Agent begins to take over cross-border e-commerce operations: A reconstruction of industry entry barriers
```
Author | Huang Yu
"After Accio Work went online in March, I first topped up 6,000 yuan, and recently added another 30,000 yuan annual fee for my team." He Jiakun, an Alibaba International platform merchant born after 2000, told Wallstreetcn that he dared to spend so much on AI because his shop's current 80% growth depends on AI.
In He Jiakun's team, work that previously required dedicated personnel — posting products, weekly data reports, infringement checks, etc. — is now all done with one click by Accio Work, allowing one person to manage 3–4 shops at the same time.
Accio Work is a vertical Agent for the cross-border e-commerce industry launched by Alibaba International.
Wallstreetcn learned from Alibaba International's internal sources that more than a month after Accio Work's launch, the daily token call volume for merchants on Alibaba International doubled.
As AI's "working ability" improves, merchants are ever more willing to pay.
The reason is not complicated. Cross-border e-commerce is itself a highly process-driven, standardized, and data-oriented internet business. From product selection, listing, SEO optimization, to client follow-up, quotations, reviews, logistics coordination, a huge amount of work has clear SOPs (Standard Operating Procedures).
This means it's inherently suitable for AI takeover.
But more worth watching, for the past decade-plus, the core business model of e-commerce platforms was "selling traffic" — merchants buy exposure, ads, search ranking, and transaction opportunities from platforms.
As Agents dive into operational processes, e-commerce platforms are shifting from "traffic distributors" to "providers of AI management systems".
From product selection, operation to customer management, more and more work originally handled by merchant teams is being completed by platform-provided Agents.
In addition, Wallstreetcn found that because B2B cross-border merchants face more complex transactional scenarios, their willingness to pay for Agents like Accio Work is stronger than that of merchants doing C-end (consumer) business.
In recent years, cross-border e-commerce has increasingly turned into a "big company business" — platform rules are ever more complicated, traffic ever more expensive, operations ever more intensive, compressing the survival space for small and medium sellers.
The emergence of Agents seems to be breaking this survival predicament, helping seasoned sellers reduce costs and increase efficiency, while lowering the entry threshold for some rookies.
How AI participates in operations
A mature cross-border e-commerce team often requires collaboration across multiple roles: operations, design, SEO, customer service, ad placement, data analysis, etc. Especially for B2B cross-border trade, with more SKUs, longer inquiry chains, more complex customer needs, the demands for operational capability are higher.
Agents are changing what's originally the most labor-consuming and standardized operational flows in cross-border e-commerce.
He Jiakun feels this directly. In his view, Agents first change "shop operation efficiency".
Previously, from product selection, listing, to keyword optimization, inquiry analysis, client follow-up, infringement checks, and later, data review — many steps relied heavily on manual labor; with Accio Work, a large volume of repetitive work is now automated.
"Before, one operator could only manage one store; now, one person can manage five stores at the same time." He Jiakun told Wallstreetcn. Posting products, weekly reports, and infringement checks all can be done with one click.
Wallstreetcn found that the Accio Work platform has several professional Agents, such as "E-commerce expert", "Online store operation", "One-click dropshipping", etc., and it customizes many dedicated "skill packages" for business fields, covering e-commerce and supply chains, marketing and content creation, research and insights, finance and accounting, and more.
While enhancing efficiency, the more crucial value of Agents is that they now begin to help merchants spot business opportunities.
One of the most core and easy-to-mess-up steps in cross-border e-commerce is product selection.
Li Jiale, born in 2002, has already sold Chinese wedding supplies to the US and Europe through Alibaba International, achieving 30 million yuan in sales in 3 years.
Li Jiale said: "Previously, every new product required investing real money to test the market; now Accio Work can quickly help me with market research and category analysis, saving a large amount on trial-and-error costs."
Making the most of AI also requires skills. Li Jiale doesn’t directly ask AI "which product sells well", but first sets the target market and customer traits, letting AI reverse-deduce the suitable product direction.
For example: "What might be lacking for outdoor wedding users aged 25-35 on the US West Coast, with a unit price under 200 dollars?" "What is the biggest pain point for European wedding companies with minimalist styles?"
What AI provides is often not the popular red-ocean SKUs, but the gaps of demand in segmented markets.
"Previously, these things depended more on experience and intuition; now AI will first scan the market for you," Li Jiale said.
Wang Teng, 21, is a typical "AI native entrepreneur". She started her industrial machinery export business in March 2025. In just over a year, she has made her store run in the industry's top 1%.
AI is involved in almost all aspects of Wang Teng's company's operations. But what truly made her feel "AI understands foreign trade" was the emergence of Accio Work.
Wang Teng told Wallstreetcn that when she previously used general AI for shop analysis, she had to manually copy chat records, upload screenshots, and repeatedly explain what platform and scenario she used.
"Back then, AI felt like someone talking to you through glass. You had to feed it a lot of background before it could understand." Wang Teng said. Accio Work truly understands what screens foreign trade workers are in front of every day, which buttons they click.
At its core, this is also one of the biggest differences between cross-border e-commerce Agents and generic AI.
The real strength of Agents isn’t necessarily just model ability, but who commands the industry's workflow.
For cross-border e-commerce, what’s truly valuable is not a generic AI that can chat, but a management system that knows how merchants operate their stores daily, handle inquiries, place ads, and follow up with clients.
Compared to pure large-model companies, platforms like Alibaba International itself own backend merchant behavior, transaction chains, supply chain coordination, customer inquiries, and operating process data, so they can more easily embed Agents into real business flows.
