ThoughtSpot vs Metabase: which to choose

When a ThoughtSpot alternative fits Metabase best

Metabase fits best when the trigger is a ThoughtSpot renewal or per-user cost that no longer matches how business users actually work, or when business questions must stay on servers you control. The decision is less about whether ThoughtSpot is capable and more about whether your users need a search-led, AI-assisted experience or a structured query builder they can use directly. Start by counting how many people ask questions versus how many people only view saved Liveboards, because that split drives both the license cost and the operating model.

ThoughtSpot still fits when search-led exploration is central to how people work, when users expect natural-language questions over a governed semantic model, or when an enterprise contract is already signed and the total cost is acceptable to the business. It also fits when the organization has invested in its own ThoughtSpot content and the migration effort would not pay back in a reasonable period. Metabase fits when most self-service questions can be answered through its query builder and models, when your team is comfortable defining metrics in a shared layer, and when you are willing to run the application on PostgreSQL in your own infrastructure. The Open Source edition is free to self-host with unlimited users and dashboards, but it does not include SSO or row and column level permissions.

The decision turns on two questions. First, which plan carries the identity and permission features your organization requires: SSO through SAML or OIDC, row and column level permissions, and advanced LDAP group sync are in the Pro or Enterprise plans, so confirm current tiers on the Metabase pricing page before budgeting. Second, who will run upgrades, backups of the application database and migrations, since Metabase applies schema migrations automatically on upgrade. If no one on your team owns that work, factor a support or retainer arrangement into the comparison from the start.

Which option fits which scenario

This table maps common ThoughtSpot usage patterns to the option that usually fits better, with the condition that makes each advantage matter. Read it against your own user mix, identity setup and data location, since a single organization often has several rows that point in different directions.

Scenario	Better fit	Why
Most users ask simple questions of governed, curated data	Metabase	The query builder and models let non-technical users filter and group a governed dataset without writing SQL, which matters when the semantic layer is already defined and stable.
Users type free-text questions into a search bar	ThoughtSpot	Search-style natural language questions are ThoughtSpot's core interaction. Metabase relies on structured questions, so this advantage matters most where users expect to type open questions and get charts back.
SSO and row-level permissions are required	Metabase Pro or Enterprise	Metabase's Open Source and Starter plans do not include SSO or row and column level permissions, so the paid plan is the deciding cost when these controls are mandatory.
Data must stay inside your own network	Metabase Open Source, self-hosted	Running Metabase on your own servers keeps query results and application data in your network. Confirm which edition provides the air-gapped deployment you need, since the pricing page limits it to Enterprise.
The team has no in-house analytics engineers	ThoughtSpot	ThoughtSpot's modeling and search features reduce the amount of data modeling that business users depend on engineers to perform. Metabase still needs someone to run the application database, upgrades and backups.
An enterprise contract is mandatory for procurement	ThoughtSpot or Metabase Enterprise	Procurement may require a vendor agreement with support terms. Metabase Enterprise is priced by quote, and ThoughtSpot's terms vary by deal, so ask both vendors for written support and data processing terms.
Embedded analytics in a customer-facing product	Metabase	Metabase documents embedding options in its embedding introduction. Confirm the authentication and licensing model for external users with the vendor before committing, since embedding needs can differ from internal use.

How Metabase's query builder compares with ThoughtSpot search

The core difference is how a non-technical user starts a question. In ThoughtSpot, a business user types or speaks a search-style phrase such as a metric and a region, and the engine turns it into a query against the semantic model. That natural-language search is a genuine ThoughtSpot strength, and teams that depend on ad hoc questions asked in plain words should weigh it heavily. In Metabase, the user works through a point-and-click query builder: they pick a table or a saved model, choose columns, filters, summaries and groupings, and see the result. Metabase has no equivalent of free-text search over your data, so the quality of the answers depends on how well your models and metrics are defined before anyone opens the tool.

That dependency is the main operational consequence. Metabase models are curated, saved queries with clear column names and descriptions, and metrics are reusable definitions that analysts can reference across questions. A business user who finds a well-named model with a documented metric can answer most routine questions without help. A user who finds raw tables with cryptic column names will struggle, and the answers they get may be wrong in ways that are hard to spot. Plan for an analyst or data owner to build and maintain these models, and count that time as part of the total effort, not as a side task.

