# Databricks Series D Financial Model Databricks is the unified analytics platform (Serverless Spark PaaS + Data Science Workspace SaaS) that bridges the gap between big data infrastructure and AI/ML applications. - Canonical: https://finamodel.com/startups/databricks-series-d - Excel download: https://finamodel.com/startup-models/databricks-series-d.xlsx - Category: AI/ML - Model type: SaaS ARR / Valuation - Funding round: Series D - Funding: $140M - Founded: 2017 - Geography: Global (cloud-agnostic: AWS, Azure, GCP) with on-prem expansion planned. [DECK slides 15, 17] - Customer: B2B ## About the company Databricks combines a serverless Spark platform with a collaborative Data Science Workspace. The cloud-agnostic product runs across AWS, Azure, Google Cloud, and on-premises environments, offering governed data connectors, notebooks, dashboards, and reports for enterprise analytics and AI workloads. At the Series D deck date, Databricks had more than 500 customers, $21.9 million ARR, $118,000 average customer ARR, 136% net dollar retention, and 69% gross margin. Its Q1 2017 ARR was growing 248% year on year, with $36.7 million forecast for Q4. The model uses quarterly ARR waterfalls for annual enterprise contracts: new customers, expansion, and churn. It translates ARR into recognised revenue and cloud COGS, models retention, sales efficiency and LTV:CAC, then forecasts technical headcount, operating cash flow, and deployment of its $56.1 million cash balance. ## What's included - 5-year monthly revenue build with stage-appropriate growth assumptions - Full P&L, headcount plan, and operating-expense schedule - Cash-flow statement, runway, and burn-rate tracking - Valuation via exit multiple with a DCF cross-check - Returns analysis with MOIC and IRR - Unit economics including CAC, LTV, payback, and cohort retention ## Product & value proposition - Two-layer platform: - **Databricks Serverless Spark Platform** (PaaS): auto-tuning Spark with 10–40x speedup vs. open-source Apache Spark; custom data connectors; SOC2 & HIPAA compliant governance. - **Databricks Data Science Workspace** (SaaS): collaborative notebooks, real-time dashboards, periodic reports - democratizes Spark access for data teams. - Cloud-agnostic: runs on AWS, Azure, GCP, and on-prem (YARN/Cloudera/Hortonworks/Kubernetes). - Built by the original Apache Spark creators at UC Berkeley; Spark is the de facto standard for enterprise AI/ML at scale. - Addresses the "AI gap": the hardest part of AI is big-data infrastructure (Google NIPS 2015 framing). ## Market - Cloud computing market: Gartner estimate of $200B by 2020. - Big data: "90% of the data created in last 2 years." - ML/AI: described as having "just scratched the surface of use-cases." - Spark ecosystem traction (proxy for addressable market): Summit attendees 1,100 (2014) → 3,900 (2015) → 5,100 (2016); Meetup members 12K (2014) → 66K (2015) → 300K+ (2016). - No formal TAM/SAM/SOM breakdown provided. ## Revenue model - Subscription ARR - enterprise contracts billed annually. Implied by ARR metrics and average customer ARR. - Average Customer ARR = $118K as of 3/31/17. - 500+ customers across all major verticals (Ad/Marketing, Media, Healthcare/Pharma, Enterprise Software, Public Sector, Financial Services, Industrial/IoT, Retail/CPG). - Channel: direct enterprise sales (CRO background: Axway, Stanford MBA; CMO: Alteryx pre-IPO). - Partnership/cloud marketplace channel mentioned via Michael Hoff (SVP BD, ex-Tableau, Azure). ## Traction & metrics - ARR: $21.9M as of 3/31/17 - ARR annual growth: 248% (3/31/17 vs. 3/31/16) - 5x growth Q1'17 vs. Q1'16 - ARR quarterly progression (actuals + forecast): - Q1'16: $6.3M | Q2'16: $7.9M | Q3'16: $12.3M | Q4'16: $16.7M - Q1'17: $21.9M | Q2'17: $25.6M (forecast) | Q3'17: $31.0M (forecast) | Q4'17: $36.7M (forecast) - T2D3 comparison: T2D3 target ARR Q2'17 = $18M vs. Databricks actual/forecast = $25.6M - tracking ahead of T2D3. - Q2'15: T2D3 $2M / Databricks $1.9M; Q2'16: T2D3 $6M / Databricks $7.9M; Q2'17: T2D3 $18M / Databricks $25.6M - Customer count: 500+ - Average Customer ARR: $118K (up 112% yr/yr) - Net Dollar Retention: 136% (3/31/17 vs. 3/31/16) - Cash on hand: $56.1M as of 3/31/17 - Gross Margin: 69% (Q1'17) - LTV:CAC ratio: 4.1 (3/31/17 TTM) ## Unit economics - Gross Margin: 69% Q1'17 - LTV:CAC: 4.1 (TTM as of 3/31/17) - Net Dollar Retention: 136% - strong expansion motion; existing customers grow significantly. - Average Customer ARR: $118K, up 112% yr/yr - driven by both new logos and expansion. - Payback period: Not explicitly stated. With 69% GM and LTV:CAC of 4.1, implied payback ~14–18 months (typical for enterprise SaaS at this NRR), but not confirmable from deck. ## Competition / moat - Moat framing: Databricks created Apache Spark (Berkeley