Databricks — research brief

Researched: 2026-09-20 (CT)
Primary focus: Lakehouse + Mosaic AI / Data + AI Platform; consumption via DBUs + committed-use discounts; multi-cloud (AWS, Azure, GCP).

Labels: FACT = Databricks primary page / company newsroom; SECONDARY = press; INFERENCE = synthesis from facts.

Company & mission

FACT — Databricks is “the data and AI company.” Mission: simplify and democratize data and AI. HQ San Francisco; offices worldwide. Source: About us.
FACT — Founded in 2013 by the original creators of the lakehouse architecture and open-source Apache Spark™, Delta Lake, MLflow, and Unity Catalog (origins in UC Berkeley AMP Lab / open source). Source: About us, Press kit.
FACT — More than 20,000 organizations worldwide — including Block, Comcast, Condé Nast, Rivian, Shell and 70% of the Fortune 500 — rely on the Databricks Data + AI Platform. Press kit also lists adidas, AT&T, Bayer, Mastercard, Unilever among named customers. Sources: About us, Press kit.
FACT — 1,200+ global cloud, ISV, and consulting partners. Source: About us.
FACT — Platform surfaces called out in Aug 2026 company PR: Lakehouse, Unity Catalog, Lakeflow, Mosaic AI / Agent Bricks, Lakebase, Genie, Unity AI Gateway. Source: Aug 13, 2026 press release.

Products (what they sell)

FACT — Unified Data + AI Platform on an open lakehouse: governed data, analytics, AI agents, and apps. Source: Our approach, About us.
FACT — Pricing unit: Databricks Unit (DBU) — a normalized unit of processing power used for measurement and pricing; DBUs consumed depend on compute resources and data processed. Separate Databricks Storage Unit (DSU) for storage SKUs. Source: Pricing FAQ.
FACT — Commercial model: Pay as you go (no up-front costs; per-second granularity) and Committed Use Contracts (discounts / benefits for usage commitments, including cross-cloud flexibility). Source: Pricing.
FACT — Classic compute: customer also pays their cloud provider for VMs/storage/network; serverless SKUs fold infrastructure into the DBU price (per docs/pricing FAQ framing). Azure Databricks pricing is set by Microsoft. Source: Pricing FAQ.
FACT — Model Serving publishes GPU DBU/hour rates (examples): T4 10.48; A10G×1 20.00; A10G×8 290.80; A100 80GB×8 628.00; H100×1 100.00; H100×8 800.00. Premium on Azure maps to Enterprise on AWS/GCP. Source: Model Serving pricing.
FACT — Free trial of the Data + AI Platform; cloud infra still billed by the cloud provider if using the customer’s cloud account. Community Edition is a free, limited Spark learning environment. Source: Pricing FAQ.

Scale metrics (company disclosures)

FACT — Aug 13, 2026 company newsroom: crossed $7B revenue run-rate, >80% YoY growth in Q2; positive adjusted free cash flow LTM; Lakehouse (data warehousing) >$1.5B run-rate growing >100% YoY; Lakebase >$100M run-rate; >1,000 customers at >$1M run-rate; >100 customers at >$10M run-rate; closed $5B strategic funding at $190B valuation (led by Coatue with Blackstone, MGX, T. Rowe Price accounts, Sixth Street Growth, and others). Source: Company PR Aug 13, 2026.
FACT — Feb 9, 2026 company newsroom (earlier checkpoint): $5.4B revenue run-rate, >65% YoY; AI products $1.4B run-rate; NRR >140%; >800 customers >$1M; >70 customers >$10M. Source: Company PR Feb 9, 2026.