Do not decide on the feature list alone. Take three or four real questions from your own business users, the ones they actually ask each week, and test both tools with them. In ThoughtSpot, check how well the search phrasing works on your current Models and whether answers match your finance or operations numbers. In Metabase, build the same questions with the query builder on a test copy of the data and time how long a non-analyst takes to reach a correct answer. Record where each tool needed help. That evidence will show whether your users need search-style freedom or structured, governed questions.

Metabase vs ThoughtSpot: capabilities side by side

This table compares how ThoughtSpot and self-hosted Metabase handle the capabilities that matter for self-service exploration, from question asking to permissions and support. Read the Metabase column with its edition in mind, since SSO, SCIM and row-level permissions depend on the plan you buy. Confirm each ThoughtSpot cell with your ThoughtSpot account team before relying on it.

Capability	ThoughtSpot	Metabase (self-hosted)
Question-asking interface	Search-style natural language questions over the semantic model, which is the product's core interaction. Confirm current search and AI features with ThoughtSpot.	Visual query builder for non-technical users, plus SQL editor for analysts. Business users build questions without writing SQL.
Semantic model and metrics	Models built from tables, views and SQL views, exported and managed as TML. Confirm how TML export maps to your objects.	Models and metrics defined in the application, with a semantic layer built from saved questions. Governance depends on how you curate them.
Dashboards	Liveboards hold charts and answers for shared viewing.	Dashboards hold questions and filters. Free Open Source edition includes unlimited dashboards.
Embedding	Embedded analytics offered as a product capability. Confirm embedding terms and licensing with ThoughtSpot.	Embedding documented in the Metabase embedding introduction. Confirm which embedding methods your plan supports before design.
Identity (SSO, SCIM)	Confirm supported identity providers and provisioning options with ThoughtSpot for your edition.	Google sign-in and LDAP on Open Source and Starter. JWT, OIDC, SAML and SCIM require Pro or Enterprise, per the Metabase pricing page.
Row and column permissions	Confirm row-level security and group controls with ThoughtSpot for your deployment.	Row and column level permissions require Pro or Enterprise. Not available on Open Source or Starter.
Air-gapped deployment	Confirm with ThoughtSpot whether your self-managed release supports your network model.	Air-gapped deployment is listed only on the Enterprise plan. Confirm on the Metabase pricing page.
Application database	Metadata and platform storage run inside the ThoughtSpot deployment. Confirm the architecture for your edition.	Stores metadata in PostgreSQL recommended for production, or MySQL or MariaDB. Embedded H2 is for demos only.
Source of truth for data	Queries run against your warehouse through ThoughtSpot connectors. Confirm supported sources for your edition.	Queries run against your data sources directly. Metabase does not copy the warehouse, and the application database holds only Metabase metadata.
Support	Vendor support depends on your ThoughtSpot contract. Confirm response terms in writing.	Community support on Open Source. Support is included on Pro and Enterprise plans per the pricing page. Confirm terms for your plan.

Where the data lives and who controls the platform

Moving from ThoughtSpot to Metabase changes where queries run, where metadata sits and who holds the keys to the platform. The cards below cover the decisions that follow from those changes, from hosting and identity to compliance evidence and licensing.

Where queries and metadata live

Metabase sends queries to your existing warehouse or database and keeps its own metadata, such as questions, dashboards, models, users and permissions, in an application database. For production, the Metabase docs recommend PostgreSQL. Map which tables and columns the application database will hold, confirm that it sits inside your network boundary, and include it in backup and retention plans alongside the warehouse itself.

Hosting model and network placement

ThoughtSpot is delivered as a vendor-hosted cloud service or as self-managed ThoughtSpot Software. Self-hosted Metabase runs as a JAR or Docker container in your cloud account, data center or air-gapped network, so you choose the network zone, egress rules and patch windows. Confirm whether your environment needs air-gapped operation, because the Metabase pricing page places air-gapped deployment only in the Enterprise plan.

