BI · Analytics

Apache Superset: open-source BI platform for building products

Apache Superset: open-source BI for data sovereignty—self-hosted, data stays on your perimeter, SQL Lab, 50+ charts, embedded SDK. KT.Team runs OSNO-VA, management accounting with embedded n8n automation layer.

Superset isn't just another dashboard builder—it's a platform for building products: Apache 2.0, self-hosted, data stays on your perimeter.

Get a consultation

Apache Superset overview: SQL Lab, charts and dashboards

Our clients

FSK Group

SMLT

Tochno

Sber City

Relief Center

Askona

Snezhnaia Koroleva

Muztorg

TVOE

Yandex

Lenta

International perfume and cosmetics brand

RAEC

L'Etoile

Apache 2.0open license unmodified—your code, no vendor lock-in, no sanctions risk

v6 → v7v6 released December 2025, v7 coming H1 2026: the project is actively developed

40+supported databases and connectors: ClickHouse, PostgreSQL and the full SQL stack for DWH

50+chart types including geospatial and cross-dashboard filters built-in

Industry solutions

What you can do with Apache Superset

Management accounting and finance Build P&L views and management dashboards on self-hosted BI with data on your infrastructure Management reporting and P&L trusted for decision-making

Product and SaaS Embed Superset dashboards into your service via embedded SDK with JWT and role-based access Embedded analytics inside your product

Retail and e-commerce Build sales and inventory analytics on top of ClickHouse and PostgreSQL Real-time sales analytics on self-hosted BI

BI localization Replace Power BI or Tableau with self-hosted Superset including dashboard and access migration Migrate your BI infrastructure to open-source without sanctions risk

Capabilities

Apache Superset capabilities

ClickHousePostgreSQL1C via ETLSuperset: SQL Lab, charts, embedded SDKRLS / JWT (access & security)Dashboards and embeddingn8n automationOSNO-VA: management accounting

Left: sources (ClickHouse, PostgreSQL, 1C via ETL); center: Superset (SQL Lab, 50+ charts, embedded SDK) with RLS/JWT; right: consumers—dashboards and embedding, n8n automation layer, and OSNO-VA product.

OSNO-VA: a product on Superset, not consulting

KT.Team runs its own product on Apache Superset—management accounting OSNO-VA with embedded n8n under the hood: users transition seamlessly from dashboards to integrations. Most vendors sell Superset as dashboard consulting; we maintain a live product.

SQL Lab: powerful SQL editor

Class-leading SQL editor: analysts write queries, save them as datasets and charts—the entire flow from raw data to dashboard stays within BI.

50+ chart types and cross-filter

50+ visualization types including geospatial and native cross-dashboard filters—build complex management analytics without third-party plugins.

Embedded SDK with JWT

Production-ready dashboard embedding in your product with JWT authentication—Superset serves as the analytics layer of a SaaS service, not just an internal portal.

40+ data connectors

Connect to 40+ databases; ClickHouse and PostgreSQL cover the DWH layer—Superset sits atop your existing data stack.

RLS and enterprise access

Row-Level Security and enterprise authentication (OpenID, LDAP, OAuth-SSO). RLS works via WHERE injection and requires thoughtful data schema—we design this upfront.

Approach

How we implement Apache Superset

Without modifying the core

We don't fork or patch Apache Superset core. It stays on a standard updated version while business logic runs in separate microservices; platform updates don't break your customizations.

International Standards, Not Homegrown Hacks

Where a mature international solution exists, we use it instead of inventing our own protocol or platform. Before writing code, we study how the problem is already solved in the industry.

Transferability

The solution is loosely coupled and documented: it can be handed over between teams and contractors without rewriting. You are not tied to us.

AI compatibility

Apache Superset in the AI perimeter

OSNO-VA: BI + embedded n8n automation

OSNO-VA embeds n8n under Superset's hood: users flow from dashboards into n8n integrations for transaction posting, data source connectors, triggers and broadcasts. BI shifts from 'view' to 'act'.

Superset as an MCP tool for agents

Expose dashboards and SQL layer to an agent via MCP: LLM queries metrics and explains them in natural language without touching production schema.

Data on your infrastructure tailored to your data model

Self-hosted Superset keeps data within your perimeter: you can integrate local and external LLMs for analytics without exporting sensitive data.

AI layer and KT.Team platform team

Production Superset requires a real platform team and automation—exactly what KT.Team does daily and packaged as a product, not one-off consulting.

Context 2026

What changed in the market

KT.Team runs a product on Superset, not consulting

OSNO-VA: management accounting on Apache Superset with embedded n8n—users transition seamlessly from dashboards to integrations. Most vendors sell Superset as dashboard consulting; KT.Team maintains a live product.

What you need to run Superset in production

Production Superset requires five processes (web, worker, beat-scheduler, Redis, Postgres-metadata). Docker Compose handles ~20 users; production needs Kubernetes/Helm. RLS uses WHERE injection; PDF export and mobile are limited. Best fit: mid-sized team with a part-time platform engineer and SQL-literate analysts.

Active project, license unchanged

Superset is actively maintained: Apache 2.0 license, v6 in December 2025, v7 in H1 2026, dozens of contributors monthly. With Power BI and Tableau gone, it's a mature data sovereignty path—self-hosted, data on your perimeter, even a registry fork Superset TA exists.

Used by large enterprises

Apache Superset is used by enterprise teams like VkusVill and Ozon, plus thousands of companies globally—platform users of Superset, not KT.Team clients. Engine maturity proven at scale.

Honestly

Pros and cons

Pros

Open source under Apache 2.0 with no vendor lock-in or sanctions risk: self-hosted, data stays on your infrastructure—the foundation for replacing Power BI and Tableau.
Strong SQL Lab, 50+ chart types and production-ready embedded SDK with JWT—build not just internal portals but entire products on Superset.
KT.Team runs OSNO-VA, its own product on Superset with embedded n8n—a live practice, not one-off dashboard consulting.
Production Superset operation (Kubernetes, platform, automation)—core expertise of KT.Team's AI-native and integration teams.

Cons

Production Superset requires five services and Kubernetes for scale: you need a real platform team, not just an out-of-the-box install.
RLS is implemented via WHERE injection and requires thoughtful data schema design; PDF export and mobile experience lag behind commercial BI platforms.
Sweet spot: a mid-sized team with a part-time platform engineer and SQL-literate analysts; for very small teams, Metabase onboards faster.