Partners & GTM

FACT — Runs on AWS, Azure, GCP (and references SAP availability on pricing pages). Azure Databricks is a first-party Microsoft service with Microsoft billing. Sources: Pricing, Model Serving.
FACT — Partner program: Bronze → Silver → Gold → Platinum; co-sell, marketplace, funding programs. Source: Partner Program, 2026 Partner Program Guide PDF.
INFERENCE — Dual GTM: product-led trial / Community Edition for developers + land-and-expand consumption (DBUs) + enterprise sales and committed-use contracts for large accounts; cloud marketplaces (esp. Azure) are major enterprise channels.

Metrics we are NOT inventing

Do not invent seat prices — Databricks is primarily consumption (DBUs), not seat SaaS; list prices vary by cloud, region, SKU, and tier and are shown on product pricing explorers.
Do not invent headcount or exact ARR beyond what company PR states as “revenue run-rate.”
Do not treat third-party blog DBU dollar examples as FACT unless mirrored on databricks.com.

Teaching arc (for canvas)

Free trial / Community Edition → pay-as-you-go DBUs (Jobs, SQL, All-Purpose, Mosaic AI / Model Serving) → Committed Use discounts → Enterprise / multi-cloud commits via sales & Azure first-party.

Lakehouse unifies data engineering + SQL analytics + governance (Unity Catalog); Mosaic AI / Agent Bricks / Genie / Lakebase expand AI attach so usage (and DBUs) grow with agent and analytics workloads.

