{
  "_meta": {
    "schema": "https://www.drawdecisiontree.com/decision-dag.schema.json",
    "source": "https://www.drawdecisiontree.com",
    "description": "DrawDecisionTree.com is a free tool for building, sharing, and embedding interactive decision trees. This file is the machine-readable export of a published decision tree. The `dsl` field contains the original source in the Decision DAG DSL; the `dag` schema is documented at the URL in `schema` above.",
    "links": {
      "interactive": "https://www.drawdecisiontree.com/t/drawdecisiontree/data-analytics-tool.html",
      "embed": "https://www.drawdecisiontree.com/embed/path/drawdecisiontree/data-analytics-tool",
      "dsl_reference": "https://www.drawdecisiontree.com/decision-tree-dsl-reference.html",
      "guides": "https://www.drawdecisiontree.com/guides",
      "schema_docs": "https://www.drawdecisiontree.com/decision-dag.schema.json",
      "author_trees": "https://www.drawdecisiontree.com/trees/drawdecisiontree"
    },
    "generated_at": "2026-05-29T12:05:39.267Z"
  },
  "author": {
    "handle": "drawdecisiontree",
    "first_name": "Andrew",
    "last_name": null,
    "avatar_url": "1d32d828-b6ca-40ec-bdd7-771fe7b9c36a/avatar-1778531481027.svg",
    "display_name": "Andrew"
  },
  "file": {
    "id": "c9dd71d8-2d0e-42ca-bd0f-124e2194d383",
    "name": "Which BI and analytics tool should I choose?",
    "public_slug": "data-analytics-tool",
    "updated_at": "2026-05-12T16:53:43.587978+00:00",
    "url": "https://www.drawdecisiontree.com/t/drawdecisiontree/data-analytics-tool.html",
    "json_url": "https://www.drawdecisiontree.com/t/drawdecisiontree/data-analytics-tool/tree.json",
    "dsl_url": "https://www.drawdecisiontree.com/t/drawdecisiontree/data-analytics-tool/tree.dag"
  },
  "meta": {
    "description": "Select the right business intelligence or analytics tool for your team's technical profile, cloud environment, and operational constraints. This tree weighs skill level, hosting preferences, and the importance of a governed semantic layer to surface the best-fit platform.",
    "mode": "elimination",
    "entry": "Q1",
    "tags": [
      "data",
      "business intelligence",
      "analytics",
      "reporting",
      "data engineering"
    ],
    "image": "https://images.unsplash.com/photo-1460925895917-afdab827c52f?w=1200&q=80"
  },
  "questions": [
    {
      "id": "Q1",
      "text": "What is the primary technical skill level of the main users of this tool?"
    },
    {
      "id": "A",
      "text": "Low-code — business users, drag-and-drop first [LOOKER, TABLEAU, POWER_BI, METABASE]"
    },
    {
      "id": "B",
      "text": "Technical — SQL-fluent analysts or engineers [LOOKER, SUPERSET, DBT_NOTEBOOK, METABASE]"
    },
    {
      "id": "Q2",
      "text": "Does your organisation have a strong preference for, or existing contract with, a specific cloud provider?"
    },
    {
      "id": "A",
      "text": "Microsoft / Azure ecosystem [POWER_BI]"
    },
    {
      "id": "B",
      "text": "Google Cloud / BigQuery ecosystem [LOOKER]"
    },
    {
      "id": "C",
      "text": "Cloud-agnostic or multi-cloud [TABLEAU, METABASE, SUPERSET, DBT_NOTEBOOK]"
    },
    {
      "id": "Q3",
      "text": "Is a fully managed SaaS deployment required, or is self-hosting acceptable?"
    },
    {
      "id": "A",
      "text": "Must be fully managed SaaS [LOOKER, TABLEAU, POWER_BI, METABASE]"
    },
    {
      "id": "B",
      "text": "Self-hosted is acceptable or preferred [SUPERSET, DBT_NOTEBOOK, METABASE]"
    },
    {
      "id": "Q4",
      "text": "Is a governed semantic layer or centralised data modelling capability a priority?"
