for AI, and the alliance is the gas station. Once everything starts here, there will be no more fighting alone," Liu Yizhi said. He also extended an invitation to the entire industry: join the alliance, work together to build industry standards, share technological achievements, and jointly explore new AI frontiers.
The successful hosting of this conference carries significance beyond merely "establishing an alliance"—it simultaneously set in motion three things related to the elementization of automotive aftermarket data: defining standards through a white paper, defining products through edge-side models, and defining profit-sharing and compliance through blockchain plus trusted spaces. This provides a replicable set of industry infrastructure for the automotive aftermarket as it transitions from the work-order era into the model subscription era, and opens a new phase of accelerated release of intrinsic value for automotive data enterprises.
Automotive Data Accounts for About 5% of Total Data, with Enormous Untapped Circulation Value Potential
Against the backdrop of China's accelerated push to build a nationally integrated data market, automotive data has become a scarce "high-value mineral deposit" among new production factors, and its value foundation lies in the massive scale of the industry. From a macro perspective, the total output value of China's entire automotive industry chain in 2023 was approximately 11 trillion yuan, surpassing real estate for the first time to become the country's largest economic pillar, accounting for nearly 10% of GDP. Such a massive industrial scale naturally breeds abundant data resources. This is evident from the data announced by CATARC at this conference: among all types of data nationwide, transportation data accounts for 8%, of which automotive data accounts for over 70%, corresponding to automotive data representing approximately 5% of total social data. This 5% data share harbors enormous potential for circulation value.
At the conference, Launch Tech used its own development as an example to demonstrate the potential of macro-level data circulation, which is being accelerated through micro-level scenarios. As a global leader in automotive diagnostic equipment, Launch Tech's hardware devices cover 234 countries and regions, with nearly 3.9 million active units, cumulative connections to over 420 million vehicles, and 1.2 million daily diagnostic reports. This massive volume of diagnostic data has been deeply integrated into various links of the industry chain, including R&D, repair, insurance, and used cars, serving as a vivid footnote to the transformation of automotive data from "resource" to "value." Specifically, on the R&D and quality side, Launch Tech uses the ECU-FDI fault density index to cross-tabulate by vehicle series, vehicle age, mileage, region, and climate, obtaining detailed fault data for different vehicle series. This can feed back to OEMs to help them use data to identify common faults in specific models or systems and improve product design and warranty strategies. On the repair and parts side, Launch Tech uses global diagnostic data to form a closed loop of "diagnostic report → fault code → repair solution → parts SKU." Repair shops can stock parts directly based on diagnostic conclusions rather than relying on experience-based blind procurement, which can directly reduce inventory costs for repair shops. On the insurance and anti-fraud side, Launch Tech's automotive diagnostic and repair data can build a multi-dimensional verification and early-warning system. By relying on historical fault codes and repair records to identify deliberately staged accidents for insurance fraud, risk control shifts from post-event to pre-event. At the same time, it can integrate accident and modification data and connect to theft databases to verify a vehicle's true identity. On the used-car and finance side, Launch Tech's data allows vehicle condition, mileage, accidents, modifications, and original parts records to be directly anchored to residual value. Maintenance time-space trajectories support dynamic vehicle locating for loan vehicles and long-cycle traceability, strengthening the defense line for post-loan asset preservation. The monetization of data in the aforementioned automotive aftermarket scenarios is just one facet of the value of automotive data elements. As CATARC emphasized at the conference, cross-entity data fusion can better support core scenarios such as intelligent connected vehicles and vehicle-road collaboration. The accelerated development of these scenarios will further release the value of data assets and drive the formation of new quality productive forces. In the autonomous driving field, perception and decision-making data flowing back from vehicle operations is becoming a core driver of algorithm iteration. In the vehicle-road collaboration field, the fusion of vehicle-side and roadside data enables the transportation system to move from "single-point intelligence" to "global optimization." In the energy interconnection field, charging, discharging, and battery data from new energy vehicles are supporting vehicle-grid interaction and virtual power plant scheduling. From aftermarket repair, insurance, used cars, and finance to front-end autonomous driving training and vehicle-road collaborative scheduling, the application scenarios of automotive data already cover the entire industry chain, forming a complete value map. The circulation radius of data directly determines its value ceiling—the more data circulates, the more its value multiplies. This is a "rich mine" that urgently needs to be jointly developed by the entire industry.