Identity and provisioning by plan

Single sign-on and user provisioning are where plan choice matters most. Per the Metabase pricing page, SSO, row and column level permissions and usage auditing sit in Pro and Enterprise, while Open Source and Starter rely on Google sign-in and LDAP. JWT, OIDC, SAML and SCIM provisioning also require Pro or Enterprise. Check which identity provider and group-sync model your ThoughtSpot tenant uses today before choosing a plan.

Compliance evidence and who produces it

Self-hosting moves configuration, access reviews, logging and backup evidence into your own control. DeployOpen prepares systems, controls and evidence for compliance assessments, but it does not issue audit reports, attestations or certifications. Independent auditors and certification bodies do that work. Ask your auditor which Metabase settings, logs and change records they expect to see before you cut over.

Licensing and the commercial code path

Metabase is open core. Code outside the enterprise directory is AGPL-3.0, while code inside it falls under the Metabase Commercial License, and the open-source Docker image is separate from the enterprise image. Review which features your team depends on, confirm they exist in the open-source build, and record which upgrade path you would need if a feature sits behind a paid plan.

Upgrade ownership and version support

On ThoughtSpot, the vendor controls the release cadence of the hosted service. With self-hosted Metabase, your team owns upgrades, which run application database migrations automatically. Metabase LTS versions are supported for 14 months, while non-LTS versions usually reach end of life after 2 months. Back up the application database before each upgrade and plan your release path against the version support page.

What each Metabase plan includes and costs

The Open Source edition is free to use and is self-hosted by you on your own servers or cloud accounts. It is the version you get from the open-source Docker image (metabase/metabase), and it has no per-user charge, so cost comes from the infrastructure, the application database and the staff who run it. Features are the core query builder, dashboards, models and self-service questions, with no limit on users or dashboards.

Price: free to download and run.
Users: unlimited users and dashboards.
Sign-in: Google sign-in and LDAP are supported, but SSO through JWT, OIDC or SAML is not.
Not included: support from Metabase, row and column level permissions, multi-tenant data segregation, usage analytics and auditing, and air-gapped deployment.
Running cost: servers, PostgreSQL backups, upgrades and your own internal support.

Before relying on this tier for ThoughtSpot users, confirm on the Metabase pricing page which access controls you need, since row and column permissions are the usual gap for teams with restricted data.

What ThoughtSpot and self-hosted Metabase cost to run

Running Metabase yourself replaces a per-user or credit-based subscription with a set of components you pay for directly: infrastructure, the application database, plan fees if you need SSO or permissions, and staff time. The cards below separate one-time migration work from recurring operating costs so you can build a fair comparison from your own user counts and usage.

ThoughtSpot subscription and quote

ThoughtSpot is sold per user or on a credit basis, and enterprise pricing is set by quote, so no list price is given here. Request a written quote that states user counts by role, the credit assumptions behind any usage model, the deployment type (cloud or self-managed) and the renewal term. Ask how overage and additional credits are billed, since that is where recurring cost often grows. This is the recurring line you are replacing, so every other card should be compared against it.

Metabase plan fees

The Open Source edition is free to self-host with unlimited users and dashboards. Single sign-on, row and column level permissions, multi-tenant data segregation and usage analytics and auditing sit in the paid Pro and Enterprise plans, according to the Metabase pricing page. Starter is $100 per month, or $90 per month billed yearly, with 5 users included, and Pro is $575 per month, or $517.50 per month billed yearly, with 10 users included, with extra users at $6 (Starter) or $12 (Pro) per month. Confirm current figures on the pricing page before budgeting, because permission needs are often the deciding factor.

Server and cloud infrastructure

Metabase runs as a JAR, so the recurring infrastructure cost is the virtual machines or containers, storage, backups and network egress in your own cloud account or data center. Costs rise with the number of concurrent dashboard viewers, query load on your warehouse, and high availability requirements. Plan for a staging environment so upgrades can be tested before production. Ask your platform team to price these components in your own account, since no hosting figures are supplied for this comparison.

