Intelligent investment research products are evolving from "writing analysis" to "conducting research."

When AI Can Verify 24 Financial Data Points Item by Item, Intelligent Investment Research Finally Gains Its "Fundamental Capabilities"

Intelligent investment research products are evolving from "writing analysis" to "conducting research."

In the past, we judged whether a financial AI product was effective primarily by whether it could quickly generate company profiles, industry trends, and investment logic. However, in professional investment research, fluent language is merely the outermost capability. The true determinant of research quality lies in whether the system can accurately extract financial data, standardize reporting periods and accounting principles, display calculation processes, and allow every key conclusion to be verified.

In a recent cross-product evaluation of conversational intelligent investment research tools, we tested multiple products using the same set of questions, the same information cutoff date, and identical default conversation modes. The evaluation covered ten professional scenarios: financial data extraction, profit quality analysis, peer comparison, banking analysis, company valuation, event-driven research, industrial chain analysis, macro variable transmission, natural language stock selection, and multi-turn error correction.

Scoring considered not only answer completeness but also factual accuracy, source traceability, financial methodology, logical rigor, uncertainty handling, multi-turn memory, and research usability.

Among these, a Ningde Time financial data verification question particularly tested the fundamental capabilities of intelligent investment research products. QVeris Bot delivered a complete response to this question.

Why Is This Question So Challenging?

We asked QVeris Bot to compile six core financial metrics for Ningde Time across 2023, 2024, 2025, and Q1 2026: revenue, net profit attributable to shareholders, non-GAAP net profit attributable to shareholders, net operating cash flow, ending inventory, and ending total assets.

This required the system to process six indicators across four reporting periods, totaling 24 financial data points.

But the task went beyond "finding numbers." It also needed to clarify that annual data represented full-year cumulative values, while Q1 2026 data covered January 1 to March 31, 2026; distinguish between period-based metrics (revenue, profit, cash flow) and point-in-time metrics (inventory, total assets); standardize monetary units; and avoid mixing in TTM (Trailing Twelve Months) data, annualized data, or forecasts.

On this foundation, it also needed to calculate the net profit margin attributable to shareholders for each reporting period, as well as the ratio of net operating cash flow to net profit attributable to shareholders.

After the first round, we further required it to verify each number, unit, and reporting period item by item, list calculation formulas, explain data sources, and cross-verify results across different interfaces.

Compared to general company analysis, this type of question leaves little room for linguistic embellishment. Any period mismatch, unit error, or calculation deviation would undermine the entire financial data sheet's usability.

QVeris Bot Delivered All 24 Data Points

QVeris Bot converted all amounts to billions of RMB and organized them into three levels: income statement, cash flow statement, and balance sheet.

Data showed Ningde Time's revenue was 400.92 billion RMB in 2023, 362.01 billion RMB in 2024, and 423.70 billion RMB in 2025, with Q1 2026 revenue at 129.13 billion RMB.

Profit performance was more pronounced: net profit attributable to shareholders reached 44.12 billion RMB in 2023, 50.75 billion RMB in 2024, and 72.20 billion RMB in 2025, with Q1 2026 reaching 20.74 billion RMB.

Corresponding non-GAAP net profit attributable to shareholders was 40.09 billion RMB, 44.99 billion RMB, 64.51 billion RMB, and 18.09 billion RMB, respectively.

More valuable than simply listing numbers, QVeris Bot calculated the net profit margin attributable to shareholders using a unified formula:

Net profit margin attributable to shareholders = Net profit attributable to shareholders ÷ Revenue × 100%

Results showed Ningde Time's net profit margin attributable to shareholders rose from 11.00% in 2023 to 14.02% in 2024, further reaching 17.04% in 2025, and 16.06% in Q1 2026.

This data clearly reveals a key trend: while revenue fluctuated, Ningde Time's profitability continued to strengthen, with profit growth significantly outpacing revenue growth.