This means, in the future, competition between Agents may not only be about model ability, but also about platform ecosystem and industrial data capability.
Cross-border E-commerce "One-personization"
Agents are also changing the organizational structure of cross-border e-commerce teams.
In traditional foreign trade teams, the number of clients a salesperson can truly analyze each day is limited.
Li Jiale’s team calculated: In the past, a salesperson could analyze at most 30 clients in depth per day; now, with AI, you can analyze 100 clients concurrently and automatically identify potential high-value orders.
Once, Li Jiale’s team asked Accio Work to analyze a week’s inquiries, and AI directly identified two “about to close” big orders, totaling over $3,000.
"It essentially guides salespeople early on, assists them to evolve mid-term, and reminds them to review later." Li Jiale says that at least 80% of salespeople’s time and energy can be saved.
Moreover, Accio Work also helps merchants further with posting, advertising, promoting stores, and autonomously assists overseas clients in finding Chinese suppliers and initial communications.
Since the start of this year, as products like OpenClaw have exploded, the concept of "One Person Company" (OPC, one-person business) is heating up rapidly among entrepreneurs.
Cross-border e-commerce is becoming one of the industries where this model lands fastest, because Agents are lowering the entry threshold for cross-border e-commerce.
He Jiakun told Wallstreetcn that when he started his company in 2023, even hiring just one employee, he needed at least 100,000 yuan for startup capital in a year.
"Now, with top-ups to Accio Work, it might only take a little over 40,000 yuan a year to start cross-border e-commerce, and you can do much more than before when I spent 100,000 yuan."
After Accio Work went online on Alibaba International, He Jiakun clearly felt more juniors coming to ask him about personal entrepreneurship in cross-border e-commerce.
From actual usage experience, as long as you send Accio Work a creative idea, it can autonomously run through market analysis, product design, store decoration, and product posting.
Overseas users say they could "manually" set up a ready-to-launch online store in just 30 minutes.
According to Wallstreetcn, through Accio Work, He Jiakun also guided a fourth-year student doing operations in his team to build a complete independent site in 3 days — from domain name, templates, product listing, copywriting, multi-language translation to SEO configurations, all accomplished in Accio Work.
He Jiakun revealed that building the independent site consumed only about 2,000 points of Accio Work — something that would cost a professional team around 10,000 yuan.
However, AI lowering the threshold does not mean cross-border e-commerce will become easier to make money.
On the contrary, as more merchants start using Agent-like tools, skills like product analysis, SEO optimization, client follow-up, etc. might rapidly become standardized.
Many operational techniques previously reliant on accumulated experience are being quickly "equalized" by AI.
In a sense, AI is both lowering the industry threshold and possibly further intensifying internal competition.
Once a large number of operational SOPs are taken over by AI, the core of cross-border e-commerce industry competition may revert to supply chains, factories, fulfillment ability, inventory management, and financial strength.
That is, what AI lowers is mostly the "operation layer threshold", not the industrial capability threshold.
Furthermore, not all merchants can immediately turn AI into revenue.
An industry insider told Wallstreetcn that although Agents reduce some operational difficulty, for AI to reliably drive business, you still need to understand platform rules, supply chain rhythms, and client needs.
Some small and medium merchants still run into problems using it — can't write prompts, can't set up workflows, AI outputs unstable results, etc.
In other words, Agents are more like advanced assistants, not fully automatic money-making machines.
Scalable commercial application scenarios
Over the past year, the biggest problem in the Agent industry has always been — user retention and sustained payment.
Many AI products are "lively in trial", but not many users really pay long-term, while cross-border merchants are extremely sensitive to ROI.
As long as AI can improve GMV, reduce staff, increase shop numbers, or lift conversion rates, they're willing to pay continuously.
Speaking of Accio Work expenses, Wang Teng and He Jiakun both revealed their companies pay about $100 a month, within acceptable range.
Wang Teng believes that, jobs which used to need 10 people can now be done by 3 with AI. She achieved about 2 million in revenue in 2025, and sets her 2026 target at 9 million.
He Jiakun says, with AI assistance, a new employee can achieve "monthly sales of 1.6 million yuan" performance within a few months.
The enhancement of individual efficiency makes cross-border e-commerce easier to form a commercial closed loop than many office scenarios, since from product selection, listing, to client analysis, script generation, and data review, nearly every step has clear rules and historical data.
Especially for B2B cross-border trade, the transaction chain is more complex. From inquiry, quotation, contract, logistics to after-sales, many processes are highly structured.
But compared to "AI helping merchants improve efficiency", perhaps more noteworthy is: platforms are further binding merchants' operational processes.
Previously, merchants mainly bought traffic and orders from platforms; in future, platforms may further provide "AI employees" to merchants.
From operations, customer service, to product selection and marketing, more business actions are relying on platform Agents.
This means Agents are not just tools, but may also become new platform infrastructure.
As merchants’ operational processes are more deeply embedded into platform AI systems, the relationship between platform and merchants may be further strengthened.
Of course, AI is still not omnipotent.
In supply chain management, on-site inspections, big client relations, complex customization, local cultural understanding, and key business decisions — AI is still unable to truly replace humans.
But even so, the changes Agents are bringing to cross-border e-commerce are already underway.
Risk Warning and DisclaimerMarkets have risks, invest with caution. This article does not constitute individual investment advice, nor does it take into account individual users' specific investment goals, financial situation, or needs. Users should consider whether any opinions, views, or conclusions in this article fit their own situation. Invest accordingly, at your own risk. ```