| Reporting and Visualization | | |
| Variety of Visualizations | 16 basic visualization options. | 25 visualization options. Custom visualizations (box plot, histogram, etc) are also available - enabled by Vega-lite integration. |
| How non-technical users self-serve | Via a graphical query builder (Metabase’s Question). Users still have to pick and join data, filter, and perform group-by and summary before creating a visualization.
No drag-and-drop interface. No SQL code surfaced for validation. | Via drag-n-drop report builder. Users explore data and build dashboards using curated datasets - which are maintained and governed by the data team for better accuracy and consistency.
For every report, Executed Query is shown for validation if needed. |
| Row Limits | The download limit is 1 million. Display limit: 2,000 without aggregation, and 10,000 with aggregation. | Up to 1,000,000 rows for Data Exploration / Report Creation (i.e., in each chart). Can export full data into CSV/Excel file. |
| Report sharing | Push dashboards to Slack and email. | More advanced. Push dashboards to Slack, Email, Google Sheets, SFTP, and Webhooks. |
| Data Alerts | Available, but with limited options. | Available and more advanced. Comes with Dashboard Filters and Business Calculations. |
| Customer-Facing Analytics | | |
| Access Control for Embedded Dashboard | Limited - with only filter-based restrictions available. Documentation regarding this aspect is relatively sparse. | Yes, easy to use and easy to share. Better access control with row-level permission settings. Detailed documentation about embedded analytics security. |
| Public Shareable Link | Available but not secure. The only option to control public shareable links is to manually disable/enable them. | Available with more advanced configuration. Permission settings for Dataset, Data models, Fields. Password protection also available. |
| Click Behavior for Embedded Dashboards | Viewers can change filters only. | Viewers can change filters or drill through to navigate between widgets and reports within an embedded dashboard. |
| Data Analysts’ Experience | | |
| Semantic Layer for metric definitions | Not available. But can work around this limitation by integrating Metabase with Cube | Available. Semantic Layer allows analysts to manage business logic centrally and define reusable data models. |
| Reusing common SQL queries | No. As more and more queries are added, data team ends up with repeated SQL queries all over the place, making it difficult to know where the correct version is. | Available. Analysts easily define reusable SQL queries for data exploration. |
| Write analytics as code | No. Metabase is not designed for a code-based analytics workflow. | Yes. Analysts can define business logic and data models as code using Holistics AML- which allows analytics logic to be governed with Git Version control, enables code review, and most importantly, promotes reusability. |
| dbt integration | Available. | Available |
| Data modeling | Analysts add metadata to database tables and specify relationships, using a graphical UI. | Analysts define business metrics using SQL formulas, then map relationships between tables and transform raw tables into reusable data models for consistent usage. Can use both GUI and code view for modeling. |
| Governance | | |
| User Access Control | Row-level Access Control is available in Enterprise Plan. Column-level Permission is not supported. | Supported both Row-level Permission and Column-level Permission. |
| Usage Monitoring | Not Available. | Usage monitoring dashboard is available for admins to have a bird-eye view of who using what. |
| Version Control | Not Available. | Yes. With Git Integration. |
| Compliance and security | GDPR, SOC2 Type II. | GDPR, SOC2 Type II. |