    },
    {
      "id": "A",
      "text": "YES — consistent metric definitions and a governed model are critical [LOOKER, DBT_NOTEBOOK]"
    },
    {
      "id": "B",
      "text": "NO — speed to dashboard is more important than centralised modelling [TABLEAU, POWER_BI, METABASE, SUPERSET]"
    }
  ],
  "outcomes": [
    {
      "id": "LOOKER",
      "label": "Looker"
    },
    {
      "id": "TABLEAU",
      "label": "Tableau"
    },
    {
      "id": "POWER_BI",
      "label": "Power BI"
    },
    {
      "id": "METABASE",
      "label": "Metabase"
    },
    {
      "id": "SUPERSET",
      "label": "Apache Superset"
    },
    {
      "id": "DBT_NOTEBOOK",
      "label": "dbt + Notebook (Code-First)"
    }
  ],
  "dsl": "dag: Which BI and analytics tool should I choose?\nversion: 1.0.0\nimage: https://images.unsplash.com/photo-1460925895917-afdab827c52f?w=1200&q=80\ndescription: Select the right business intelligence or analytics tool for your team's technical profile, cloud environment, and operational constraints. This tree weighs skill level, hosting preferences, and the importance of a governed semantic layer to surface the best-fit platform.\ntags: data, business intelligence, analytics, reporting, data engineering\nentry: Q1\nmode: elimination\n\nQ1: What is the primary technical skill level of the main users of this tool?\n  hint: Think about who will own the day-to-day report building and dashboard maintenance — not the data engineers who set the platform up, but the analysts and business stakeholders who will live in it. \"Low-code / business user\" means people comfortable with drag-and-drop interfaces and Excel-level logic. \"Technical / SQL-fluent\" means analysts or engineers who are happy writing queries, using Git, and thinking in data models. Choosing the wrong level here leads to tools that are either too intimidating or too limiting for the actual user base.\n  A: Low-code — business users, drag-and-drop first [LOOKER, TABLEAU, POWER_BI, METABASE]\n  B: Technical — SQL-fluent analysts or engineers [LOOKER, SUPERSET, DBT_NOTEBOOK, METABASE]\n\nQ2: Does your organisation have a strong preference for, or existing contract with, a specific cloud provider?\n  hint: Power BI is deeply integrated with the Microsoft ecosystem (Azure, Teams, SharePoint, Active Directory) and offers the best value if you are already paying for Microsoft 365. Looker is a Google Cloud product and benefits from native BigQuery integration and GCP IAM. Tableau has strong connectors across all major clouds but carries a premium licence cost. If you are on AWS and cost-sensitive, open-source tools deployed on EC2 or ECS may be the most pragmatic choice.\n  A: Microsoft / Azure ecosystem [POWER_BI]\n  B: Google Cloud / BigQuery ecosystem [LOOKER]\n  C: Cloud-agnostic or multi-cloud [TABLEAU, METABASE, SUPERSET, DBT_NOTEBOOK]\n\nQ3: Is a fully managed SaaS deployment required, or is self-hosting acceptable?\n  hint: Managed SaaS means the vendor handles infrastructure, upgrades, and security patching — ideal for teams without a dedicated platform engineering function. Self-hosting gives you full control over data residency, customisation, and cost, but requires operational investment. Superset and dbt + notebooks are open-source and typically self-hosted; Metabase offers both a managed cloud tier and a self-hosted Community edition. If your data governance policy prohibits sending query results to third-party servers, self-hosting may be mandatory.\n  A: Must be fully managed SaaS [LOOKER, TABLEAU, POWER_BI, METABASE]\n  B: Self-hosted is acceptable or preferred [SUPERSET, DBT_NOTEBOOK, METABASE]\n\nQ4: Is a governed semantic layer or centralised data modelling capability a priority?\n  hint: A semantic layer lets you define metrics, dimensions, and business logic once in a central place so that every report and dashboard uses consistent definitions — eliminating the \"two reports, two different revenue numbers\" problem. Looker's LookML and dbt's metrics layer are purpose-built for this. Tableau and Power BI have partial semantic layer capabilities through calculated fields and datasets, but they require more discipline to keep consistent. If different teams currently define the same KPI differently, a strong semantic layer is likely non-negotiable.