Cockroach Labs

1.1 Identity and business model

Cockroach Labs was founded in 2015, is headquartered in New York City, and remains described in the retained company directory as a Series F company. Its core product, CockroachDB, is an operational database built on a transactional, strongly consistent key-value store and exposed through a familiar SQL interface. The commercial proposition is resilient, horizontally scalable transaction processing with PostgreSQL compatibility, offered across managed and self-hosted deployment patterns. In September 2026 the company extended that proposition with Cockroach Continuum, an agentic database-cloud layer intended to pool capacity and coordinate operations across a database estate. PostgreSQL's position as the most-used database in Stack Overflow's 2024 developer survey supports the familiarity of the interface, but does not establish CockroachDB adoption. The product statements remain company-issued positioning, not independently measured performance or customer economics.[CO001, CO002, CO006, CO015]

1.2 Leadership, founders, and governance visibility

The founding team is reported as Spencer Kimball, Peter Mattis, and Benjamin Darnell, all former Google engineers. A September 2026 company release names Kimball as co-founder and CEO, while Tracxn lists Nathan Rodriguez as Co-CEO and Darnell as co-founder and Chief Architect. These records show founder continuity across executive and technical leadership, but they do not explain the practical division of authority implied by the CEO and Co-CEO titles. Nor do they establish current duties for co-founder Peter Mattis. No retained source provides the current board roster, voting rights, investor protective provisions, succession plan, or a complete executive roster, so governance and key-person dependence remain diligence items rather than settled facts.[CO003, CO004, CO005]

1.3 Capital history and observable scale

The strongest financing anchor is the December 2021 Series F: Cockroach Labs announced $278 million of new funding at a $5 billion valuation and said the round brought lifetime funding to $633 million. Tracxn reports the same lifetime total across seven rounds, while Latka totals $579.6 million across only four rounds beginning in 2019; the different period and round denominator explains why the latter is not a substitute for lifetime funding. Scale evidence is weaker. Tracxn reports 706 employees in September 2026, versus Latka's 715 for November 2025. Latka labels $128.3 million as a December 2023 revenue estimate sourced to Growjo and explicitly says the company has not published revenue. The company's January 2026 release says ARR grew year over year but supplies no amount, and the last retained quantified customer disclosure is more than 200 customers in December 2021. None of these figures establishes current ARR, audited revenue, cash, profitability, or runway.[CO007, CO008, CO009, CO010, CO011, CO012, CO013, CO014]

1.4 Milestones and diligence implications

The public chronology shows a decade-long progression from founding and venture financing into enterprise product expansion: four disclosed rounds from 2019 through 2021, a licensing change in 2024, company-reported commercial momentum entering 2026, and the Continuum launch in September 2026. The chronology also exposes the principal underwriting limitation. The capital valuation is almost five years old at the run date, while the current operating picture relies on company assertions and third-party directory estimates. The 2024 licensing change reportedly prompted open-source community resistance, adding a concrete adoption and ecosystem risk. Investment diligence should therefore prioritize current financial statements, cohort metrics, capitalization and control documents, and customer evidence before treating product momentum as enterprise-value support.[CO007, CO008, CO013, CO015, CO016]

1.5 Exhibits

2.1 Market boundary and substitutes

Cockroach Labs competes in the distributed SQL portion of the broader cloud database market: relational systems that preserve SQL and transactional consistency while distributing storage, compute, and replicas across nodes or regions. The relevant spend includes database software or managed-service consumption, support, implementation, migration, and operating infrastructure for mission-critical transactional workloads. It excludes general analytics warehouses, standalone vector databases, and ordinary single-region relational deployments unless they are being evaluated as substitutes. The practical status quo remains PostgreSQL or another incumbent relational database, often extended with replicas, sharding, failover tooling, and application-layer routing. That boundary matters because broad cloud-database forecasts overstate Cockroach Labs' immediately addressable opportunity, while distributed-SQL estimates include hyperscaler products and larger vendors that compete for the same modernization budget.[CM001, CM002, CM003, CM004]

