--- id: "en/ai-data-assistant" title: "Intelligent data assistant" language: "en" description: "Intelligent Data Assistant for querying, analysis, cross-file statistics, and chart generation in Excel and third-party data sources." canonical_url: "https://www.babel.cc/public/babelwiki/en/ai-data-assistant.html" markdown_url: "https://www.babel.cc/public/babelwiki/ai/en/ai-data-assistant.md" alternates: {"en": "https://www.babel.cc/public/babelwiki/ai/en/ai-data-assistant.md", "zh": "https://www.babel.cc/public/babelwiki/ai/zh/ai-data-assistant.md"} reading_note: "文档记录不等于当前部署能力；以实际版本、授权模块与配置为准。保留原文的可选、限制及待确认说明，勿补猜接口。 / Documentation is not a guarantee for every deployment. Preserve optional-module requirements, limitations and unresolved API contracts." --- # Intelligent Data Assistant Intelligent Data Assistant for querying, analysis, cross-file statistics, and chart generation in Excel and third-party data sources. ![Excel Data Assistant in the Zhichao AI application list](https://www.babel.cc/public/babelwiki/assets/images/excel-assistant-entry.png) *Open Zhichao AI and select Excel Data Assistant (Excel数据助理) on the left. This guide follows that interface from file selection to preview, questions and export.* ## First task: check a summary of one spreadsheet **Prepare:** Choose a small spreadsheet you can access, with clearly defined columns and a few values you can check manually. Start with one file; no third-party data source is needed. 1. Open **Zhichao AI → Excel Data Assistant (Excel数据助理)**. 2. Select **Choose my cloud files (选择我的云盘文件)** on the right to use a file stored in BabelBird, or **Click or drag a file to upload (点击或拖拽文件上传)** to add a local file. 3. Under **File Preview (文件预览)** on the left, check the filename and data. Use the worksheet tabs at the bottom to view a sheet. The current interface accepts `csv`, `xls` and `xlsx`, with up to **3 files**, **10M per file**, and only the **first three valid sheets** processed. 4. In the question box on the right, specify the filename, worksheet, columns, row range and filters. First ask the assistant to state the data range, row count and key columns. Then request analysis, for example: “In sales.xlsx, use the Monthly sheet, range A1:F101, with row 1 as headers. Total sales by region, state the row count and grand total, and explain how blanks are handled.” Replace these names and ranges with your actual data. 5. Check sample groups, totals and units against the original before asking for a chart. Select **Analysis Results (分析结果)** on the left, then **Export as document (导出为文档)** to save the result. The answer also provides Copy, Copy as Markdown and Save as document actions. 6. **Completion check:** The file, sheet and analysis range are correct, sampled results match the original, and chart units and filters are correct. ![Cloud-file and local-upload choices in Excel Data Assistant](https://www.babel.cc/public/babelwiki/assets/images/excel-assistant-import.png) *The right panel shows both import options and the file-count, size, format and worksheet limits. These are limits of the illustrated Excel Data Assistant, not general cloud-drive upload limits.* ## Check worksheets, analysis scope and processing status {#data-scope-status} ![File preview, worksheet tabs and a completed Excel Data Assistant answer](https://www.babel.cc/public/babelwiki/assets/images/excel-assistant-preview.png) *The filename appears above the preview, with worksheet tabs at the bottom. The question, reasoning status and answer appear on the right. This example compares survival rates by passenger class.* - **File and sheet:** Check the filename, column headings and bottom worksheet tabs. If the target sheet is not among the first three valid sheets, prepare an analysis copy containing the needed worksheets and import it. - **Analysis range:** State the sheet, cell range, columns and filters in the question, and ask the answer to identify the range actually used. The preview’s `A1` name box and cell selection let you inspect the spreadsheet; the selection alone does not establish the AI’s analysis scope. - **Read verification:** Once the preview opens, ask for the filename, sheets, column names, row count and a few sample rows, then compare them with the original. A working preview does not prove that the AI has read every value. - **Answer status:** “Completed deep thinking, time taken…” (已深度思考，用时……) and the generated answer show the outcome of that response. They do not indicate whole-file ingestion or full-sheet indexing. The supplied interface does not show a separate “all data read” indicator. **Empty or inconsistent results:** First check the limits of 3 files, 10M per file and the first three valid sheets, then permissions, column names, date filters, blank values and number formats. If reading still fails, provide the administrator with the filename, sheet, question and error message. Verify results manually before using them in formal reports. ## View and save analysis results {#data-results} ![Analysis Results and Export as document in Excel Data Assistant](https://www.babel.cc/public/babelwiki/assets/images/excel-assistant-results.png) *Select Analysis Results (分析结果) to see the summary and chart, and Export as document (导出为文档) to save them. The conversation remains on the right for follow-up questions.