PostgreSQL application database

Metabase stores its own metadata, questions, dashboards and permissions in an application database. The embedded H2 database is for demos, and PostgreSQL is the recommended choice for production. You will need to run it with backups, patching and monitoring, either on a managed service in your cloud or on a database team's servers. Moving from H2 uses a one-time load command that must match the Metabase version that last updated the H2 file, so budget that migration step separately from ongoing database operations.

Administration and upgrade time

This is the recurring staff cost that a subscription hides. Someone must monitor the service, manage users and groups, rotate credentials, and apply Metabase upgrades. Metabase runs database migrations automatically, but the documentation advises backing up the application database first, reducing a cluster to one node for major upgrades, and upgrading release by release from versions older than 40. Non-LTS releases typically reach end of life after 2 months, while LTS releases are supported for 14 months, so choosing an LTS line lowers upgrade frequency.

One-time migration effort

Migration is the main one-time cost. It includes exporting and rebuilding Models, Liveboards and Answers as Metabase models, metrics and dashboards, redesigning search-style questions for the query builder, configuring SSO and permissions, and training users. Effort rises with the number of Liveboards, the complexity of ThoughtSpot Models, and how many business users depend on ad hoc search. Ask a migration partner for a fixed-scope written plan that lists these items, and confirm whether the ThoughtSpot TML exports will be used as the specification.

Running Metabase yourself on PostgreSQL

The operating effort for self-hosted Metabase centers on its application database, the store where Metabase keeps questions, dashboards, users and settings. Metabase ships with an embedded H2 database for demos and recommends PostgreSQL for production. MySQL 8.4.0 or MariaDB 10.6.0 and later with utf8mb4 are also supported. Choose the database before go-live, because moving off H2 later means running the documented load-from-h2 command with the same Metabase version that last wrote the H2 file. Your team will own PostgreSQL backups, patching, sizing and high availability, and the cost of that work is the part of the self-hosted total that ThoughtSpot's subscription currently covers.

Upgrades are where most of the recurring work sits. Metabase runs application database migrations automatically when it starts a new version, so take a backup of the application database before every upgrade. For major upgrades, the upgrading guide recommends reducing a cluster to a single node because of a migration lock. Versions older than 40 must be upgraded release by release rather than jumped to the latest. Build a test environment that restores a recent application database backup and runs the upgrade path before production, and confirm the exact steps for your target version in the Metabase upgrading documentation.

Plan your release cadence against the support model. Metabase LTS versions are supported for 14 months, while non-LTS versions usually reach end of life after two months. Metabase 64 is an LTS release published on 16 September 2026 with end of life on 1 December 2027, and Metabase 65 was released on 12 October 2026. An LTS line gives a longer window between upgrades, but you still need to schedule the backup, test and cutover work for each one. Decide which line your business users can tolerate, confirm current end-of-life dates on the Metabase version support page before you commit, and budget for at least one planned upgrade within the window you choose.

What a move from ThoughtSpot to Metabase involves

The sequence turns ThoughtSpot's governed models, Liveboards and user habits into a Metabase deployment on PostgreSQL that business users can query and that the organization can run and audit itself.