2.2 Sizing through multiple lenses

Public estimates point to a large but definition-sensitive opportunity. DataIntelo estimates the global distributed SQL database market at $7.2 billion in 2025 and $24.8 billion in 2034, while Coherent Market Insights estimates the broader cloud database market at $26.00 billion in 2026 and $73.03 billion in 2033. Business Research Insights gives a narrower distributed-database estimate of $7.74 billion by 2033 at a 10.5% CAGR. These values should not be added or treated as interchangeable: they use different product boundaries, base years, geographies, and forecast methods. The evidence therefore supports a distributed-SQL beachhead inside a larger cloud-database spend pool, but it does not support a company-specific SAM or SOM. Cockroach Labs' market capture depends on converting workloads where global consistency, availability, data locality, and operational simplicity justify migration cost.[CM005, CM006, CM007, CM008, CM009]

2.3 Buyers, adoption drivers, and constraints

The primary economic buyer is usually an enterprise platform, database, or application-infrastructure owner with a mandate to reduce outage exposure and support growth across regions. Developers and application teams are users; security, compliance, and data-governance teams influence the purchase where residency or regulated workloads matter; finance and procurement own the recurring cloud or enterprise commitment. Adoption is triggered by a new globally distributed application, a legacy single-primary bottleneck, a cloud or hybrid-cloud migration, an AI workload, or a regulatory requirement for locality and resilience. The counterweight is switching cost: schema and feature compatibility testing, topology redesign, workload modeling, replication and egress charges, and the need for distributed-database expertise. Customer examples show that migration can be staged, but public evidence does not establish a universal payback period or buyer-level willingness to pay.[CM010, CM011, CM012, CM013, CM014]

2.4 Market diligence implications

The market case is strongest when the workload cannot be served economically by a simpler single-region relational database and when downtime, inconsistency, or data-residency failures have material business cost. Cockroach Labs can use PostgreSQL compatibility, managed deployment, multi-region operation, and partner distribution to lower adoption friction, but these advantages do not remove the need to prove total cost of ownership. Independent market pages forecast double-digit growth, yet they do not disclose Cockroach Labs' obtainable share, win rate, retention by segment, or the share of database budgets that can move from incumbent platforms. The next diligence step is a workload-level model comparing compute, replication, egress, migration labor, support, and avoided outage cost, segmented by regulated enterprise, high-scale digital product, and ordinary transactional application.[CM015, CM016]

3.1 Competitive landscape and buyer alternatives

The relevant competitive set is broader than other distributed-SQL vendors. PostgreSQL is the default status quo and internal-build substrate, while Amazon Aurora Global Database and Google AlloyDB offer managed PostgreSQL-compatible paths inside hyperscaler estates. Vitess adds application-transparent sharding and near-zero-downtime resharding for MySQL workloads. CockroachDB competes most directly when a buyer needs one logical relational system across regions, writes in multiple locations, and strong transactional semantics. That positioning is attractive for payments, retail, gaming, and other always-on workloads, but it is not a universal replacement for a single-region PostgreSQL deployment or a read-heavy managed service.[CP001, CP002, CP003, CP004]

3.2 Capability, trust, and switching comparison

CockroachDB's core capability edge is architectural rather than a single feature: its official architecture documentation describes symmetric nodes, range replication, quorum writes, and Raft-based consensus, while the comparison material contrasts automatic distribution and geo-partitioning with PostgreSQL's manual sharding and routing burden. PostgreSQL counters with nearly four decades of development, broad extensions, standards conformance, and a large community. The buyer therefore trades operational control and ecosystem breadth for integrated multi-region behavior. Compatibility lowers migration friction, but it does not eliminate testing, operational retraining, or the risk that extensions and workload assumptions differ.[CP005, CP006, CP007, CP008]

