對話式數據分析功能可讓您與資料對話，並由 Gemini for Google Cloud 提供技術支援。對話式數據分析功能可協助不具商業智慧專業知識的使用者，以一般自然語言 (對話) 提出資料相關問題，進一步瞭解靜態資訊主頁。對話式數據分析功能適用於 Looker (Google Cloud Core) 和 Looker (原始) 執行個體，以及 Looker Studio Pro 訂閱方案的 Looker Studio。
以下對話範例顯示使用者如何以自然來回的方式與對話式數據分析互動。在本例中，使用者提出以下問題：「Can you plot monthly sales of hot drinks versus smoothies for 2023, and highlight the top selling month for each type of drink?」(請繪製 2023 年熱飲和冰沙的每月銷售量，並標示出每種飲品的銷售量最高月份)。對話式數據分析會產生折線圖，顯示 2023 年熱飲和冰沙的每月銷售量，並醒目顯示 7 月是這兩類產品的銷售量最高月份。
s
如本範例對話所示，對話式數據分析會解讀自然語言要求，包括使用「銷售量」和「熱飲」等常見字詞的多部分問題，使用者不必指定確切的資料庫欄位名稱 (例如 Total monthly drink sales) 或定義篩選條件 (例如 type of beverage = hot)。對話式數據分析會說明主要發現、解釋推理過程，並提供包含文字的答案，以及適用的圖表。為鼓勵深入分析，對話式數據分析功能也可能會建議後續問題。
[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間：2025-09-08 (世界標準時間)。"],[],[],null,["\u003cbr /\u003e\n\n|\n| **Preview**\n|\n|\n| This product or feature is subject to the \"Pre-GA Offerings Terms\" in the General Service Terms section\n| of the [Service Specific Terms](/terms/service-terms#1).\n|\n| Pre-GA products and features are available \"as is\" and might have limited support.\n|\n| For more information, see the\n| [launch stage descriptions](/products#product-launch-stages).\n\nConversational Analytics is a chat-with-your-data feature that is powered by [Gemini for Google Cloud](/gemini/docs/overview). Conversational Analytics empowers users with no expertise in business intelligence to go beyond static dashboards and ask data-related questions in regular, natural (conversational) language. Conversational Analytics is available in Looker (Google Cloud core), Looker (original) instances, and within Looker Studio as part of a Looker Studio Pro subscription.\n\nThe following sample conversation shows how a user can interact with Conversational Analytics in a natural, back-and-forth way. In this example, the user asks the following question: \"Can you plot monthly sales of hot drinks versus smoothies for 2023, and highlight the top selling month for each type of drink?\" Conversational Analytics responds by generating a line graph that displays the monthly sales of hot drinks and smoothies for 2023, highlighting July as the month with the highest sales for both categories.\n\n\ns\n\nAs this sample conversation illustrates, Conversational Analytics interprets natural language requests, including multi-part questions that use common terms like \"sales\" and \"hot drinks,\" without requiring users to specify exact database field names (like `Total monthly drink sales`) or define filter conditions (like `type of beverage = hot`). Conversational Analytics describes its key findings, explains its reasoning, and provides an answer that includes text and, where appropriate, a chart. To encourage deeper analysis, Conversational Analytics may also suggest follow-up questions.\n\nLearn [how and when Gemini\nfor Google Cloud uses your data](/gemini/docs/discover/data-governance).\n| As an early-stage technology, Gemini for Google Cloud\n| products can generate output that seems plausible but is factually incorrect. We recommend that you\n| validate all output from Gemini for Google Cloud products before you use it.\n| For more information, see\n| [Gemini for Google Cloud and responsible AI](/gemini/docs/discover/responsible-ai).\n\nKey features\n\nConversational Analytics includes the following key features:\n\n- **Use Conversational Analytics in Looker** : Access [Conversational Analytics in Looker](/looker/docs/studio/conversational-analytics-looker) to ask natural language questions about your Looker Explore data within a Looker (original) instance or a Looker (Google Cloud core) instance.\n- **Use Conversational Analytics in Looker Studio** : Access [Conversational Analytics in Looker Studio](/looker/docs/studio/conversational-analytics-looker-studio) to ask natural language questions about data from [supported data sources](/looker/docs/studio/conversational-analytics-looker-studio#supported-data-sources). Requires a Looker Studio Pro subscription.\n- **Create and converse with data agents** : With [data agents](/looker/docs/studio/conversational-data-agents), you can customize the AI-powered data querying agent by providing context and instructions that are specific to your data, which helps Conversational Analytics generate more accurate and contextually relevant responses.\n- **Enable advanced analytics with the Code Interpreter** : The [Code Interpreter](/looker/docs/studio/conversational-analytics-code-interpreter) within Conversational Analytics translates your natural language questions into Python code and executes that code. Compared to standard SQL-based queries, the Code Interpreter's use of Python enables more complex analysis and visualizations.\n\n\nSetup and requirements\n\nTo use Conversational Analytics within a Looker instance, you and your Looker instance must meet the following requirements:\n\n1. Gemini in Looker must be enabled for the Looker instance.\n - To access these features in a Looker (original) instance, a Looker admin must [enable Gemini in Looker](/looker/docs/admin-panel-platform-gil) in the Looker (original) instance settings. The instance must be on Looker 25.2 or later and be Looker hosted. We recommend that customers participating in Lookers [Extended support release program](/looker/docs/standard-extended-support-release-program-overview) update to Looker 25.6 or later to use Conversational Analytics.