Built by Claude FableDOCN | Thematic Exposure: 3/10
Data caveat: Daloopa is not connected this session (see company_context.md), so no daloopa.com/src citations exist. DigitalOcean reports one segment. The split below is built from management-disclosed ARR in the FY2026Q2 call (transcripts/DOCN_FY2026Q2.txt) and FMP revenue. Market share and TAM come from web sources and are directional only.
Segment table (FY2026Q2 run-rate)
| Segment | % of Revenue | Market Share | TAM | Theme Growth |
|---|---|---|---|---|
| Core cloud (Droplets, Kubernetes, storage, managed DBs for developers, SMBs and digital natives) | ~79% (~$890M of ~$1.12B run-rate) | ~0.2% of global cloud infrastructure | ~$575B annualized (Q2 2026 spend $143.4B, Synergy via secondary sources) | ~+40% industry; DOCN core cloud mid-teens |
| AI / inference cloud (inference engine, GPU droplets, agents) | ~21% (AI customer ARR $234M, +200% YoY) | ~0.7% of neocloud market | Neocloud ~$35B in 2026, rising to ~$237B by 2031 (Mordor, 46% CAGR) | >40% |
Sources: Q2 revenue $281.2M, +29% YoY; AI customer ARR $234M; inference over 70% of AI ARR; 85% of AI ARR from inference plus core cloud, not bare metal (DOCN_FY2026Q2 transcript, FMP earnings). Run-rate = Q2 revenue x4. Cloud shares: tech-insider, shattered.io. Neocloud TAM: DCD, SDxCentral.
Key competitors
- Core cloud: AWS (~28%), Azure (~20%), Google Cloud (~15%). Together they hold 63% of global cloud infrastructure, and that oligopoly does not include DOCN. In the developer and SMB "alt-cloud" niche, DOCN competes with Akamai/Linode, Vultr, Hetzner and OVHcloud.
- AI cloud: CoreWeave (Q2 2026 revenue $2.58B, more than 10x DOCN's total company revenue, ~$104B backlog), Nebius, Lambda and Crusoe. On inference specifically, it also faces API providers (Together, Fireworks, Baseten, OpenRouter-routed endpoints) and the hyperscalers' own inference services.
Gate questions
- Competitors with >15% share? In global cloud, three players (AWS, Azure, GCP) are at or near 15% or more, and DOCN is not one of them at ~0.2%. Neocloud and inference are fragmented. CoreWeave leads, and there is a long tail of 10+ GPU clouds plus API providers. DOCN is a niche participant in both.
- Could a customer replace DOCN within 12 months? Yes, for most of the base. Droplets, Kubernetes and object storage are commodity primitives that run on open standards, and management's own pitch is "zero lock-in" with open source at every layer. Stickiness is rising at the top: $1M+ customers had 0 churn in Q4 2025 and 115% NDR, and ~70% of $100K+ AI customers attach core cloud. Blended D&E NDR is only ~102%, though (DOCN_FY2025Q4). Inference is the easiest workload to move, since it is one API call behind routers like OpenRouter and Vercel's AI gateway.
- Price setter or taker? Price taker. DOCN wins on price and performance and on simplicity versus the hyperscalers (character.ai saw ~50% lower cost per token; DOCN had the #1 Artificial Analysis output speed on DeepSeek V3.2). Its "valuemaxxing" positioning sells lower cost per unit of intelligence. GPU and token pricing follows the market.
OLIGOPOLY HARD GATE: FAILED. DOCN has no segment with >30% share and is not one of <=3 players controlling >70% of any market, so the ceiling is 5/10.
Qualitative
- Why the position is durable (to the extent it is): DOCN has 12 years of developer brand and 4M+ active developers. It offers an integrated full stack (compute, storage, DBs, networking, inference, agents) that bare-metal neoclouds lack. AI natives that scale fast (Cursor, Higgsfield, Ideogram, Daytona, OpenCode) want one integrated environment with predictable pricing. The highest-spend cohorts grow fastest: $1M+ ARR +214% YoY, now 23% of ARR versus 9% a year ago (DOCN_FY2026Q2).
- Who the buyers are: Developers, SMBs and "digital native enterprises" (62% of ARR, +30% YoY as of Q4 2025), plus a fast-growing cohort of AI-native software companies whose largest COGS line is inference.
- What could replace it: Hyperscaler startup-credit programs and simplified tiers, CoreWeave/Nebius moving up-stack into managed inference (CoreWeave inference ARR is already >$100M, targeting >$250M by end of 2026), and inference API providers. Capacity is also a constraint: demand exceeds supply, and DOCN must raise capital (an $888M equity raise in Q1 2026) to compete for megawatts against far larger balance sheets.
Score rationale
The themes are excellent: cloud is growing ~40% and inference/neocloud ~46% CAGR, and DOCN is accelerating (revenue growth went from 18% to 22% to 29% over three quarters). But the rubric scores share and structure, not TAM. DOCN has <1% share in every market it competes in and sells largely substitutable services as a price taker. That puts it at the 1-3 end, and the ~10-20% share needed for 3-4 is clearly not met. The theme growth and real differentiation (integrated inference stack, top-cohort stickiness) lift it from 1-2 to 3.
Assessment: DigitalOcean is riding two of the fastest-growing themes in technology, AI inference and cloud for AI-native and digital-native companies, and its own trajectory is improving sharply: revenue +29% YoY, AI ARR +200%, record incremental ARR. However, it is a sub-1% share participant in markets dominated by the AWS/Azure/GCP oligopoly, with CoreWeave at 10x its scale in GPU cloud. Its offering is built on open, portable primitives that most customers could replace within a year, and it competes on price and performance rather than setting prices. By the principle "don't settle for the 2nd, worse the 3rd," DOCN is not a market leader. It fails the oligopoly hard gate and scores 3/10 on thematic exposure: great tailwinds, no dominant position.