Industry Pain Points Highlight the Inevitability of Establishing an Automotive Data Alliance
Although automotive data resources harbor enormous potential for circulation value, for the past two decades, automotive aftermarket data has long slumbered in enterprises' internal BI reports, playing a supporting role in "assisting decision-making." Repair shop work orders, diagnostic equipment fault codes, insurance claim records, used-car inspection reports—these data have each fought their own battles and never truly entered cross-entity circulation, with the compound value of data locked behind firewalls. The aftereffects of this "data slumber" have been sharply magnified during the industry's transformation period. As the automotive aftermarket bids farewell to traditional extensive operations and enters a new stage driven by data, empowered by AI, and characterized by ecological symbiosis, industry pain points have become increasingly evident: severe data silos, inconsistent standards, fragmented applications, and AI models lacking fuel. These issues have constrained high-quality industry development. Particularly under data silos, AI models trained by individual enterprises have weak generalization ability and are difficult to truly deploy. If this predicament of "data without circulation, models without fuel" cannot be resolved within the window of industry transformation, it will evolve from an internal constraint into a survival crisis for enterprises in the future. More urgently, the window is narrowing. We are currently in a period of technology-driven transformation led by AI, with the wave of large models and edge-side AI sweeping through the automotive industry. Whoever masters high-quality, multi-dimensional data will hold the initiative in next-generation competition. If industry leaders cling to a "data moat" mentality and are content with one-time profits from hardware sales, they will face the risk of being "boiled like a frog in warm water." The only path to breakthrough is proactive self-innovation and open cooperation, channeling enterprise data into the industry foundation to ensure lasting success. This is precisely the core motivation behind the establishment of the Automotive Data Alliance: through alliance mechanisms, data scattered across OEMs, diagnostic providers, repair chains, insurance, used cars, and other links will be aggregated and circulated in a compliant manner, providing sufficient "fuel" for AI models and driving the aftermarket from "each fighting alone" to "ecological symbiosis."
Alliance Produces Edge-Side/Local Small Model Capabilities, Distributing Returns Based on Data Contribution
As establishing an automotive data alliance becomes an inevitable industry trend, a key question emerges: in what form will the alliance be established, and how will it operate? As a global leader in automotive diagnostic equipment and the core supporting unit of this conference, Launch Tech has already made significant progress in data value mining and has provided a model answer to this question through its own practice. The core of this answer was fully presented at this conference—Launch Tech demonstrated its advantages in data scale, compliance layout, and AI analysis capabilities, proving the feasibility of transforming automotive data from "raw material" to commercial value, and detailed the operating model of the "Automotive Data Alliance." Leveraging its hard power of cumulative connections to over 420 million vehicles and 1.2 million daily diagnostic reports, supplemented by 250 million repair work orders in the United States and a compliant layout across three major global data centers, Launch Tech defined the industry's "ECU-FDI fault density index" and released at this conference the "Automotive Electronic Control System Quality White Paper," completed by AI in three days, laying the foundation of trust for the alliance's establishment. Launch Tech proposed that the Automotive Data Alliance is an ecological collaborative organization based on technical trust, not a simple data exchange pool. In terms of member structure, it strictly limits participation to industry leaders, inviting only the top two in each sub-sector. The 25 enterprises attending the inaugural meeting cover OEMs, insurance, parts platforms, and government data groups. At the data access level, it adopts "thematic data spaces" and blockchain technology to achieve "usable but invisible, controllable and measurable," abandoning direct exchange of raw data. At the rights confirmation and profit-sharing level, smart contracts automatically execute and distribute returns based on data contribution. Given the domestic legal framework prohibiting data trading, the alliance processes data into compliant products such as diagnostic reports and risk warnings, ensuring that monetization paths are legal. In terms of organizational mechanisms, starting with the white paper release, it regularly shares PPTs and attendee lists, establishes a senior executive community, and promotes specific cooperation through "throwing out a brick to attract jade plus in-depth discussion." Notably, the core output of the alliance is not raw data but edge-side/local small model capabilities. Edge-side models can generate diagnostic results locally within 15-20 seconds without internet connectivity, balancing speed and privacy, and continuously evolve through data feedback loops. The blockchain profit-sharing system automatically records data sources in model calls, ensuring all parties receive returns based on contribution, allowing OEMs, repair shops, parts suppliers, and others to profit from "slumbering data" and break free from silo dilemmas. As Jiang Shiwen, Senior Vice President of Launch Tech, summarized: "Everyone holds a piece of a treasure map, but without sharing, the treasure will never be found." The establishment of the Automotive Data Alliance marks the industry's transition from each fighting alone to ecological symbiosis. This is not only an upgrade of China's automotive aftermarket business model but also a key step for the industry to seize the initiative in global competition in the AI era. In the future, the alliance will further address potential challenges such as rights confirmation and traceability, privacy protection, anti-monopoly, and enterprise participation motivation, continuing to break through with federated learning, transparent rules, and already-implemented cross-entity collaboration cases, driving industry data from single-point applications to cross-domain integration.
Promoting Digital Transformation of Member Enterprises, Accelerating Revaluation of Data Asset Value
The deeper significance of the data alliance is not just giving enterprises "one more data channel," but reclassifying data from a cost item in internal reports to a circulable, billable, and compoundable asset, thereby changing corporate valuation logic. Using Launch Tech as a sample, its growth path of data value revaluation is already clearly reflected in the company's performance. In the past, Launch Tech focused on diagnostic equipment sales, recognizing revenue upon hardware delivery. However, the equipment lifecycle is typically 3-5 years with long repurchase cycles, making it difficult to generate sustained cash flow from selling equipment alone. But the company's current business structure has clearly changed. In the first half of 2026, Launch Tech's software business revenue was approximately 120 million yuan, up 67.9% year-on-year, accounting for 11% of total revenue. Data business revenue was approximately 15.4 million yuan, up 93.6% year-on-year, the fastest growth among all businesses. Remote diagnostics and other AI service revenue was approximately 16.1 million yuan, up 45.7% year-on-year. What distinguishes these businesses from hardware is that software subscriptions, diagnostic model calls, and data interfaces are all paid periodically. After a device is sold once, it can continue generating revenue from upgrades, queries, reports, and intelligent agent calls, with high-frequency customer usage and predictable enterprise revenue. The alliance's core output products shifting to edge-side/local small models plus subscription licensing is an important measure to upgrade one-time sales into long-term services. What supports the accelerated tilt of Launch Tech's revenue structure toward software, data, and AI services is the completeness of the company's data source structure and dimensions. Launch Tech's diagnostic data runs through the entire vehicle lifecycle—from entry fault codes, repair solutions, and parts replacements to subsequent recalls and used-car residual values, forming a complete data chain. It also covers different brands and new energy vehicle models, enabling comparison of differences across systems. It has also accumulated work orders, climate, and road condition samples from different regions domestically and internationally. Combined with time dimensions such as 1.2 million daily diagnostic reports and nearly 3.9 million active devices, the model can continuously iterate and optimize by vehicle age, mileage, region, and season. The ECU-FDI fault density index and the "Automotive Electronic Control System Quality White Paper" released by Launch Tech at this conference essentially distill such complex data into reusable quality indicators, transforming Launch Tech from "selling equipment with reports attached" to "selling diagnostic capabilities." After the data alliance is established, it will rewrite the traditional ledger of "data for internal use only." In the past, data was reflected in financial statements more as storage and maintenance costs. Under the alliance's blockchain smart contract system, multi-dimensional data contributed by enterprises—diagnostics, work orders, claims, used cars, parts, and more—will automatically distribute returns based on call frequency and model improvement effects, directly converting into model call rights, subscription revenue-sharing rights, and even joint intellectual property rights for industry white papers. Data is no longer a slumbering cost item but an asset that can continuously generate returns. Once this profit-sharing model runs successfully, it will trigger a self-reinforcing growth flywheel: the larger the enterprise's data volume and the more complete its dimensions, the higher the accuracy of the trained edge-side models, the more frequently they are called by repair shops, insurance companies, used-car dealers, and parts platforms, and the more exponentially marginal returns rise. Data then transforms from a "burden" into a "money printer." For enterprises in the alliance, this will usher in a deep restructuring of commercial valuation logic. The data assets of alliance member enterprises will be revalued from "numbers in self-use reports" to "licensable, tradable model assets." On this basis, as the weight of software, AI, and data businesses in financial statements continues to rise, market valuation logic will completely break free from the traditional hardware PE framework and leap to a composite model of "device entry point + subscription revenue + data barriers." Once the compliant profit-sharing mechanism is implemented, the AI output share of leading enterprises will rise rapidly. A wave of data-driven value revaluation will arrive at an accelerated pace.
For more heavyweight Hong Kong stock news, download the Zhitong Finance app. For more Hong Kong stock and overseas wealth management information, please visit www.zhitongcaijing.com (search for "Zhitong Finance"); to join the Zhitong Hong Kong stock investment group, please add Zhitong customer service WeChat (ztcjkf).
Comments