origins); deepest technical expertise in the ecosystem. - Platform lock-in: collaborative workspace + serverless infra = stickiness across data engineering, data science, and analytics teams. - Cloud-agnostic positioning differentiates from single-cloud-native offerings (AWS EMR, Azure HDInsight, GCP Dataproc). - On-prem expansion addresses Hadoop/Cloudera/Hortonworks incumbents. - Named competitors: Not explicitly named in deck. Implicitly: legacy data warehouses (Netezza, Oracle), Hadoop vendors (Cloudera, Hortonworks, MapR), cloud storage. - "AI gap" framing positions Databricks as the only unified solution across ETL, ML, and streaming. ## Team & funding ask / use of funds - Team: - Ali Ghodsi - CEO & co-founder; PhD/MBA, Adjunct Professor UC Berkeley - Patrick Wendell - VP Engineering & co-founder; UC Berkeley MSc, Princeton BS - Ron Gabrisko - CRO; Cyclone (pre-revenue → Axway acquisition), Stanford MBA/MS - Rick Schultz - CMO; Alteryx (pre-revenue → IPO), Oracle VP Product Marketing - John Winkenbach - SVP Finance; CFO Jobvite, VP Finance Technorati - Hatim Shafique - CCO; AppDynamics (pre-revenue → IPO) - Michael Hoff - SVP BD & Partnerships; Tableau VP Channels, EMC Global VP Sales, MSFT GM Windows Azure - Cash on hand pre-raise: $56.1M (3/31/17). --- ## Recommended financial model - **Archetype + why:** SaaS ARR subscription model with usage-based expansion layer. Databricks sells annual enterprise contracts with strong net dollar retention (136%), so the correct model is an ARR waterfall (new ARR + expansion ARR − churn ARR) driving a P&L, not a transactional revenue model. The 69% gross margin, LTV:CAC of 4.1, and cloud infrastructure cost structure are all consistent with a cloud SaaS P&L. - **Forecast horizon & granularity:** Quarterly for Years 1–2 (2017–2018), annual for Years 3–5 (2019–2021). The deck already provides quarterly ARR actuals through Q1'17 and management forecasts through Q4'17, so Q-level is natural. - **Key drivers & assumptions:** | Driver | Value | Source | | -- | -- | -- | | ARR at model start (Q1'17) | $21.9M | - | | ARR annual growth rate (historical) | 248% | - | | Q4'17 ARR target | $36.7M | - | | Average Customer ARR | $118K | - | | Customer count | 500+ | - | | Net Dollar Retention | 136% | - | | Gross Margin | 69% | - | | LTV:CAC | 4.1 | - | | ARR growth rate Year 2 (2018) | ~150% | Deceleration from 248% consistent with T2D3 double-double-double trajectory and scale | | ARR growth rate Year 3 (2019) | ~100% | Continued scale; Series D capital deployed into S&M | | ARR growth rate Years 4–5 | 60–80% | Normalization as base grows; comparable SaaS at $100M+ ARR | | NDR long-term | 120–130% | Slight moderation from 136%; still strong expansion motion | | Gross margin long-term | 72–75% | Modest improvement as platform scales; cloud infra costs optimized | | S&M % of revenue | 45–55% | Typical high-growth enterprise SaaS; consistent with LTV:CAC of 4.1 | | R&D % of revenue | 25–35% | Deep technical product; Spark ecosystem investment | | G&A % of revenue | 8–12% | Standard enterprise SaaS overhead | | New customer adds per quarter | ~30–50 | Implied from 500+ customers over ~10 quarters of operation | | Average contract length | 1 year | Standard enterprise SaaS; not stated in deck | - **Scenarios (Base / Bull / Bear - which variables flex):** - **Base:** ARR growth decelerates per table above; NDR holds at ~128%; GM improves to 73% by Year 3. - **Bull:** NDR sustains at 136%+; on-prem expansion adds a new revenue stream; cloud partnerships accelerate new logo adds; ARR growth stays >100% through 2019. - **Bear:** Enterprise sales cycle elongates; ARR growth decelerates to ~80% in 2018; NDR compresses to 115% as budget scrutiny increases; GM pressure from cloud infra costs. - Flex variables: new ARR growth rate, NDR, gross margin, S&M efficiency (CAC). - **Required sheets / outputs:** 1. **Assumptions** - all drivers in one place, color-coded vs. 2. **ARR Waterfall** - beginning ARR, new ARR, expansion ARR, churn ARR, ending ARR (quarterly then annual) 3. **Customer Cohort Schedule** - new logos per period, average ACV, expansion rate by cohort 4. **P&L** - revenue (ARR → recognized revenue), COGS, gross profit, S&M, R&D, G&A, EBITDA, net income 5. **Cash Flow** - operating cash flow; key: deferred revenue build (annual prepay), capex light 6. **KPI Dashboard** - ARR, ARR growth %, NDR, gross margin, LTV:CAC, average customer ARR, cash runway 7. **Scenario Toggle** - Base / Bull / Bear switcher driving all sheets ## Frequently asked questions ### Is the Databricks Series D financial model free? Yes. The Databricks Series D model is a free Excel download with live formulas. ### Can I change the assumptions? Yes. The workbook is editable and its live formulas recalculate when assumptions change.