Português
EnglishDingTalk AI Tables Launch First "10-Million Hot Rows in a Single Table"
By Shen Songnan
The arrival of the 17th Double 11 has once again thrust the e-commerce industry into its annual high-intensity operational cycle. From marketing promotions and livestream operations to warehousing, logistics, customer service, and after-sales support, every link is tightly interwoven, with data generation and real-time data-driven demands growing exponentially.
In this context, multi-threaded operations have become the norm for merchants. A brand representative recently told Tianxia Wangshang that their birthday coincides with Double 11: "For the past few years, I haven't really celebrated my birthday—everyone at the event is busy except for the birthday person."
Behind the year-on-year growth in promotional sales and the continuous development of brands are both consumer demand itself and the evolution of supporting tools. On November 5, DingTalk AI Tables launched the industry's first technology capable of handling 10 million hot rows in a single table, enabling enterprises to aggregate, compute, and drive massive amounts of data in real time within a single sheet—eliminating the previous need to manually split and merge multiple tables due to excessive data volumes. This capability perfectly aligns with the e-commerce industry's urgent need for real-time data processing during the Double 11 period.
In fact, prior to this, Tianxia Wangshang had observed that companies such as Semir, Intime Department Store, and designer brand Almond Rocks had already begun using "a single table" as the core medium for their daily operations.
Semir: From Manual Data Entry to Real-Time Data-Driven Operations
In the apparel sector, capturing and responding to consumer feedback is directly tied to identifying hit products and mitigating inventory risks. As a highly emotional and fashion-oriented category, consumer reviews, product showcases, and even subtle emotional shifts reflect the most authentic signals of fashion trends—especially during promotional periods, when massive volumes of inquiries and feedback surge like a tidal wave, putting businesses to the test.
Lv Wanlong, a customer service supervisor at Semir Co.'s customer service center, and his team know all too well the pain of slow data response during Double 11. In the past, the team relied on Excel spreadsheets, requiring manual entry of multidimensional information—a lag that was dramatically amplified during promotional periods. Every round of data feedback required regenerating charts and analyses, then passing the results across departments, making the process slow and cumbersome. Moreover, customer service feedback came from diverse sources, including chat logs, audio recordings, images, and social media content, making it difficult to distill long-term, analyzable data under traditional workflows.
Semir uses DingTalk AI Tables
Today, Semir has built a "real-time data-driven" business mid-platform through deep integration of RPA (robotic process automation) and AI Tables.
First, real-time feedback captures public sentiment across all channels. After automatically collating brand content from platforms like Xiaohongshu, RPA writes the data directly into DingTalk AI Tables and sends real-time notifications. The AI Tables can automatically identify sentiment and precisely categorize issues—whether related to products, logistics, or promotions.
"Say we sell 10,000 units, and 500 customers share their experiences. Over 90% of the feedback includes positive tags like 'beautiful' or 'comfortable.' As soon as this data comes in, we push it in real time to the operations team for decision-making, prompting additional stock orders or adjustments to marketing campaigns," Wanlong explains.
The real-time collection and analysis of user feedback enables Semir to quickly adapt resources amid Double 11's multi-wave sales rhythm, shortening the "user voice–operations decision" loop to the shortest possible duration. After all, as the promotional period stretches out, predicting sales peaks becomes much more challenging. For example, a sudden cold snap in late October, combined with promotional discounts, rapidly boosted winter apparel sales for brands like Semir and Bosideng. Such unexpected events not only create pressure on inventory preparation but also strain pre-sale consultation and after-sales resources at brand flagship stores on Tmall. Now, by combining sales trends, weather data, and other external factors, the AI Tables can predict operational needs in real time, helping teams flexibly adjust staffing across different stages.
For Semir, the value of AI Tables lies in building a "real-time data-driven" business mid-platform that transforms massive, dynamic information into actionable insights under high-concurrency conditions. And the seamless operation of this system relies on DingTalk AI Tables' ability to process millions of interactive data points in seconds.
Tianxia Wangshang believes that behind the "10-million-hot-rows" technology narrative, DingTalk AI Tables hold a deeper significance for the e-commerce industry: "This is undoubtedly the table that understands the e-commerce industry best." China's e-commerce sector is thriving, vast, and complex, with hundreds of industries spanning everything from front-end marketing to back-end inventory and after-sales. The data chains are long, and the data formats are intricate. To "manage" such a complex data network, a platform must have deep roots in the e-commerce industry and a long-term understanding of its dynamics—and this is precisely where DingTalk, born out of Alibaba, holds a distinct advantage in its DNA.
Intime Department Store: Breaking Down Collaboration Silos for Cross-Organizational Efficiency
While Semir has solved the problem of data real-time performance, Intime Department Store is using the same table to tackle the complexity of large-scale, cross-organizational collaboration.
A major retail livestream often involves multiple independent entities, including brands, platforms, shopping malls in different locations, and operations teams. Each party uses different systems, and data formats vary, making it easy for errors to occur during the transmission and alignment of critical elements such as product information, pricing, coupons, and schedules. Even minor changes require repeated cross-organizational confirmations.
According to Li Kai, who is responsible for department store content operations at Intime, this is a "collaboration nightmare."
Recently, Li Kai oversaw a large livestream that covered more than 60 shopping malls, involved over a dozen beauty brands, and featured more than 80 types of coupons. Following traditional practices, he had to set up multiple temporary communication groups to collect and verify information—essentially shuttling back and forth between one data island after another...
Intime uses DingTalk AI Tables
Now, Li Kai uses AI Tables as a unified information hub for all collaborators:
Brands and mall staff update information in the table, and the data is automatically aggregated in real time. Behind this is the AI Tables' "two-way linking" capability: for example, when product information in Table A is modified, all related entries in Table B—such as inventory dashboards or scheduling plans—are automatically updated. This fundamentally addresses a core pain point in modern work collaboration.
Today, with the help of AI Tables, Li Kai has fully automated key work processes. For instance, the AI Tables can automatically filter out prohibited words in product information and add location-specific tags to products for different malls with a single click, ensuring data consistency and standardization from the source. Whenever a livestream ends, the data analytics results provided by the table are pushed in real time to the work group, cutting the original days-long review period down to just minutes.
Li Kai explains that even the action of pushing the review results to relevant business groups is completed "on his behalf" by the AI Tables. In addition, the settlement of livestream hosts' salaries has also been automated—by linking livestream duration with scheduling data, the system automatically generates salary statements each month, making manual reconciliation a thing of the past.
The boost in efficiency is directly reflected in business scale. When Li Kai first joined Intime last year, his team could only support about 20 malls at full capacity. Now, with the standardized processes enabled by the AI Tables, the same number of staff can efficiently manage livestream operations for more than 60 malls nationwide, with plenty of room to spare. "I feel confident handling over 100 malls—I'd say any retailer today simply must start using this table," he says.
Almond Rocks: Streamlining the Full Lifecycle of Influencer Collaboration to Create a Business "Cockpit"
The day before a weekly meeting, Almond Rocks' commerce and operations teams are likely to pull an all-nighter: faced with a list of over 6,000 influencers scattered across different folders, along with logistics tracking numbers for influencer merchandise samples and screenshots of campaign performance, the sheer volume and disarray of the information can be overwhelming for frontline employees.
The information is spread across different Excel spreadsheets, social media chat logs, and emails, and just organizing it into a weekly report feels like draining one's energy.
Almond Rocks started with a pair of socks, aiming to become a Chinese original designer brand that combines quality, design, and affordability. However, the efficiency of its collaborations with external influencers and the brand's own management radius had begun to constrain its further growth. Today, DingTalk AI Tables are helping this brand-centric company achieve greater operational agility.
Almond Rocks' cockpit
Take the "influencer seeding" scenario, for example. Of the more than 6,000 influencers in the company's database, only 4–6 commerce staff were originally responsible for managing them, and most of their time was spent sorting data, creating spreadsheets, and checking information. Now, the commerce team can see all the influencers they need to manage in a single AI Table, and the system sorts the influencers to identify those that require priority attention. Fragmented information such as the number of times each influencer has received samples and their performance outcomes is also presented in a centralized view.
In the livestream scenario, in the past, operations staff had to switch back and forth between 4–5 platforms to check livestream conversion data, manually integrating and cleaning the data, resulting in a fragmented workflow. Now, with an RPA plugin, the AI Tables can automatically integrate conversion data from multiple platforms, creating a unified data dashboard that allows operations staff to intuitively compare this week's sales figures and details with last week's. Within Almond Rocks, this is defined as a "cockpit"—a dashboard that brings together all dimensions of the brand's operations data.
Moreover, Zhang Qi explains that the AI Tables can also help the brand generate heat maps of influencer performance to identify long-term collaborators; dynamically monitor individual product sales data and inventory turnover rates to promptly identify hit products. It is reported that Almond Rocks has recently begun using the tables to track competitors' activities, helping the company stay abreast of external competition...
Zhang Qi tells Tianxia Wangshang that half of Almond Rocks' employees now work on AI Tables. With the promotional season underway, he notes, "For example, if we want to ramp up production of a well-performing product, how many units should we ship and promote this week and next week? In the past, the data was lagging and couldn't keep up with the fast-paced promotional environment (posting after Double 11 would be meaningless). Now, this data flows in real time, making it perfectly suited to the e-commerce industry's need for rapid adjustments."
When more than 6,000 influencers, cross-platform livestream data, and ever-changing inventory information all converge in one place and can flow in real time, the brand's ability to respond to market rhythms truly keeps pace with the speed of e-commerce.
AI Tables: Evolving from Tools to a New Form of Productivity
Semir's "data real-time performance," Intime's "process collaboration," and Almond Rocks' "operational agility"—these three brands' AI initiatives, unfolding from different angles, collectively demonstrate a trend: AI Tables are becoming a new paradigm for supporting e-commerce operations.
It is therefore only natural that this year is being hailed as the "first Double 11 powered by AI Tables."
Tianxia Wangshang believes that behind DingTalk AI Tables' series of applications in complex scenarios such as e-commerce, retail, and large-scale promotions lies, at its core, a closed-loop application of AI capabilities within Alibaba's own ecosystem: At the top level, Alibaba has built a foundational AI large-model technology stack; the technology is embodied in a series of products, including DingTalk AI Tables; and e-commerce businesses—especially the extreme Double 11 scenario, with its multi-threaded, data-driven nature—provide the most intense and immediate feedback environment for testing AI applications.
The successful implementation and flowering of these AI capabilities not only empower the industry but also provide an excellent model for extending these scenarios and business applications. One reasonable speculation is that AI Table capabilities could be further extended to other complex internal business operations, such as Cainiao's logistics network (warehousing, transportation, and delivery) or Hema's retail operations (inventory management and sales tracking)...
Thus, the application of "this one table" during Double 11 has a value that goes far beyond the e-commerce industry and promotional scenarios—it clearly outlines the path by which Alibaba translates its top-level AI technology into new modes of production through specific products within its ecosystem.
In the e-commerce sector—a field in which China is among the world's most mature and leading industries—AI Tables are completing a transition from "tool" to "new form of productivity." This is both a showcase of Alibaba's AI capabilities and a blueprint for its future large-scale AI empowerment of both internal and external ecosystems.
DomTech is DingTalk's officially designated service provider in Macau, specializing in providing DingTalk services to a wide range of customers. If you'd like to learn more about DingTalk platform applications, please contact our online customer service directly, or call +852 95970612, or email us at cs@dingtalk-macau.com. We have an outstanding development and operations team with extensive market service experience, ready to provide you with professional DingTalk solutions and services!