The vendor presents these as examples of enterprises applying AI to everyday work; the reported functions and outcomes have not been independently verified.
In recent years, AI products have emerged one after another. However, for enterprises, only practical and implementable AI solutions can create real value.
The vendor describes Feishu AI deployments in R&D, production, sales and service. Whether they improve progress, quality or response efficiency should be judged against each organisation’s baseline and pilot results.
From Yangtze River Pharmaceutical's target research to NIO's production anomaly handling, and from PROYA's new product research to XPeng's intelligent meetings, with Feishu, more companies report applying AI in ways that, in their own accounts, have meaningfully changed how they work.
Yangtze River Pharmaceutical
AI + R&D
Obtain comprehensive data insights with just one command
Target research is a critical starting point in pharmaceutical R&D. In the past, Yangtze River Pharmaceutical’s scientific teams had to sift through vast amounts of literature and databases—any oversight could delay the pace of new drug development.
Now, leveraging Feishu AI Assistant and AI capabilities within Feishu Multidimensional Tables, Yangtze River has built an AI-powered target research bot. With just a simple instruction, AI extracts key information from a curated set of sources, completing searches, table creation, and chart generation. Researchers can open a document for consolidated data views; whether this materially improves R&D efficiency should be measured against a research team’s own baseline.
NIO
AI + Production
Real-time data aggregation and tracking of abnormal events
On NIO’s production lines, filtering alarm messages used to be a pain point for team leaders. By the team's own account, they previously spent up to half an hour each day searching through Feishu groups just to get started.
Now, powered by the AI features in Feishu Multidimensional Tables, the system is designed to capture group messages, aggregate data in real time, identify vehicle models and processes with one click, and deliver scheduled visual dashboards—capabilities a buyer should confirm in a plant trial before rollout. How many man-hours alarm filtering actually saves is worth measuring against a plant's own baseline first.
Dongming Petrochemical
AI + Inspections
Automated follow-up and closure to enhance workshop safety management
At Dongming Petrochemical, identifying potential hazards during inspections was once a major challenge. Complex production conditions and varying technician skill levels often led to missed issues, while compiling safety case reports sometimes took nearly half an hour.
According to the vendor case, Feishu Multidimensional Tables digitised inspection records, routed issues to responsible staff and used AI to assist follow-up and photo recognition; a plant should verify recognition accuracy, routing speed and safety-management impact in a controlled pilot.
Ecovacs
AI + Customer Voice
AI-assisted labelling of new reviews
Previously, Ecovacs relied on manual labeling for large volumes of overseas user reviews and fault descriptions, overwhelming its customer service team.
Now, Feishu offers a new approach. By importing reviews into Feishu Multidimensional Tables, AI can handle translation and summarization of review batches, automatically label new comments, and assign service tickets. It can also convert users’ natural-language descriptions of device malfunctions into technical tags, supporting multilingual labelling for overseas users. Before relying on any of this, a service lead should run a pilot: label a sample of real reviews manually and by AI, and compare the agreement rate against the team’s own baseline.
Meiyijia
AI + Intelligent Customer Service
Automatically breaks down FAQs to shorten response times
With a nationwide store network, Meiyijia’s customer service staff face significant pressure to respond in real time, while complex information also increases the workload for maintaining knowledge bases.
According to the vendor's account, Meiyijia centralised knowledge bases across regional divisions and business modules within roughly two weeks; the implementation period and operating results have not been independently verified. AI automatically parses FAQs (frequently asked questions) and supports intelligent Q&A. Whether response times improve at scale, a retail operator should measure its own average response and resolution rates before and during any pilot.
Eifini
AI + Store Management
Automatically evaluates training quality to improve frontline performance
In the past, Eifini struggled to ensure consistent training effectiveness across its store network due to reliance on manual spot checks, and sales staff couldn’t receive timely feedback or guidance.
The vendor case says store associates use Feishu group video calls and Multidimensional Tables AI to extract keywords from training videos and assist with scoring; buyers should verify extraction accuracy and the review process with their own samples. Meanwhile, Feishu Multidimensional Tables aggregate training data in real time from stores nationwide (per the vendor), giving headquarters consolidated visibility; any effect on frontline sales performance should be verified against store-level sales data during a pilot.
Insta360
AI + Channels
Intelligently process partner emails to identify potential business opportunities
Previously, Insta360 required dedicated personnel to monitor channel partnership emails—an expensive approach that often resulted in delayed responses, negatively impacting collaboration.
The vendor case says Insta360 syncs email data into Feishu Multidimensional Tables, where AI fields assist with content analysis, initial screening and task assignment. The case also reports substantial monthly time savings, but the figure has not been independently verified and should not be treated as a forecast for another organisation.
PROYA
AI + Marketing
One-click extraction of selling points for effortless new product insights
PROYA often faced challenges with scattered information and inefficient knowledge accumulation during new product research. Traditional documents and spreadsheets made quick searches difficult and hindered easy sharing.
Using Feishu Multidimensional Tables, the team built a “New Product Insight Dashboard.” After entering product details, they can filter by category or keyword and use AI to instantly extract selling points and generate word clouds for intuitive trend analysis. Tasks that previously took half a day may now be completed faster, with insights preserved as shared data assets—teams should time their own baseline first.
Anker Innovations
AI + Business Management
One table to track all strategic opportunities
Anker Innovations’ strategic opportunity tracking often involves hundreds of topics, requiring extensive statistics, summaries, and meeting discussions. Tracking efficiency was challenging due to diverse business lines and cross-regional collaboration.
The vendor case says Anker uses Feishu AI Assistant and Multidimensional Tables to organise strategic-opportunity progress for responsible owners to confirm. Buyers should verify the available integrations and baseline current reporting hours before measuring any saving in alignment effort.
XPeng Motors
AI + Meetings
Automatically generated meeting minutes for faster follow-up
At XPeng Motors’s scale, meeting volume had grown so large that reducing it seemed as difficult as redesigning a car model.
In this context, XPeng uses Feishu AI to generate meeting minutes and push action items to owners automatically. For any organisation considering the same step, the useful question is whether action-item follow-through and decision turnaround actually improve once minutes and tasks are generated automatically—metrics worth tracking from the first pilot month.
Click here to read more and contact us to customize a free productivity solution for your enterprise, making AI a productivity tool right at your fingertips.