Cash Flow Adds Dimension to Profit Quality

Whether a manufacturing company's profits represent high quality cannot be judged solely by the income statement; it also requires verifying whether profits convert to operating cash flow.

Ningde Time's net operating cash flow was 92.83 billion RMB in 2023, 96.99 billion RMB in 2024, and 133.22 billion RMB in 2025, with Q1 2026 at 33.68 billion RMB.

QVeris Bot further compared using the following formula:

Operating cash flow to net profit ratio = Net operating cash flow ÷ Net profit attributable to shareholders

Corresponding results were 2.10x, 1.91x, 1.85x, and 1.62x.

In other words, during all four reporting periods, Ningde Time's operating cash flow exceeded its net profit attributable to shareholders for the same period. Combined with profit margin changes, the company not only achieved profit scale growth but also maintained strong operating cash generation capability.

When intelligent investment research tools proactively connect two financial statements, they provide more than isolated numbers—they offer an analytical framework ready for further exploration.

Inventory and Total Assets Reveal Expansion Pace

On the balance sheet, Ningde Time's ending inventory increased from 45.43 billion RMB in 2023 to 59.84 billion RMB in 2024, further reaching 94.53 billion RMB in 2025, with Q1 2026 ending at 108.94 billion RMB.

Corresponding total assets were 717.17 billion RMB, 786.66 billion RMB, 974.83 billion RMB, and 1,046.33 billion RMB, respectively. By the end of Q1 2026, the company's total assets had surpassed 1 trillion RMB.

Inventory changes may relate to raw material reserves, work-in-progress scale, finished goods stockpiling, and business growth. By placing inventory, revenue, profit, and cash flow on the same sheet, researchers can more easily identify metrics requiring continued tracking.

For example, subsequent analysis could further examine inventory composition, turnover efficiency, impairment provisions, and the alignment between inventory growth and orders or production.

Second-Stage Verification Demonstrates Research-Grade AI Value

In subsequent verification, QVeris Bot clarified that original data was in yuan, with tables displaying values divided by 100 million and rounded to two decimal places.

Revenue, net profit attributable to shareholders, and non-GAAP net profit attributable to shareholders corresponded to income statement items; net operating cash flow came from the cash flow statement; inventory and total assets corresponded to balance sheet items. Derived metrics were calculated using the formulas above.

It also performed multi-path verification of key data. For example, comparing revenue and net profit attributable to shareholders across different financial interfaces; inferring total assets via liabilities and debt-to-asset ratio; estimating total operating cash flow through per-share operating cash flow and total shares outstanding.

Verification results showed consistent data across all groups, with unit conversions and field mappings mutually verifiable.

This process is crucial because the core of professional research isn't merely "obtaining a number," but knowing which reporting period the number belongs to, which financial statement item it originates from, what calculations were applied, and whether it can be re-verified through an alternative path.

From this perspective, QVeris Bot delivered not just a financial statement, but a foundational draft ready for subsequent research workflows.

Intelligent Investment Research Is Evolving from "Writing Views" to "Building Evidence"

This Ningde Time financial data verification question demonstrated several capabilities of QVeris Bot in structured financial research: it can standardize accounting principles across multiple reporting periods, batch-organize core financial metrics; distinguish between period-based and point-in-time data; complete profit margin and cash flow quality calculations; and continue verifying sources, formulas, and consistency in subsequent conversations.

What truly consumes analysts' time is rarely writing a sentence like "the company has long-term competitiveness," but rather compiling base sheets, verifying accounting principles, checking units, identifying anomalies, and determining whether a conclusion has sufficient evidence.

When AI can reliably complete these foundational tasks, it ceases to be merely a content generation tool and begins to integrate into the research workflow.

Evaluating an intelligent investment research product can ultimately be distilled into four questions: Which reporting period does this number belong to? What accounting principles were applied? What calculation methodology was used? Can it be re-verified through an alternative path?

QVeris Bot's complete response to this Ningde Time financial data verification question provides a noteworthy sample: the value of intelligent investment research is gradually shifting from "writing views for people" to "building evidence for views."