Português
EnglishIn the retail industry, whether online or offline, every major promotion is a full-company operation. Customer service must hold the front line; logistics must secure the back end; and operations must ensure that livestreams proceed as planned. Amid the chaos, the most vulnerable link is often the efficiency and accuracy of information flow.
How can you handle several times the workload in a limited amount of time?
Internet brand ZHR Zeze and lifestyle retailer 55°N FIFTYFIVE share a common goal: to free up time from repetitive tasks and focus energy on what truly matters.
They chose DingTalk AI Tables to reorganize their respective business processes. What’s truly impressive is not the tool itself but their determination and actions to "catch" their business with technology before peak traffic arrives.
ZHR Zeze: Turning customer service reps from "mules" into problem solvers
ZHR Zeze is an internet-based brand specializing in women's casual shoes, operating across multiple e-commerce platforms. The online sales cycle is fast, and after-sales support is complex.
The head of the customer service team, Jiu’er, not only coordinates customer service across multiple stores but also acts as a "messenger + dispatcher," accurately conveying customer requests to finance, operations, warehousing, quality inspection, couriers, and other departments and business partners.
During peak pre- and post-sales periods, the biggest fear for customer service is that customer complaints get overlooked—requests are manually sent to different department groups, courier interception relies on manual monitoring, progress tracking depends on follow-ups, and feedback results must be synchronized with customers to form a closed loop. The entire process involves extensive collaboration and follow-up, and during peak business periods, the workload becomes so overwhelming that it feels "never-ending."
With more than a decade of e-commerce experience, Jiu’er used AI Tables to overhaul the process.
Return order management is a core pain point for every e-commerce retail company. Traditionally, a return order involves multiple departments and steps: customer service must query the order system, verify the address, track the logistics status, and notify warehouse staff to receive the package; when warehouse staff receive a customer’s return, they must confirm the logistics tracking number with customer service, determine whether the item can be accepted, and update customer service on the status of the returned item. When issues arise—such as customers returning the wrong item or damaged goods—extensive communication among the warehouse, customer service, and the customer is required to close the loop on an order.
With AI Tables, the process is completely different.
Customer service reps simply enter the courier tracking number into the AI Table. AI fields automatically fetch the latest logistics status—delivered, in transit, or abnormal—for instant visibility. Intercepting a courier no longer requires manual monitoring; the table tracks and evaluates status automatically.
When warehouse staff identify an issue with a returned item, they @ the automation assistant in the group chat, provide the tracking number and the type of rejection, and the information instantly populates the table. The system automatically queries the order and store name, then assigns the task to the appropriate after-sales team. After-sales updates the resolution in the table, and the system automatically syncs the information back to the group chat. Automation replaces much of the manual work, eliminating tedious communication and virtually eliminating the risk of oversight.
Swipe left and right to see how AI Tables manage return orders
Negative feedback closed loop: From "finding the right group" to "automatic task assignment"
The workflow has become simpler, and cross-team collaboration has become more efficient. The most noticeable improvement is in managing negative feedback tasks.
The customer service team at ZHR Zeze needs to relay feedback about issues such as shoes running too large or small, courier damage, or incorrect product images to operations, logistics, and other departments. There are more than 30 different groups across various platforms, stores, and departments, and just "finding the right group" and "following up on results" wastes a significant amount of customer service time.
Now, customer service reps send feedback to the AI assistant, and the information is automatically captured in the AI Table. AI fields automatically classify the issue type, match it with the responsible person, and precisely route the feedback to the appropriate group while sending notifications. Customer service reps no longer have to search through message threads to find the right group. From the moment feedback is generated to the time it reaches the responsible person, a complete closed loop is formed; feedback information that was once scattered across multiple group chats is now centralized in a single table, making it easier to analyze and optimize issues.