3.3 Pricing, packaging, and distribution power

Public evidence does not establish Cockroach Labs' realized contract pricing, discounting, or gross margin, so the comparison must separate public list prices from economic proof. RFP Wiki reports CockroachDB Cloud tiers spanning request-unit and provisioned-vCPU models, while PostgreSQL has no license fee but shifts cost into infrastructure, DBA labor, and separately assembled high-availability tooling. Spanner, Aurora, and AlloyDB monetize managed operations and hyperscaler distribution; AlloyDB publishes starting compute, memory, storage, and network rates, but those figures are not directly comparable to an enterprise CockroachDB contract. The value case is strongest where avoided operational complexity and outage exposure justify higher multi-region baseline spend and migration cost.[CP009, CP010, CP011, CP012]

3.4 Moat durability and displacement risks

The moat is strongest in workloads that require multi-region writes, serializable consistency, locality controls, and failure survival without assembling a distributed-systems toolchain. The assigned comparisons are not independent benchmarks, but they do report recurring buyer concerns about multi-region cost and distributed-system tuning. PostgreSQL's adoption remains a material distribution advantage, with Stack Overflow reporting PostgreSQL as the most-used database in its 2024 survey. DataIntelo places Cockroach Labs alongside Google and AWS in the leading competitive set, underscoring that hyperscaler services can bundle managed operations, AI features, replicas, and migration tooling. Likely displacement therefore comes from narrower fit, price pressure, or buyers accepting active-passive recovery instead of paying for multi-active semantics.[CP012, CP013, CP014, CP015, CP016]

3.5 Exhibits

4.1 Monetization model and revenue quality

Cockroach Labs monetizes a distributed SQL database through managed cloud offerings and self-hosted enterprise software rather than through a single disclosed revenue stream. A 2026 company release says ARR growth was driven by enterprise customer growth and deeper adoption among existing customers, but it does not publish the ARR amount, growth rate, renewal rate, or mix. Historical independent coverage describes Dedicated as a managed cluster, Serverless as consumption-based, and Self-Hosted as a customer-managed deployment. That mix can support expansion from usage into enterprise contracts, but it also makes list pricing, recognized revenue, and gross margin difficult to infer without contracts and a current finance pack.[CI001, CI002, CI003, CI004]

4.2 Pricing, GTM, and sales-efficiency proxies

Public pricing evidence is directional rather than a realized-price dataset. The independent technical-company profile says Dedicated pricing is based on cluster resources and Serverless charges for storage and Request Units; a current cost review reports three tiers, a free Basic tier, and custom pricing, while warning that contract terms and negotiated discounts affect the outcome. The 2026 company release identifies IBM as an OEM partner and describes deeper enterprise adoption, which supports a partner-assisted motion. However, no public source supplies CAC, payback, sales-cycle duration, quota attainment, implementation fees, or channel margin. These are material underwriting gaps, not values to estimate from list prices.[CI005, CI006, CI007, CI008]

4.3 Cost structure and capital adequacy

The product architecture implies a meaningful service-delivery cost base: distributed replication, multi-region operation, support, migration, and cross-region data transfer all sit alongside ordinary cloud infrastructure costs. An independent cost review specifically identifies migration, professional services, and cross-region transfer as costs beyond a license fee. Historical funding coverage reports $633 million of cumulative funding and a $5 billion valuation in the 2021 Series F, while current third-party profile data reports $579.6 million across four rounds and labels revenue and headcount estimates explicitly. These figures are not a current cash balance or runway. No assigned source discloses cash, burn, debt, capex, liquidity preference, or a next-round trigger, so capital adequacy remains unresolved.[CI009, CI010, CI011, CI012, CI013]

