> For the complete documentation index, see [llms.txt](https://docs.powermonitor.com.br/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.powermonitor.com.br/en/power-monitor/qualidade-de-dados.md). # Data Quality Overview of the Data Quality module: Environment X-Ray, lineage, dictionary, best practices scores (models and reports) and AI score, environment inventory, workspace X-ray and data exposure. The **Data Quality** module is the part of Power Monitor that looks **inside** your semantic models and reports: how they were built, where the data comes from, who consumes it, whether they follow best practices, whether they are ready for AI and whether there are credentials exposed in the code. It also brings the **Environment X-Ray**, an overall diagnosis that gathers the problems of the whole organization on one screen. While **Governance** shows the artifact inventory and **Monitoring** shows consumption and execution, Data Quality answers questions such as: * "What is the overall health of my environment and what should I fix first?" * "If I change this table in the source database, which models and reports will break?" * "Is this model well documented enough for Copilot to answer questions about it?" * "Does this report follow design, governance and accessibility best practices?" * "Is there a password or token written directly in the Power Query of any model?" **How to access:** side menu **Data Quality**. The module screens are **read-only** and are available to any user profile (Administrator or User), with one exception: the **Environment X-Ray** is exclusive to **Administrators**. The data displayed respects your user's **workspace scope**: if an administrator has restricted your visibility to certain workspaces, you will see only the models, reports and items from those workspaces. An administrator can also hide specific pages for a user in **Users**; the block also applies to queries made through the product APIs. The predefined **Audit** profile comes with this module's screens blocked (an administrator can allow them in **Users**). ## The module screens | Screen | What it answers | | ----------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- | | [Data Lineage](/en/power-monitor/qualidade-de-dados/linhagem-de-dados.md) | Where a model's data comes from and who consumes it (sources, gateways, reports, apps and composite models) | | [Environment X-Ray](/en/power-monitor/qualidade-de-dados/raio-x-do-ambiente.md) | Overall diagnosis of the organization: a 0 to 100 score and problems by severity across six areas, with a link to fix each one (Administrators only) | | [Data Dictionary](/en/power-monitor/qualidade-de-dados/dicionario-de-dados.md) | Starting from a physical object (gateway, server, database, table, column), which models and reports use it | | [Best Practices Score](/en/power-monitor/qualidade-de-dados/score-de-boas-praticas-bpa.md) | Adherence of a model to 55 best practice rules (Best Practice Analyzer, Microsoft guidelines and Power Monitor heuristics) | | [Report Best Practices Score](/en/power-monitor/qualidade-de-dados/score-de-boas-praticas-de-relatorios.md) | Adherence of a report to 61 rules on page design, data binding, governance, lifecycle and accessibility | | [AI Score](/en/power-monitor/qualidade-de-dados/score-de-ia.md) | How prepared a model is for Copilot and natural-language questions | | [Environment Inventory](/en/power-monitor/qualidade-de-dados/inventario-do-ambiente.md) | One row per semantic model in the tenant, with dozens of structure, health, usage and score metrics | | [Workspace X-Ray](/en/power-monitor/qualidade-de-dados/raio-x-de-workspace.md) | How many artifacts of each type exist in each workspace | | [Data Exposure](/en/power-monitor/qualidade-de-dados/exposicao-de-dados.md) | Credentials written in model code and other leakage risks | {% hint style="info" %} **Model Cleanup** (tables, columns and measures that no report uses) now lives in the **Performance** module: see [Model Cleanup](/en/power-monitor/performance/limpeza-de-modelo.md). The old address keeps working. {% endhint %} {% hint style="info" %} All the module's analyses work on the **data already collected** by Power Monitor (model inventory and mapping, report structure, permissions, usage). None of them connects to your semantic model in real time or changes anything in Power BI/Fabric. If a model was changed recently, wait for the next collection to see the effect here. See [Mapping](/en/power-monitor/mapeamento.md). {% endhint %} ## Features by screen Each page in the module has a **Features** section that presents, with a screenshot and step by step, each feature of the screen. Summary: | Screen | Main features | | ----------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | [Data Lineage](/en/power-monitor/qualidade-de-dados/linhagem-de-dados.md) | **Table and Columns** and **Power BI Items** tabs, table/column search with source column, item search, diagram with details and highlighting, zoom, full screen and export (PNG, JSON, text) | | [Environment X-Ray](/en/power-monitor/qualidade-de-dados/raio-x-do-ambiente.md) | Overall score, findings by severity, trend, cards for the six areas with a link to the screen that fixes each problem, **Refresh analysis**, **Summarize with AI** and export | | [Data Dictionary](/en/power-monitor/qualidade-de-dados/dicionario-de-dados.md) | Gateway › Server › Database › Table › Column tree, search at every level, "where it is used" diagram, zoom and export (PNG, JSON, CSV) | | [Best Practices Score](/en/power-monitor/qualidade-de-dados/score-de-boas-praticas-bpa.md) | Model selection, overall score, nine categories (click to filter), findings table with **Recommendation** and the **Source** link of each rule, **Rules not evaluated**, catalog version, **Export CSV**, **Run again** and the tab in the model details | | [Report Best Practices Score](/en/power-monitor/qualidade-de-dados/score-de-boas-praticas-de-relatorios.md) | Report selection (Power BI or paginated), five categories, origin badge of each rule, **Rules not evaluated** and coverage, **Export CSV**, **Run again** and the **Best Practices** tab in the report details | | [AI Score](/en/power-monitor/qualidade-de-dados/score-de-ia.md) | Model selection, overall score, six weighted categories, findings table with filters and **Items per page**, **Rules not evaluated**, **Export CSV**, **Run again** and the tab in the model details | | [Environment Inventory](/en/power-monitor/qualidade-de-dados/inventario-do-ambiente.md) | KPI cards, general search (including by table, column, measure or report), header filters, sorting, **Items per page** and model details | | [Workspace X-Ray](/en/power-monitor/qualidade-de-dados/raio-x-de-workspace.md) | Search, filter coming from the Governance Dashboard, matrix by artifact type, sorting, actions menu and workspace details | | [Data Exposure](/en/power-monitor/qualidade-de-dados/exposicao-de-dados.md) | Risk type filter (credentials and broad risks), masking notice, summary and findings per model, notice and configuration of **Microsoft standard artifacts** | ## Concepts you will come across **Data lineage:** the path that data takes from the original source (database, file, Lakehouse) to the report, passing through the semantic model. It helps measure the impact of a change. **Composite model:** a semantic model that consumes another Power BI semantic model (DirectQuery to a semantic model). In the lineage, it appears as a *Source model* or *Dependent model*. **Storage mode:** how the model stores its data: **Import**, **DirectQuery**, **DirectLake**, **Dual** or **Mixed** (tables with different modes). **Score:** a rating from 0 to 100 that summarizes adherence to a set of criteria. Bands used in the model and report scores: **80 or more = Good**, **60 to 79 = Attention**, **below 60 = Needs attention**. In the Environment X-Ray the bands are **Healthy** (80 or more), **Attention** and **Critical** (below 50). **Best Practice Analyzer (BPA):** a set of modeling best practice rules that originated in Tabular Editor. Power Monitor's best practices scores also use rules from Microsoft guidelines and its own heuristics; in the model score, the **Source** link shows where a rule comes from and, in the report score, each rule also carries an origin badge. **Rules not evaluated:** rules the score could not calculate because data is missing (for example, a report without a captured structure). They stay out of the rating and are listed with the reason. **RLS / OLS:** Row-Level Security and Object-Level Security (which hides tables or columns from certain roles). **Hard-coded credential:** a password, token or key written directly in the model's code (Power Query/M or DAX) instead of being protected in a connection or gateway. **Microsoft standard artifacts:** items that Microsoft itself creates in the tenant (Fabric Capacity Metrics, usage metrics models, the Admin monitoring workspace). The analyses ignore them by default; see [Microsoft standard artifacts](/en/power-monitor/qualidade-de-dados/exposicao-de-dados.md#microsoft-standard-artifacts). ## How to use ### How to open a screen in the module 1. In the side menu, click **Data Quality** to see the module's screens. 