\n - To access these features in a Looker (Google Cloud core) instance, a user with the [Looker Admin](/iam/docs/understanding-roles#looker.admin) (`roles/looker.admin`) IAM role must [enable Gemini in Looker](/looker/docs/looker-core-admin-gemini) in the Looker (Google Cloud core) instance settings in the Google Cloud console.\n2. The Trusted Tester capabilities must be enabled to use Conversational Analytics during the preview period.\n3. A Looker admin must grant you a Looker role that contains the [`gemini_in_looker` permission](/looker/docs/admin-panel-users-roles#gemini_in_looker) for the models that you're querying. This permission is available as part of the default [Gemini role](/looker/docs/admin-panel-users-roles#gemini). Additional permissions may be necessary to carry out the tasks that use Gemini assistance. You must also have a role that contains the [`access_data`](/looker/docs/admin-panel-users-roles#access_data) permission for the model that you are querying.\n\nTo use Conversational Analytics in Looker Studio, you must meet the following requirements.\n\n1. You must be a user under a [Looker Studio Pro subscription](/looker/docs/studio/looker-studio-pro-subscription-overview). Looker Studio Pro licenses are [available at no cost](/looker/docs/admin-panel-platform-lsp#accept) to Looker users.\n2. An administrator must have [enabled Gemini in Looker for Looker Studio](/looker/docs/studio/enable-and-disable-gemini-in-looker-for-looker-studio).\n3. The [**Trusted Tester features**](/looker/docs/studio/enable-and-disable-gemini-in-looker-for-looker-studio) must be enabled to use Conversational Analytics during the preview period.\n\nKnown limitations\n\nConversational Analytics has the following known limitations.\n\nLimitations on visualizations\n\nConversational Analytics leverages [Vega-lite](https://vega.github.io/vega-lite) for conversation chart generation. The following Vega chart types are fully supported:\n\n- Line chart (one or more series)\n- Area chart\n- Bar chart (horizontal, vertical, stacked)\n- Scatter plot (one or more groups)\n- Pie chart\n\nThe following Vega chart types are supported, but you may encounter unexpected behavior when rendering them:\n\n- Maps\n- Heatmaps\n- Charts with tooltips\n\nChart types that exist outside the Vega catalog are not supported. Any charts that are not specified in this section are considered unsupported.\n\n\nLimitations on data sources\n\nConversational Analytics has the following data source limitations:\n\n- For Looker data, Conversational Analytics can return a maximum of 5,000 rows per query.\n- Conversational Analytics doesn't support BigQuery's [Flexible Column Names](/bigquery/docs/schemas#flexible-column-names) feature.\n- Conversational Analytics doesn't work well with data sources that have [field editing in reports](/looker/docs/studio/edit-fields-in-your-reports) disabled because this setting prevents Conversational Analytics from creating calculated fields.\n- When the data source is Looker, Conversational Analytics cannot set the value of a filter-only that is defined using the LookML [`parameter`](/looker/docs/reference/param-field-parameter) parameter.\n- While Conversational Analytics generally supports connections to [Looker (Google Cloud core) instances with private connections configurations](/looker/docs/looker-core-networking-options#private_ip_connections), Conversational Analytics does not support Looker (Google Cloud core) instances that are configured to use [CMEK](/looker/docs/looker-core-cmek) or VPC Service Controls.\n- Using Conversational Analytics to connect to a private connections Looker (Google Cloud core) instance using Looker Studio Pro when that Looker (Google Cloud core) instance is inside a VPC Service Controls perimeter is not a supported configuration and does not meet VPC Service Controls compliance requirements.\n\nLimitations on questions\n\nConversational Analytics supports questions that can be answered by a single visualization, for example:\n\n- Metric trends over time\n- Breakdown or distribution of a metric by dimension\n- Unique values for one or more dimensions\n- Single metric values\n- The top dimension values by metric\n\nConversational Analytics doesn't yet support questions that can only be answered with the following types of complicated visualizations:\n\n- Prediction and forecasting\n- Advanced statistical analysis, including correlation and anomaly detection\n\nMore [advanced questions](/looker/docs/studio/conversational-analytics-code-interpreter#suggested-questions), such as forecasting, can be answered when the Code Interpreter is enabled.\n\nProvide feedback\n\nYou can provide feedback to Google about individual responses in Conversational Analytics by selecting one of the following options:\n\n- thumb_up **Good response**: Indicate that the response was helpful.\n- thumb_down **Bad response**: Indicate that the response was not helpful.\n\n| **Note:** Gemini in Looker is in preview with limited support. We encourage you to share your feedback to help us improve. To report bugs or issues, reach out to [Looker Support](/looker/docs/best-practices/looker-support-details) and be sure to include the following details:\n|\n| - A clear description of the problem and the expected behavior\n| - Steps to reproduce the issue\n| - Any additional relevant details"]]
