The 45.47-trillion-yuan foreign trade sector still relies on sales personnel to manually source customers, draft business development emails and vet buyers, and Tradewhale aims to integrate this entire business chain into a suite of AI tools.
For small and medium-sized foreign trade enterprises, work including finding customers, writing development emails and judging the credibility of buyers is still largely completed manually by sales staff. Tradewhale, an AI assistant, integrates all these functions into one set of tools. The product has been launched and charges subscription fees per account. According to the team, there are more than 40 stable paying users and 1 enterprise with clear cooperation intention at present. The company is seeking angel round financing or incubator cooperation.
The full-year 2025 data released by the General Administration of Customs of China in January 2026 shows that the total value of China's import and export of goods trade reached 45.47 trillion yuan, a year-on-year increase of 3.8%, hitting a record high; among which exports were 26.99 trillion yuan, a year-on-year increase of 6.1%. There are more than 780,000 business entities with import and export records. According to the "2025 Statistical Communiqué on National Economic and Social Development" of the National Bureau of Statistics, the import and export volume of private enterprises accounted for 57.3% of the total import and export value, with a year-on-year increase of 7.1%. The basic market of exports remains stable, but the customer acquisition methods and experience inheritance of front-line teams have not been optimized along with the expansion of business scale.
The pain points of sales staff are very specific. Target companies can often be found, but the key contacts in charge of procurement and decision-making cannot be located, the contact information and social media access are also unstable, and the outreach process is often interrupted halfway. Even if the list of companies is obtained, there is a lack of sufficient background information to judge the strength and cooperation intention of the targets, resulting in a large number of ineffective follow-ups. Searching for customers and writing development emails take up a lot of working hours. Enterprise owners face another layer of problems: order closing is highly dependent on a small number of senior sales staff, experience is lost with personnel turnover, and it is difficult to build a talent pipeline for new employees; there is no accessible basis in advance for the fluctuation of order volume, as well as the risks of buyer payment and performance, and most problems can only be dealt with after the fluctuations occur.
Existing tools on the market do not cover the whole process of this business chain. Tungee has a large coverage of enterprise data, and its product focus has long been on information query. Xiaoman and Fumang have built a complete foreign trade customer management process. Tradewhale does not compete with them on the scale of database, but entrusts the work of writing emails, judging targets and conducting follow-ups after information query to AI and local Agent, and its products are first sold to domestic teams with actual export performance.
The launched capabilities are divided into four parts. The clue center is used to retrieve self-owned trade data, generate HS and UN Comtrade product reports, and access social media customer acquisition and AI search. Outreach and transformation functions include email marketing, AI-generated development emails, and desktop Agent installed on sales staff's computers to handle repetitive tasks such as searching for clues and writing development emails. Customer management is divided into public clue pool and private library, supporting claiming, grading and follow-up data precipitation. Agent skills are split into 10 links according to foreign trade operations: customer acquisition, outreach and follow-up, customer background check, inquiry transformation, negotiation and quotation, contract terms, logistics and customs declaration, payment collection and bad debt handling, customer retention, and team management, with a total of 49 skill packages.
For sales staff, the search function is used to supplement contact information and social media information, the background check function is used to reduce ineffective follow-ups, and the desktop Agent keeps repetitive operations on the local device. For enterprises, the standard subscription first stores clues and follow-up data in the system; the customized version converts the practices of senior sales staff into reusable skills, and stores core data in the enterprise's private domain according to the requirements of the enterprise. Skills related to credit, sanction lists, payment collection and import cycle correspond to the advance check of buyer risks and order fluctuations, not a guarantee of deal closing. The overseas version is still under planning, and the current products serve Chinese export teams.
The charging model is mainly based on subscription, with enterprise customization as a supplement for higher unit price. The basic membership is 158 yuan per account, with 20 Agent conversation times per day; the premium membership is 365 yuan per account, for core sales staff who need deeper data collection, with 60 Agent conversation times per day. The seat discount is 15% off for 3 to 4 accounts, 25% off for 5 to 9 accounts, 50% off for 10 or more accounts, and an additional 20% off for annual payment. The team said that sales are promoted according to the process of demonstration, trial use and then payment, and real subscription orders have been received so far. This set of figures shows that some people are willing to pay for higher customer acquisition efficiency. The 1 intended enterprise has not signed a contract yet, which cannot be counted as a landed enterprise customer.
When calculating the market space, a commonly used figure cannot be directly applied. According to the World Trade Organization's "Global Trade Outlook and Statistics", the global trade volume of AI-related goods in 2025 increased by 21.9% year-on-year, such goods accounted for about one sixth of global goods trade, and contributed about 40% of the trade increment of that year. This statistical caliber covers AI-related goods such as chips and servers, rather than the proportion of foreign trade companies purchasing AI customer acquisition software. The demand more relevant to Tradewhale comes from more than 780,000 business entities with import and export records, especially small and medium-sized export teams with limited budgets that have to find customers on their own. Recent users are the sales staff and enterprise owners of these teams. Only when the overseas version is launched, the same set of customer acquisition, background check, outreach and Agent capabilities will be provided to more overseas buyers and sellers to reduce language and information asymmetry, rather than the global network that has been connected at the present stage.
Founder Li Jingyue is responsible for technology, product and commercialization. According to the team, he was previously engaged in large model training algorithm development at Cheetah Mobile in the early stage, with educational background in computer science and business English at Oregon State University. He received angel investment from MiraclePlus in 2024, participated in the incubation of Charles Xue Fund in 2025 and worked on AI and blockchain applications, and returned to the AI application scenario to found Tradewhale in 2026. Co-founder Ma Wei is responsible for channels and enterprise customers. Graduated from Southwestern University of Finance and Economics with an accounting major, he has about 11 years of front-line foreign trade experience, once worked at Alibaba International Station and XT Cross-border Payment, and then started his own business. The team size once reached about 50 people, and became a strategic partner of Shell and Deeye. The technical side and channel side are thus separated: one is responsible for model and Agent development, and the other is responsible for connecting the actual process of foreign trade enterprises.
The product derives functions from actual foreign trade operations, rather than developing a general conversation model first and then adding a foreign trade label to it. Before the launch, two main issues were dealt with: how to make the skills fit the scenarios of customer acquisition, email writing, background check, quotation and payment collection, as well as the requirements of enterprise customers for private data. The progress that can be confirmed at the present stage is that the functional version has been monetizable, with more than 40 paying users and 1 additional enterprise expressing cooperation intention. The team also stated that the operating capital is tight, so the angel round and incubation cooperation are put on the current schedule.
Li Jingyue said that the funds will be prioritized for domestic customer acquisition and sales, product and Agent iteration, to expand the already verified paying funnel, and focus on 1 to 2 foreign trade categories first. After building stable paid services in the domestic scenario, the overseas version will be promoted to provide the same set of capabilities to more digital foreign trade teams.
Data sources: The 2025 full-year import and export situation of the General Administration of Customs (press conference of the State Council Information Office on January 14, 2026); the "2025 National Economic and Social Development Statistical Communiqué of the People's Republic of China" of the National Bureau of Statistics; the "Global Trade Outlook and Statistics" (March 2026) of the World Trade Organization. The number of paying users, pricing, skill modules and team resumes are from the product page of Tradewhale and its angel round materials, and the resumes are stated by the team itself.