4.4 Financial underwriting verdict

The financial case is strongest on revenue-mechanism plausibility and weakest on observability. Cloud adoption, enterprise expansion, and partner distribution are consistent with a scalable recurring model, but the available evidence does not distinguish ARR from recognized revenue or establish gross margin, retention, customer concentration, cash burn, or runway. The 2023 revenue estimate of $128.3 million is explicitly marked as an estimate by a third-party profile and should not be treated as company disclosure; the 2021 valuation is historical and not a current mark. A financing decision should therefore be gated on a current income statement, ARR bridge, cohort retention, cloud unit-cost schedule, cash forecast, and debt and preferred-equity terms.[CI011, CI012, CI013, CI014, CI015]

4.5 Exhibits

5.1 Product and customer workflow

CockroachDB is positioned as a cloud-native distributed SQL database for modern, data-intensive applications. In customer workflow terms, an application connects through PostgreSQL-compatible drivers and tools, writes transactional data to a logical cluster, and can place or replicate data across regions while retaining a SQL interface. The product therefore targets systems of record where correctness, availability, and geographic distribution matter more than a single-node database's simplest operating model. The official FAQ describes transactional and strongly consistent storage, horizontal scale, automatic replication and recovery, while the Red Hat partner brief describes rolling node updates and a single logical database across regions. These are product capabilities, not proof that every workload migrates without code or topology changes.[CE001, CE002, CE003]

5.2 Product lines and maturity

The current product surface has a core database engine, managed cloud deployment options, self-hosted or customer-cloud deployment, and newer estate-level capabilities. The core engine and PostgreSQL-compatible access are documented capabilities; managed Basic, Standard, and Advanced plans are described in the assigned comparison material; and Cockroach Continuum is presented in a 2026 company release as a new database-estate operating model. The distinction matters: a stable database primitive is not equivalent to a generally available AI operations layer or autoscaling roadmap item. The Continuum release explicitly labels auto-scaling compute “Coming soon,” so product maturity should be evaluated by SKU, deployment mode, and availability status rather than by the broad platform narrative.[CE004, CE005, CE006, CE007]

5.3 Architecture and operating model

The architecture combines a transactional key-value substrate with SQL, distributed consensus, replication, and online rebalancing. Official documentation says the system scales horizontally, survives failures with minimal latency disruption and no manual intervention, and supports strongly consistent ACID transactions. The PostgreSQL-compatibility documentation narrows that promise: most PostgreSQL syntax, drivers, and ORMs are supported, but some features behave differently because they are difficult to implement in a distributed system. The assigned comparison material describes automatic range splitting, merging, rebalancing, geo-partitioning, and changefeeds; these features create a powerful operating model but also a larger configuration and cost surface than a single-node deployment. Architecture is therefore a differentiation and an implementation dependency at the same time.[CE008, CE009, CE010, CE011]

5.4 Differentiation, trust, and technical risks

Cockroach Labs differentiates on distributed transactional correctness, survivability, data locality, and PostgreSQL ergonomics. The 2026 release reports CockroachDB 25.2 throughput improvements of up to 50 percent, leader leases, buffered writes, WAL failover, and PostgreSQL-compatible vector indexing through C-SPANN; these are company-issued product claims and should not be treated as independent benchmark results. Independent comparison text reports positive feedback on resilience and compatibility, but also a learning curve for distributed performance tuning, specialist help for complex designs, and gaps in analytics, documentation, or administrative workflows. The status history is only an incident-history surface, not evidence that incidents did or did not occur. Key diligence items are production-specific reliability data, security and compliance attestations by deployment, migration exceptions, realized multi-region cost, and whether newer AI-oriented capabilities are generally available.[CE012, CE013, CE014]