Swipe left and right to see how AI Tables manage negative feedback
In the past, information in group chats was cluttered. Each department dumped its data into the group chat: How many return orders does after-sales customer service have? How many returned items did the warehouse receive? As for which after-sales team is responsible for tracking orders from which store, everyone had to sift through massive amounts of data. The whole process could easily consume 3–5 hours per person in the group chat.
Now, after-sales staff can simply open their task view to quickly handle issues. View filtering replaces manual data extraction, and the entire process shrinks from three to five hours to just a few minutes. Even more impressive, the AI fields in the AI Table can automatically determine whether a task is overdue based on the feedback time and requirements, eliminating the need for manual checks and ensuring that daily rejection information is handled promptly.
The dashboard displays the number of tasks, processing progress, and overdue alerts in real time, providing a visual overview of all information. You can instantly see how many issues remain unresolved and how much time is left. This also solves the previous problem where customer service couldn’t follow up on feedback issues in a timely manner.
During promotional periods, the customer service team significantly reduced unnecessary communication and follow-ups. The rate of missed tasks dropped to nearly zero, return processing efficiency improved by 40%, task delay rates fell by 70%, task assignment efficiency increased by 80%, and overall feedback efficiency improved by more than 70%. These aren’t just numbers—they ensure that during peak traffic, the same-sized team can handle far more customer service and after-sales issues.
"By reducing unnecessary communication and follow-ups, AI makes our team work more comprehensively and with higher standards," says Jiu’er, head of customer service at ZHR Zeze. "We can finally spend our time on what matters most."
Livestream efficiency: From 2 hours to 10 minutes
This improvement in efficiency has been noticed by other teams, such as the operations team, which needs to prepare livestreams every day.
Previously, the process was lengthy: Before each livestream, operations staff had to manually prepare scripts, organize product benefit points, and paste dozens of product details one by one.
During peak traffic, when livestream frequency doubles, one person may need to prepare several livestreams in a single day. Missing even a single product can cause problems in the livestream room.
After handing this process over to AI Tables, operations staff only need to enter the product ID, influencer details, and timing, and upload images highlighting product benefits. AI automatically recognizes the product ID, extracts selling points from the images, and generates livestream scripts. These scripts are synced directly with the customer service team, eliminating the need for manual editing. What once took 2 hours to prepare now takes just 10 minutes.
For ZHR Zeze, this change isn’t just about efficiency—it makes every detail more controllable during peak periods, allowing the company to host more livestreams with more influencers without delays caused by preparation work. By the time users watch the livestream, backend efficiency issues have already been resolved, and customer service is fully prepared to handle inquiries.
55°N FIFTYFIVE: A single table powers the logistics system for more than 30 stores
This lifestyle brand, which carries over 25,000 SKUs, 55°N FIFTYFIVE, expanded from its founding to more than 30 stores in less than three years.
This rapid business expansion has placed considerable pressure and challenges on warehouse and logistics management.
For logistics manager Liu Fengshou, the wide variety of products, the long logistics chain, and the growing number of stores and shipping frequency have made inventory counting the most time-consuming part of his job:
Employees must take photos of product labels, manually enter production dates, and calculate shelf life. When dealing with product bundles that combine multiple items, calculating the final shelf life requires repeated calculations.
With hundreds of entries to process in a single day, the error rate is high.
Losses and discrepancies between stores and warehouses are also common. Damage, shortages, and wrong items occur several times a month, and determining responsibility can take days as teams exchange screenshots, compare Excel files, and conduct manual reviews. Adding to the challenge, logistics and warehouse work is inherently difficult to quantify, and employee performance is often evaluated based on subjective judgment, making fair and consistent assessments challenging. As a startup, choosing the right management tools is also a dilemma: the system must be systematic yet lightweight and accessible.
That’s when Liu Fengshou turned to AI Tables.
He used AI Tables to build a "system" that covers the entire process—from production scheduling and shipping to in-transit tracking, delivery confirmation, claims, and reconciliation.