Inventory content, owners and dependent users
List every ThoughtSpot Model, Worksheet, Liveboard, Answer, alert and embedded use, and record who owns each one and which teams open it each week. Ask the ThoughtSpot administrator for usage data so you can separate the content people rely on from content nobody has opened for months. Also identify where the data behind each Model comes from, including the source databases and any SQL views. The step is done when you have a signed-off list that marks each object as migrate, consolidate or retire, with a named business owner for every migrate item.
Export ThoughtSpot definitions as TML
Export the Models, Views, Tables, Answers and Liveboards as TML files so the definitions become your written specification. Keep the export in version control so changes during the migration are traceable. Confirm with your ThoughtSpot administrator which object types your export permissions cover, because some objects may need a separate export path. The step is done when every migrate item from the inventory has a matching TML file stored in the project repository and someone other than the author has reviewed the join paths and column formulas.
Set up PostgreSQL and the Metabase application database
Install Metabase on infrastructure you control and point its application database at PostgreSQL rather than the embedded H2 database, which Metabase provides for demos. Connect Metabase to the same analytical data sources that ThoughtSpot queries, using read-only service accounts. Plan backups of the application database before any upgrade, since Metabase runs schema migrations automatically on upgrade. Confirm with your platform team who operates PostgreSQL, how backups are tested and which Metabase release line you will track. The step is done when a test instance connects to the data warehouse, its application database is on PostgreSQL, and a restore from backup has been demonstrated.
Rebuild models, metrics and dashboards
Recreate the governed logic from the TML specification as Metabase models and metrics, so business users pick curated tables with agreed definitions instead of raw tables. Rebuild the most-used Liveboards as Metabase dashboards, and move the rest only when a named owner asks for them. Where ThoughtSpot Models hold large amounts of formula logic or row-level rules, expect this step to take the most design work, because each rule must be restated in SQL or Metabase's query builder. The step is done when each migrated Model has a documented metric list, each priority dashboard matches agreed figures from the ThoughtSpot original for a sample period, and discrepancies are logged with an owner.
Configure SSO, groups and data permissions
Decide which identity model you will use for sign-in and map ThoughtSpot groups and privileges to Metabase groups. Check the plan requirements before you design the setup: the Metabase pricing page places SSO, row and column level permissions and usage auditing in the Pro and Enterprise plans, while Open Source and Starter do not include them. Confirm with Metabase which sign-in methods your chosen plan supports, and confirm with your security team how group membership will be provisioned and removed. The step is done when test accounts in each group see only the collections and data their role allows, and a leaver's access is removed in the test run.
Pilot with a small group of business users
Run a pilot with a small group of business users who ask ad hoc questions in ThoughtSpot today. Give them the query builder, the migrated models and the priority dashboards, and watch where they struggle. Search-style questions behave differently in Metabase, so expect users to need training on how to phrase filters and pick metrics, and collect the questions they used to ask so you can add missing models. Record the questions they cannot answer and the performance issues they report. The step is done when the pilot group can answer their routine questions without help from the data team, and the open gaps are listed with an owner and a decision.
Retire ThoughtSpot content and keep the TML archive
Once the pilot passes, move the remaining users to Metabase, announce a read-only period for ThoughtSpot content, and then retire the objects that were marked as migrate or retire. Keep the TML export and a record of the inventory decisions as the archive of what the content did, so auditors and future teams can see the original logic. Check with your contract owner the notice period and the data retention terms for the ThoughtSpot subscription before you stop access. The step is done when no active dashboards remain in ThoughtSpot for migrated users, the archive is stored with the access controls you use for other business records, and the decommission date is recorded.

Questions buyers ask before changing ThoughtSpot

What drives the cost of moving from ThoughtSpot to Metabase?

The main drivers are the people doing the work, the Metabase edition you choose, and the hosting you run. The Open Source edition is free to self-host with unlimited users and dashboards, so the license cost is zero, but it lacks SSO and row-level permissions. Metabase's pricing page lists Starter at $100 per month ($90 billed yearly) with 5 users included, and Pro at $575 per month ($517.50 billed yearly) with 10 users included, with extra users at $6 or $12 per month. Enterprise is priced by quote starting at $20,000 per year. One-time costs are the content inventory, rebuilding models and dashboards, testing and training. Recurring costs are infrastructure, PostgreSQL, upgrades and support. Ask any quote to itemize these components and state the scope of content included.

Can an internal team migrate ThoughtSpot to Metabase without outside help?

A small team with SQL skills and Linux operations experience can often manage a limited migration, but the work usually needs more than BI skills. The team must run Metabase and PostgreSQL in production, integrate identity, rebuild content while ThoughtSpot stays live, and keep users working during cutover. Many teams find the hardest part is the rebuild and the change in how users ask questions, not the installation. DeployOpen plans, implements, integrates, hardens, backs up, documents and supports self-hosted platforms in your own infrastructure. Engagements are fixed-scope, with scope, deliverables and timelines agreed in writing before work starts. Before deciding, list the people who would own the platform after go-live and check their time and experience.

How long does a ThoughtSpot to Metabase migration take?