5.5 Exhibits

6.1 Customer segments and buying jobs

CockroachDB is positioned for organizations whose customer-facing or internal systems need transactional consistency across regions, cloud portability, and resilience. The official modernization page describes buyers moving from monoliths to microservices, from VMs to Kubernetes, and toward continuous delivery, while also highlighting regulated data-locality requirements and hybrid or multicloud adoption. The resulting buyer is typically a platform, infrastructure, or application engineering team; the end user is the business application or service relying on always-on transactions. Apps Run The World identifies DoorDash, Shipt, and Salto Systems as users in its public customer database, spanning delivery, logistics, and identity/access-management contexts. These examples support vertical breadth, but the database is an aggregation and does not establish current contract status, spend, or retention.[CU001, CU002, CU003, CU004]

6.2 Adoption proof and trajectory

The strongest public adoption evidence is deployment-specific rather than a disclosed customer-count series. The Mux case describes CockroachDB as the backbone for Mux Video after a move from PostgreSQL, with multi-region support, geo-partitioning, and traffic routing used for performance, data residency, and downtime-risk objectives. The Stake case is more operationally specific: the customer reportedly moved production in about four hours with about a 5% rewrite and uses the database for authentication and a financial ledger across three regions. Cockroach Labs' January 2026 company release says growth came from enterprise customer growth and deeper adoption among existing customers, but it gives no denominator, retention rate, bookings, or customer count. These signals indicate real production adoption and expansion narratives, not a measurable cohort curve.[CU005, CU006, CU007, CU008]

6.3 Named customer proof and durability

Named references are strongest when they identify a production workload and an operational outcome. Mux is described as serving video streaming and analytics, while Stake is described as running authentication and a financial ledger; both are more probative than an unqualified logo. Apps Run The World adds a public customer-list view with DoorDash, Shipt, and Salto Systems, and gives a detailed DoorDash migration narrative involving distributed replication, staged cutovers, and high-throughput transactional services. Independent review aggregation is directionally positive on horizontal scalability, strong consistency, and managed-service reliability, but also flags learning curves, cost sensitivity, and the need for specialist help in complex designs. No assigned source supplies Cockroach Labs' GRR, NRR, churn, renewal rate, contract length, or cohort retention, so durability remains an evidence gap.[CU009, CU011, CU012]

6.4 Expansion, procurement friction, and concentration

The public evidence supports a plausible land-and-expand motion because the company says existing customers adopted more deeply. However, that is a company-issued description rather than independently audited expansion data. Third-party review text describes pricing and cluster sizing as non-trivial, with cost sensitivity for multi-region deployments and specialist support sometimes needed for complex distributed or hybrid designs. That friction can lengthen procurement and implementation, even when the technical use case is strong. The assigned sources do not disclose the top-customer revenue share, partner-sourced revenue, average contract value, implementation conversion, or the number of active production accounts. Concentration and expansion should therefore be treated as unresolved diligence items rather than inferred from named logos or positive adoption claims.[CU007, CU014, CU015, CU016]

6.5 Exhibits

7.1 Risk posture and thesis-break conditions

Cockroach Labs is positioned around workloads where availability, consistency, geographic locality, and regulatory controls matter, but the same positioning creates a demanding operating burden. Its own materials describe multi-region resilience, compliance controls, and continuous product expansion; those are product claims, not independent proof of realized service performance. A January 2026 company release reports throughput improvements and enterprise adoption, while independent profiles rely partly on estimates and historical data. The investment risk is therefore asymmetric: a few large reliability, migration, or pricing failures could damage trust faster than broad market growth can repair it. The thesis breaks if production incidents materially impair customer workloads, if migrations require materially more services than buyers will fund, or if current cash, retention, and concentration data fail diligence.[CR001, CR002, CR003, CR004]

7.2 Regulatory, legal, and data-governance exposure

The public record reviewed here does not establish a current enforcement action, material lawsuit, license dispute, or privacy incident involving Cockroach Labs. That absence is not evidence that no such matter exists. The product is marketed for regulated and geographically distributed workloads, and the company lists GDPR, PDPA, CCPA, and DORA as relevant regulatory drivers. Geographic placement and compliance features can reduce customer exposure, but they also make configuration, data-residency controls, access governance, and evidence of certification commercially important. PostgreSQL remains a widely adopted open-source alternative with its own security and replication model, so Cockroach Labs must prove that its source-available licensing, controls, and support commitments satisfy each customer's jurisdiction and procurement standard. Legal diligence should test certifications, DPAs, subprocessors, BSL obligations, open-source notices, export controls, and unresolved claims rather than infer clearance from marketing pages.[CR005, CR006, CR007, CR008]