2. Choose the screen: **Data Lineage**, **Environment X-Ray** (Administrators), **Data Dictionary**, **Best Practices Score**, **Report Best Practices Score**, **AI Score**, **Environment Inventory**, **Workspace X-Ray** or **Data Exposure**. 3. If a screen does not appear in the menu, an administrator may have blocked access to it for your user in **Users**. ### How to analyze a specific model or report The **AI Score** and **Best Practices Score** screens work the same way: 1. Type at least 2 characters in the **Search semantic model...** field. 2. Choose the model from the list of suggestions (click, or use **↑**/**↓** and **Enter**). The bar below the search confirms the model and the workspace. 3. Wait for the *Analyzing the model...* message to be replaced by the result. To redo the analysis after a new collection, click **Run again**. The **Report Best Practices Score** follows the same path, starting from the **Search for a report...** field. ### How to start from the overall diagnosis {% stepper %} {% step %} ### Environment X-Ray See the overall score and the critical and high findings across all areas (Administrators only). {% endstep %} {% step %} ### Follow the finding link Each finding leads to the screen that fixes the problem: Alerts, Gateways, Data Exposure, Capacity Cost, Permissions Dashboard and others. {% endstep %} {% step %} ### Go deeper with the scores For model quality problems, open the Best Practices Score or the AI Score of the model pointed out. {% endstep %} {% endstepper %} ### How to investigate the health of a semantic model {% stepper %} {% step %} ### Environment Inventory Find the model and see, in a single row, its structure, RLS, refresh failures, AI and best practices scores and usage efficiency classification. {% endstep %} {% step %} ### Best Practices Score and AI Score Open the model's specific problems, with the recommendation for each one. {% endstep %} {% step %} ### Model Cleanup Find out what can be removed from the model to make it smaller and simpler, in [Performance › Model Cleanup](/en/power-monitor/performance/limpeza-de-modelo.md). {% endstep %} {% step %} ### Data Exposure Confirm that there are no credentials written in the model's code. {% endstep %} {% endstepper %} ### How to assess the impact of a change at the source {% stepper %} {% step %} ### Data Dictionary Start from the gateway, server, database, table or column that will change and see all the models and reports that depend on it. {% endstep %} {% step %} ### Data Lineage Open each affected model and confirm the path to the dependent reports, apps and composite models. {% endstep %} {% step %} ### Workspace X-Ray See the composition of the workspaces involved to plan communication with the owners. {% endstep %} {% endstepper %} ### How to run a security review {% stepper %} {% step %} ### Data Exposure List the models with exposed credentials and, if you want, enable the broad risks (exfiltration, external destinations, dynamic connections). {% endstep %} {% step %} ### Environment Inventory Cross the affected models with RLS and sensitivity label. {% endstep %} {% step %} ### Permissions Audit Check who has access to the affected workspaces and items in [Permissions Audit](/en/power-monitor/auditoria/auditoria-de-permissoes.md). {% endstep %} {% endstepper %} ## Related pages * [Performance](/en/power-monitor/performance.md): Model Cleanup, Performance Assessment and Average Execution Time * [Governance › Semantic Models](/en/power-monitor/governanca/modelos-semanticos.md): inventory and details of each model (the lineage and score tabs also appear there) * [Governance › Reports](/en/power-monitor/governanca/relatorios.md): the **Best Practices** tab in the report details * [Mapping › Models and reports › Model Mapping](/en/power-monitor/mapeamento/mapeamento-de-modelos.md): the collection that feeds most of these screens * [Monitoring › Semantic Models](/en/power-monitor/monitoramento/modelos-semanticos.md): refresh failures and duration * [Audit](/en/power-monitor/auditoria.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.powermonitor.com.br/en/power-monitor/qualidade-de-dados.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.