No duration can be given responsibly without a scoped plan. The drivers are the number of Models, Liveboards and Answers in use, how many joins and formulas must be rebuilt, the number of data sources, the identity and permission rules that must match, and how many users need training. Parallel running and the number of review cycles with business owners also add time. A useful first step is exporting your content with TML so the migration is based on an inventory rather than estimates. Ask for a written plan that lists phases, dependencies, acceptance criteria and the points where your team must make decisions. Treat any timeline without that inventory as a starting assumption only.

Can Metabase replace ThoughtSpot search for business users?

Metabase covers much of self-service exploration through its query builder, models, metrics and dashboards, but it is not a like-for-like match for search-style questions. Users typing free-form questions will need to work from curated models and metrics. Before committing, confirm with Metabase which natural-language or search features, if any, are available in the edition you plan to run. A practical approach is to identify the questions users ask most often, build governed models for them, and train users on the query builder. Plan for a change in habits, and test the most common questions with real users before retiring ThoughtSpot Answers.

Which Metabase plan includes SSO and row-level permissions?

The Pro and Enterprise plans include SSO and row and column level permissions, according to Metabase's pricing page. The Open Source and Starter plans do not include these features. Starter does support Google sign-in and LDAP. JWT, OIDC and SAML, including documented setups for Auth0, Entra ID, Google, Keycloak and Okta, along with advanced LDAP group and attribute sync and SCIM provisioning, require Pro or Enterprise. Check your identity requirements against this list before choosing a plan, because identity integration is often the deciding factor. Confirm the current feature matrix with Metabase, since plan contents can change.

Can Metabase run air-gapped, and which plan does that need?

Metabase's pricing page states that only the Enterprise plan offers air-gapped deployment. Enterprise is priced by quote starting at $20,000 per year. The open-source Docker image (metabase/metabase) is separate from the enterprise image (metabase-enterprise), so confirm which image and license terms apply to your environment. An air-gapped deployment also needs an offline process for container images, updates and any driver files, plus a documented way to receive security fixes. Ask Metabase to confirm these terms in writing, and confirm your own network and change control requirements before you plan the deployment.

How are Metabase upgrades handled after go-live?

Upgrades are a recurring task, and they are easier to plan when you know the release cadence. Metabase LTS versions are supported for 14 months, while non-LTS versions usually reach end of life after about 2 months. Metabase 64 is an LTS release published on 16 September 2026 with end of life on 1 December 2027. Upgrades run application database migrations automatically, so back up the application database first. For major upgrades, reduce a cluster to a single node because of a migration lock, and upgrade release by release if your installation is older than version 40. Your team needs a test environment, a rollback plan and a change window. Ongoing upgrades, support and testing can be covered under a DeployOpen retainer.

How DeployOpen helps evaluate and implement Metabase

DeployOpen plans, implements, integrates, hardens, backs up, upgrades, documents and supports self-hosted open-source platforms inside your own infrastructure. For ThoughtSpot vs Metabase, we start with the self-service questions your business users ask today, then decide whether Metabase, with the right edition and application database, covers them in your environment.

How we scope

Inventory which ThoughtSpot Liveboards, Answers and users actually drive self-service use each week
Confirm which Metabase edition covers the SSO, row-level permissions and directory sync your identity model needs
Agree data source list, hosting target, backup approach and upgrade ownership in a written scope before work starts
Engagements are fixed-scope, with deliverables and timelines confirmed in writing before work begins

How we build

Deploy Metabase on your own infrastructure with PostgreSQL as the application database, not the embedded H2 database
Connect your data sources, configure models and metrics, and set up collections and permission groups
Test permissions, query behavior and performance against representative business questions before users see them
Document the configuration, network paths and authentication setup so your team can maintain it

How we hand over

Deliver runbooks for installation, backup and restore of the application database, and version upgrades
Walk administrators through user, group and permission management, including what each plan covers
Provide a migration record that maps exported ThoughtSpot content to the Metabase objects that replace it
Hand over admin access and documented ownership so your staff run the platform after acceptance

How we support

Ongoing support, upgrades and continuous testing are available as retainers after handover
Plan upgrades release by release, with a backup of the application database taken first
Review Metabase version support dates with you so upgrades are scheduled before end of life
Ask any provider, including DeployOpen, to confirm what a retainer covers in writing before signing

ThoughtSpot vs Metabase