The next month’s shipping plan, estimated loading time, number of shipments, and cargo volume are all recorded in this table. To accommodate different scenarios, he designed multiple views: a detailed view for stores to check all products during inventory, a Gantt chart to show shipping frequency and progress, and a calendar view that clearly marks daily pickup schedules. Different permissions are assigned to different logistics companies and business departments, so each role only sees the information relevant to them.
Swipe to see the full logistics management dashboard
In the inventory-counting phase, AI becomes a true productivity booster. Warehouse staff take photos of product labels, and AI automatically identifies the production date and shelf life, calculates the expiration time, and generates alerts. Inventory time is drastically reduced, and the error rate drops accordingly.
Automated handling of losses, discrepancies, and reconciliation
Handling losses and discrepancies has become an automated process. When store staff discover damage or missing items, they upload the delivery confirmation directly to the table. AI identifies the time on the delivery receipt and automatically determines whether the logistics shipment was delayed. If there is a delay, the system automatically triggers the claims process. What once took days to resolve now takes just minutes. All loss and discrepancy information is automatically aggregated in the table—no detail is missed.
The most troublesome reconciliation process has also been solved. Every month, Liu Fengshou reconciles accounts with more than a dozen logistics companies. Now, by setting up a "print view," he can filter the data and generate a reconciliation statement in seconds—the mall name, outbound order number, logistics tracking number, cargo volume, and settlement fees are all included. Export the data as a PDF, add a stamp, and issue an invoice—the entire process is simplified many times over.
Let the data speak: Make services measurable and improvable
Liu Fengshou believes the real value lies in data. A dashboard displays quantitative metrics such as on-time delivery rate and damage rate, and uses a scoring system to present service quality. More importantly, he incorporates real feedback from end-store employees into the system, with monthly employee feedback surveys. This way, logistics companies can see not only their own scores but also those of their competitors. Naturally, they are motivated to improve—to boost on-time delivery, reduce damage, and enhance service quality.
When rolling out this system, Liu Fengshou didn’t rush. In the first month, he entered all the data himself to help employees get familiar with the system. In the second month, he began "delegating" authority, letting employees fill in some basic information. It wasn’t until the third month that he started teaching advanced features like permission settings. Employees were quickest to embrace the most practical features—for example, handling damage reports. Previously, they had to scour a dozen group chats to find the right tracking number; now, all related information is available in the table, and the number can be found instantly. Work efficiency has indeed improved significantly.
In just three months, Liu Fengshou transformed a logistics department that relied on experience into one that operates on data, where everyone’s contributions are visible. A startup, with minimal investment, has achieved what would normally require a dedicated, professional system.
In his words, AI Tables "make repetitive work simple, simple work automatic, and automatic work accurate."
Conclusion: AI Tables are frontline teams' most reliable allies
The practices of ZHR Zeze and 55°N FIFTYFIVE show that AI Tables don’t bring about a revolutionary transformation; instead, they centralize fragmented information, automate processes that should be automated, and make tasks that need to be tracked visible.
It may seem simple, but the impact is profound: fewer repeated queries, less manual coordination, and fewer information gaps. During peak traffic, these improvements directly translate into greater capacity—when orders double and stores multiply, the same team can handle more business.
If a single环节 goes wrong, you could lose customers. These two companies have refined every环节, freeing up individual performance and boosting the overall capacity of their teams.
An AI Table that understands the business, supports collaboration, and enables closed-loop management can become frontline teams' most reliable ally—keeping pace during peaks, maintaining standards amid chaos, and driving efficiency during growth.
An AI Table that understands the business, supports collaboration, and enables closed-loop management can become frontline teams' most reliable ally:
Keeping pace during peaks,
Maintaining standards amid chaos,
Driving efficiency during growth.
This is what businesses really need from AI.
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, you can contact our online customer service directly, or call +852 95970612, or email cs@dingtalk-macau.com. We have an excellent development and operations team with rich market service experience, ready to provide you with professional DingTalk solutions and services!