\n  A: YES — consistent metric definitions and a governed model are critical [LOOKER, DBT_NOTEBOOK]\n  B: NO — speed to dashboard is more important than centralised modelling [TABLEAU, POWER_BI, METABASE, SUPERSET]\n\n[LOOKER]: Looker\n  color: #4F86C6\n  description: Looker's LookML semantic layer makes it the gold standard for organisations that need a single source of truth for metrics and dimensions across many teams and dashboards. It integrates natively with Google BigQuery and supports robust access control, version-controlled data models, and embedded analytics. Budget for a meaningful implementation phase — LookML has a learning curve and typically requires at least one dedicated analytics engineer to maintain the model. Start by modelling your highest-value entities (orders, customers, sessions) and expand the semantic layer iteratively rather than attempting a full data dictionary on day one.\n  code: TOOL_LOOKER\n\n[TABLEAU]: Tableau\n  color: #E53E3E\n  description: Tableau offers best-in-class interactive visualisation capabilities and an intuitive drag-and-drop interface that empowers business analysts to build sophisticated dashboards without writing SQL. Its Hyper in-memory engine delivers fast query performance even on large extracts, and Tableau Prep provides a visual data wrangling workflow. Licence costs are significant — plan for Tableau Cloud or Server licensing per user and budget accordingly. Establish a Tableau Server governance process early: certify published data sources, enforce naming conventions, and schedule extract refreshes to prevent a proliferation of stale, duplicated workbooks.\n  code: TOOL_TABLEAU\n\n[POWER_BI]: Power BI\n  color: #ED8936\n  description: Power BI is the most cost-effective managed BI platform for organisations already invested in the Microsoft ecosystem, with seamless integration into Teams, SharePoint, Excel, and Azure Active Directory for row-level security. Its DAX formula language and Power Query ETL capabilities give technical analysts considerable flexibility, while the drag-and-drop report canvas is accessible to business users. Enable the Premium Per User or Fabric capacity tier if you need paginated reports, large dataset support, or deployment pipelines. Establish a workspace governance model early — use certified datasets and a promotion workflow to prevent uncontrolled report sprawl.\n  code: TOOL_POWER_BI\n\n[METABASE]: Metabase\n  color: #48BB78\n  description: Metabase is the fastest tool to get non-technical business users self-sufficient with data exploration — its question builder requires no SQL knowledge, and the interface is clean enough that it rarely needs training. The open-source Community edition is free to self-host, and the Cloud plan adds SSO, SMTP, and managed infrastructure at a modest per-seat cost. It lacks a formal semantic layer, so enforce metric consistency by curating a small library of official \"Questions\" and restricting write access to a trusted analytics team. Use the SQL editor and native query embedding for power users who need more than the GUI can offer.\n  code: TOOL_METABASE\n\n[SUPERSET]: Apache Superset\n  color: #9F7AEA\n  description: Apache Superset is a powerful open-source BI platform with a rich chart library, SQL Lab for ad-hoc querying, and role-based access control — all at zero licence cost. It connects to virtually every SQL-speaking data store via SQLAlchemy and supports physical and virtual datasets for lightweight data modelling. Plan for meaningful DevOps investment: Superset requires containerised deployment (Docker or Kubernetes), regular version upgrades, and an internal support function for user issues. Define a dataset curation workflow so that analysts publish clean, documented virtual datasets rather than letting ad-hoc SQL proliferate across hundreds of ungoverned charts.\n  code: TOOL_SUPERSET\n\n[DBT_NOTEBOOK]: dbt + Notebook (Code-First)\n  color: #38B2AC\n  description: A code-first stack of dbt for data modelling and Jupyter or Marimo notebooks for analysis and visualisation gives SQL-fluent engineers maximum control, reproducibility, and version-control integration. dbt's semantic layer and documentation features ensure metrics are defined once and tested continuously, while notebooks allow rich narrative analysis with Plotly, Altair, or Matplotlib charts. This approach requires a mature engineering culture — teams need Git workflow discipline, a CI pipeline for dbt tests, and a shared notebook platform (JupyterHub, Databricks, or Deepnote) for collaboration. Use this stack when your primary audience is data scientists and engineers, and pair it with a lightweight BI tool (Metabase or Superset) for business-user access to pre-built dashboards.\n  code: TOOL_DBT_NOTEBOOK\n"
}