# Databricks — research brief Researched: 2026-09-20 (CT) Primary focus: Lakehouse + Mosaic AI / Data + AI Platform; consumption via DBUs + committed-use discounts; multi-cloud (AWS, Azure, GCP). Labels: **FACT** = Databricks primary page / company newsroom; **SECONDARY** = press; **INFERENCE** = synthesis from facts. --- ## Company & mission - **FACT** — Databricks is “the data and AI company.” Mission: simplify and democratize data and AI. HQ San Francisco; offices worldwide. Source: [About us](https://www.databricks.com/company/about-us). - **FACT** — Founded in 2013 by the original creators of the lakehouse architecture and open-source Apache Spark™, Delta Lake, MLflow, and Unity Catalog (origins in UC Berkeley AMP Lab / open source). Source: [About us](https://www.databricks.com/company/about-us), [Press kit](https://www.databricks.com/company/newsroom/press-kit). - **FACT** — More than **20,000 organizations** worldwide — including Block, Comcast, Condé Nast, Rivian, Shell and **70% of the Fortune 500** — rely on the Databricks Data + AI Platform. Press kit also lists adidas, AT&T, Bayer, Mastercard, Unilever among named customers. Sources: [About us](https://www.databricks.com/company/about-us), [Press kit](https://www.databricks.com/company/newsroom/press-kit). - **FACT** — **1,200+** global cloud, ISV, and consulting partners. Source: [About us](https://www.databricks.com/company/about-us). - **FACT** — Platform surfaces called out in Aug 2026 company PR: Lakehouse, Unity Catalog, Lakeflow, Mosaic AI / Agent Bricks, Lakebase, Genie, Unity AI Gateway. Source: [Aug 13, 2026 press release](https://www.databricks.com/company/newsroom/press-releases/databricks-grows-80-yoy-surpasses-7b-revenue-run-rate-scales). ## Products (what they sell) - **FACT** — Unified **Data + AI Platform** on an open lakehouse: governed data, analytics, AI agents, and apps. Source: [Our approach](https://www.databricks.com/our-approach), [About us](https://www.databricks.com/company/about-us). - **FACT** — Pricing unit: **Databricks Unit (DBU)** — a normalized unit of processing power used for measurement and pricing; DBUs consumed depend on compute resources and data processed. Separate **Databricks Storage Unit (DSU)** for storage SKUs. Source: [Pricing FAQ](https://www.databricks.com/product/pricing). - **FACT** — Commercial model: **Pay as you go** (no up-front costs; per-second granularity) and **Committed Use Contracts** (discounts / benefits for usage commitments, including cross-cloud flexibility). Source: [Pricing](https://www.databricks.com/product/pricing). - **FACT** — Classic compute: customer also pays their cloud provider for VMs/storage/network; serverless SKUs fold infrastructure into the DBU price (per docs/pricing FAQ framing). Azure Databricks pricing is set by Microsoft. Source: [Pricing FAQ](https://www.databricks.com/product/pricing). - **FACT** — **Model Serving** publishes GPU DBU/hour rates (examples): T4 10.48; A10G×1 20.00; A10G×8 290.80; A100 80GB×8 628.00; H100×1 100.00; H100×8 800.00. Premium on Azure maps to Enterprise on AWS/GCP. Source: [Model Serving pricing](https://www.databricks.com/product/pricing/model-serving). - **FACT** — Free trial of the Data + AI Platform; cloud infra still billed by the cloud provider if using the customer’s cloud account. **Community Edition** is a free, limited Spark learning environment. Source: [Pricing FAQ](https://www.databricks.com/product/pricing). ## Scale metrics (company disclosures) - **FACT** — Aug 13, 2026 company newsroom: crossed **$7B revenue run-rate**, **>80% YoY** growth in Q2; positive adjusted free cash flow LTM; Lakehouse (data warehousing) **>$1.5B** run-rate growing **>100% YoY**; Lakebase **>$100M** run-rate; **>1,000** customers at **>$1M** run-rate; **>100** customers at **>$10M** run-rate; closed **$5B** strategic funding at **$190B** valuation (led by Coatue with Blackstone, MGX, T. Rowe Price accounts, Sixth Street Growth, and others). Source: [Company PR Aug 13, 2026](https://www.databricks.com/company/newsroom/press-releases/databricks-grows-80-yoy-surpasses-7b-revenue-run-rate-scales). - **FACT** — Feb 9, 2026 company newsroom (earlier checkpoint): **$5.4B** revenue run-rate, **>65% YoY**; AI products **$1.4B** run-rate; NRR **>140%**; **>800** customers >$1M; **>70** customers >$10M. Source: [Company PR Feb 9, 2026](https://www.databricks.com/company/newsroom/press-releases/databricks-grows-65-yoy-surpasses-5-4-billion-revenue-run-rate). ## Partners & GTM - **FACT** — Runs on **AWS, Azure, GCP** (and references SAP availability on pricing pages). Azure Databricks is a first-party Microsoft service with Microsoft billing. Sources: [Pricing](https://www.databricks.com/product/pricing), [Model Serving](https://www.databricks.com/product/pricing/model-serving). - **FACT** — Partner program: Bronze → Silver → Gold → Platinum; co-sell, marketplace, funding programs. Source: [Partner Program](https://www.databricks.com/partners/partner-program), [2026 Partner Program Guide PDF](https://www.databricks.com/sites/default/files/2026-05/partner-program-guides-2026.pdf). - **INFERENCE** — Dual GTM: product-led trial / Community Edition for developers + land-and-expand consumption (DBUs) + enterprise sales and committed-use contracts for large accounts; cloud marketplaces (esp. Azure) are major enterprise channels. ## Metrics we are NOT inventing - Do **not** invent seat prices — Databricks is primarily consumption (DBUs), not seat SaaS; list prices vary by cloud, region, SKU, and tier and are shown on product pricing explorers. - Do **not** invent headcount or exact ARR beyond what company PR states as “revenue run-rate.” - Do **not** treat third-party blog DBU dollar examples as FACT unless mirrored on databricks.com. ## Teaching arc (for canvas) **Free trial / Community Edition → pay-as-you-go DBUs (Jobs, SQL, All-Purpose, Mosaic AI / Model Serving) → Committed Use discounts → Enterprise / multi-cloud commits via sales & Azure first-party.** Lakehouse unifies data engineering + SQL analytics + governance (Unity Catalog); Mosaic AI / Agent Bricks / Genie / Lakebase expand AI attach so usage (and DBUs) grow with agent and analytics workloads.