* ## Function description The intelligent data assistant is designed for enterprise spreadsheets, ledgers, reports and business data queries. It can perform cross-analysis around multiple Excel files, and can also be connected to third-party data sources accessed as URLs depending on the implementation. Suitable for business users who don't want to manually write formulas, filter conditions or statistical scripts. ## Main abilities - Supports cross analysis and comprehensive query of multiple Excel files. - Supports access to third-party data sources as URLs, depending on enterprise configuration. - AI can automatically program data query, cleaning, statistics and analysis based on the problem. - Charts can be generated based on query results to assist in report production or business analysis. - Supports the use of table Q&A in the Excel file preview interface or Zhichao AI section. - Can be combined with deep thinking mode to handle more complex statistics, functions and analysis tasks. - Supports multi-language interaction, suitable for cross-regional or cross-language teams. ## Applicable scenarios - Summary analysis of sales orders, inventory, project ledgers, supplier lists, customer lists and financial details. - Cross-reference between Excel files maintained by multiple departments. - Ask questions in natural language, such as "Sales this month by region", "Find overdue items", "Compare missing customers in two tables". ## Notes The scope of data processed by the intelligent data assistant should be confirmed by the enterprise. Sensitive data such as finance, personnel, and customer privacy should be used in conjunction with permissions, auditing, and data-masking strategies. ## Permissions and usage boundaries For shared requirements on source authorization, model services and human review, see [Zhichao AI Overview](https://www.babel.cc/public/babelwiki/en/ai-overview.html). The completion checks for this task do not replace your organization’s access approvals. ## Frequently asked questions {#data-faq} ### Where is Excel Data Assistant, and how do I add files? {#faq-data-import} Select Excel Data Assistant (Excel数据助理) in Zhichao AI. On the right, choose a cloud file or click/drag to upload a local file. The current interface accepts csv, xls and xlsx: up to 3 files, 10M per file, and only the first three valid sheets. See the import screenshot above. ### How do I choose a worksheet and analysis range? {#faq-data-range} Use the worksheet tabs at the bottom of File Preview to inspect data. In your question, specify the filename, worksheet, range and filters, and ask the answer to state the actual range used. Selecting cells in the preview alone does not prove that the AI’s scope is restricted to them. If a target sheet is outside the first three valid sheets, prepare an analysis copy containing it. ### How do I check that the assistant has read the data? {#faq-data-read} Check the preview, then ask for the filename, sheet, column names, row count and a few sample rows. Compare them with the original before continuing. “Completed deep thinking” describes the current answer, not full-sheet reading or ingestion. The supplied screenshot does not show a separate indicator that all data has been read. ### What should I check if results are wrong or empty? {#faq-data-check} Check file-count and size limits and the first-three-valid-sheets restriction, then permissions, sheet, row range, date filters, blanks and number formats. Ask the answer to identify the data actually used and verify samples against the original. If the issue persists, give the administrator the filename, question and error message. ### Where do I export the finished analysis? {#faq-data-export} Select Analysis Results (分析结果) on the left, then Export as document (导出为文档). The answer also offers Copy, Copy as Markdown and Save as document. Verify data scope, calculations and chart units before exporting. ## Related information - [Permission system](https://www.babel.cc/public/babelwiki/en/permissions.html) - [Security and Audit](https://www.babel.cc/public/babelwiki/en/security.html) - [FAQ: Deployment, AI and technical issues](https://www.babel.cc/public/babelwiki/en/faq-technical.html) - Public introduction: