In mid-July 2026, a hands-on Smart Q training session took place at Shenzhen Environment Water Group — one of China's largest state-owned water utilities, serving over 30 million residents. The workshop brought together operations teams and Quick BI product experts for an intensive, on-site workshop designed to show non-technical staff how to query, interpret, and report on operational data using nothing but natural language.
The premise was straightforward: if the people who manage water supply, treatment, and distribution every day could simply ask their data questions the way they ask a colleague, the entire analytics bottleneck disappears. No SQL. No waiting for the IT team to build a dashboard. No exporting to Excel and crossing your fingers.
The Host: Shenzhen Environment Water Group
Shenzhen Environment Water Group is a state-owned enterprise responsible for water supply, sewage treatment, and environmental water management for one of China's most populous and economically vital cities. With infrastructure spanning reservoirs, treatment plants, pumping stations, and distribution networks, the group handles an enormous volume of operational data every single day — flow rates, pressure readings, water quality metrics, maintenance logs, energy consumption, and more.
Like many large utilities, Shenzhen Water had invested heavily in data infrastructure over the years. Dashboards existed. Reports were generated. But the people who needed answers most — field engineers, regional managers, operations leads — still depended on a small analytics team to translate their questions into queries. That translation process took time, and in a utility where decisions about water pressure or treatment schedules can affect millions, time is not a luxury anyone can afford.
This is precisely the gap that Smart Q was built to close.
Four Capabilities, One Conversation
During the training, the Quick BI team walked participants through the four core modules of Smart Q — each designed to handle a different stage of the data-to-decision pipeline. Together, they form a complete analytics workflow that any operations professional can drive without writing a single line of code.
| Capability | What It Does | Problem It Solves |
|---|---|---|
| Q Chat | Ask data questions in plain language, get instant charts and tables | Eliminates the "submit a request, wait for IT" cycle |
| Intelligent Interpretation | Auto-analyzes data for anomalies, trends, and root causes | Replaces hours of manual number-crunching with AI-driven insight |
| Q Report | Generates structured analytical reports from multiple data sources | Turns scattered data into a polished, decision-ready document |
| Q Dashboard | Builds and beautifies dashboards through natural language commands | Lets anyone create professional visualizations without design skills |
Q Chat is the entry point — and for many participants, the most eye-opening moment of the entire session. Think of it as a conversational interface sitting on top of your existing datasets. You type a question like "What was the average daily water consumption in Futian District last month?" and Smart Q translates your intent into the appropriate SQL query, pulls the data, and renders a visualization — all in seconds. Under the hood, this is powered by an NL2SQL engine that has been fine-tuned on enterprise data patterns, achieving a reported 96.5% query accuracy rate according to industry evaluations.
Intelligent Interpretation takes over once you have data in front of you. Instead of staring at a table trying to spot what matters, you can ask Smart Q to interpret it. The system identifies anomalies (a sudden 20% drop in water pressure at one station), detects trends (rising consumption in a developing district), and performs root-cause analysis — breaking down which factors contributed most to a given change. For a utility managing hundreds of monitoring points, this is the difference between reactive firefighting and proactive management.
Q Report solves a different problem entirely: the weekly or monthly reporting grind. Operations teams at water utilities typically spend significant time compiling data from multiple systems, formatting it into spreadsheets, adding commentary, and distributing it to leadership. Q Report automates this entire pipeline. Feed it your datasets — whether from Quick BI dashboards, uploaded Excel files, or connected databases — and it produces a structured analytical report complete with charts, data summaries, and narrative insights. The training showed how a monthly operational summary that previously took a full day to assemble could be generated in minutes.
Q Dashboard rounds out the toolkit by letting users build and refine visual dashboards through conversation. "Add a year-over-year comparison for this metric." "Change the layout to show all three regions side by side." "Apply the company's color scheme." Each instruction is executed immediately, removing the dependency on BI specialists for every visual tweak.
The Training in Action: From Spreadsheets to Conversations
The most compelling part of the session was the live demonstration, where Shenzhen Water's own operational data was used as the test ground. The Quick BI team set up a realistic scenario that mirrored the daily challenges of a regional operations manager.
The scenario: A regional manager arrives at work on Monday morning and needs to understand what happened over the weekend. Water pressure readings from three pumping stations look unusual. She needs to know which stations were affected, whether there's a pattern, what might have caused it, and whether she needs to escalate.
In the old workflow, this would mean opening multiple dashboards, cross-referencing data from different time periods, perhaps requesting a custom report from the analytics team, and waiting — potentially hours — before having a complete picture. With Smart Q, the entire investigation happened in a single conversational thread.
She started with Q Chat: "Show me water pressure readings from all pumping stations over the weekend, compared to the same period last month." Within seconds, a multi-line chart appeared, clearly showing that two stations had experienced abnormal drops between 2:00 AM and 5:00 AM on Saturday.
Next, she used Intelligent Interpretation: "What could have caused the pressure drops at these two stations?" Smart Q performed a contribution analysis, identifying that the most significant correlated factor was a simultaneous spike in water demand from an adjacent industrial zone — likely an unplanned production surge.
Then came Q Report: "Generate a weekend operations summary highlighting these pressure anomalies and their likely cause." In under a minute, Smart Q produced a formatted report with the pressure trend charts, the root-cause analysis, and a recommended action item.
Finally, Q Dashboard: "Create a real-time monitoring panel that shows current pressure at all three stations with alerts if any reading deviates more than 15% from the 30-day average." A clean, color-coded dashboard appeared — ready to be shared with the entire regional team.
The room went quiet for a moment. Then someone said, "We could have used this last week when the treatment plant alarm went off at 3 AM and nobody could figure out why until Tuesday."
Setting the Standard: Data Querying Best Practices
Beyond the product demonstration, the training also established a set of data querying standards tailored to Shenzhen Water's operational context. The Quick BI team worked with the utility's data architects to define consistent naming conventions for datasets, standardize metric definitions across departments, and configure role-based access controls — ensuring that when a field engineer asks about "water quality in Bao'an District," the system pulls from the same certified dataset that the compliance team uses.
This standardization layer is often overlooked in analytics deployments, but it is critical in a utility environment. Inconsistent metric definitions — one department calculating "water loss rate" differently from another — can lead to conflicting reports and, worse, conflicting decisions. Smart Q's Knowledge Base Management feature allows administrators to codify these definitions, so every natural-language query resolves to the same authoritative calculation.
The team also configured permission tiers: field operators can query operational metrics for their assigned region, regional managers can access cross-district comparisons, and senior leadership can view enterprise-wide aggregates — all without exposing sensitive data beyond their authorization level.
A Vision for Automated Water Utility Analytics
The training concluded with a forward-looking discussion about where this technology is heading. The Quick BI product team shared their roadmap for deeper integration with utility-specific data models — including predictive analytics for demand forecasting, automated anomaly detection that triggers alerts before a human even thinks to ask, and scheduled report generation that lands in leadership inboxes every Monday morning without anyone lifting a finger.
For Shenzhen Water, the immediate goal is broader adoption: getting more operations staff comfortable with conversational analytics so that data-driven decision-making becomes the default rather than the exception. The longer-term vision is a fully automated analytics layer where Smart Q doesn't just answer questions — it proactively surfaces insights that no one thought to ask about.
Quick BI Smart Q has been recognized as the only Chinese vendor in the Gartner Magic Quadrant for Analytics and Business Intelligence Platforms for seven consecutive years, and was among the first products to pass the China Academy of Information and Communications Technology's specialized evaluation for LLM-driven intelligent data analysis tools. With its "dual-engine" architecture — combining a foundational LLM with a BI-domain-specific model — Smart Q is positioned not as a chatbot bolted onto a dashboard, but as a purpose-built analytics agent that understands enterprise data the way a senior analyst would.
Ready to Let Your Data Speak?
- Start with a free trial: Visit the Quick BI product page to activate Smart Q on your existing datasets — no infrastructure changes required.
- Request a hands-on workshop: Contact the Lingyang team to arrange an on-site or virtual training session tailored to your industry and data environment.
The era of "submit a data request and wait" is ending. The era of "just ask" is already here.
Quick BI Smart Q — conversational analytics for the enterprise. Learn more at alibabacloud.com.
Top comments (1)
Great case study on how conversational analytics is transforming utility operations. The real-time pressure anomaly detection scenario really shows the practical value of NL2SQL in critical infrastructure. Excited to see more utilities adopting this approach!