7.3 Operational, quality, and security risk

CockroachDB's architecture is intended to survive node, disk, rack, and region failures, and the company says it introduced a benchmark that evaluates partitions, disk stalls, restarts, and zonal outages. These are useful test dimensions, but the cited comparison is company-issued and does not establish production incident frequency, customer-impact duration, or independently reproducible results. Distributed consensus, online schema changes, multi-cloud networking, vector search, and migration tooling expand the surface area that must be operated correctly. CostBench identifies migration, professional services, and cross-region transfer as costs beyond licensing; that supports a practical implementation-risk hypothesis, not a complete cost model. Diligence should obtain incident history, SLO attainment, support escalations, rollback evidence, security attestations, penetration-test summaries, and customer references for the largest deployments.[CR009, CR010, CR011, CR012]

7.4 Partner, dependency, and financial-model risk

Route-to-market and platform dependencies are material because Cockroach Labs sells into an ecosystem dominated by AWS, Google Cloud, Microsoft Azure, PostgreSQL, and large database vendors. The company reports an IBM OEM relationship and marketplace or hybrid-cloud reach, but partner announcements do not quantify sourced pipeline, renewal economics, or dependency concentration. Independent public profiles estimate revenue and report historical funding and headcount, while explicitly noting that the company has not published revenue; those figures cannot establish ARR quality, cash runway, burn, gross margin, or customer concentration. Pricing sensitivity is also plausible: the product's value depends on reducing downtime and operational burden, yet cross-region transfer, migration, support, and implementation costs can change total cost of ownership. The decisive diligence asks are current cash and burn, renewal and expansion by cohort, top-account exposure, cloud and partner terms, and downside financing capacity.[CR002, CR006, CR007, CR010, CR012, CR013, CR014, CR015, CR016]

7.5 Exhibits

8.1 Investment thesis and anti-thesis

Cockroach Labs has a credible infrastructure thesis: CockroachDB combines PostgreSQL compatibility, ACID transactions, horizontal scale, multi-region resilience, and managed or self-managed deployment options. Its 2026 company release says ARR grew year over year, enterprise customers expanded, and existing customers adopted more deeply, while the product pages emphasize uptime, data locality, security controls, and deployment across AWS, Google Cloud, and Azure. These are relevant value drivers for mission-critical workloads because migration costs and reliability requirements can make a successful deployment sticky. The product thesis is therefore stronger than a generic database software pitch, but the cited growth statements are company-issued and do not disclose ARR dollars, retention, gross margin, or profitability. The anti-thesis is valuation opacity and competition. An independent IPO overview reports that the company remains private, has no public S-1, ticker, or price range, and was last reported at roughly $5 billion in a 2021 financing; a separate company-data page reports a $5 billion Series F valuation and a $128.3 million 2023 revenue estimate, while explicitly saying the company has not published its revenue. Those figures are not a current audited basis for entry. Independent comparison material also flags cost sensitivity, complexity, and the need for workload-specific total-cost modeling. Without a current price and financial denominator, a buy recommendation would imply precision the evidence cannot support.[CV001, CV002, CV003, CV004, CV005]

8.2 Financing context, entry discipline, and comparable signals

The financing record is useful as a historical reference, not a mark-to-market valuation. GetLatka reports $579.6 million raised across four rounds, including a $278 million Series F in December 2021 at a $5 billion valuation, and labels the revenue figure an estimate. The independent IPO overview instead reports approximately $630 million or more of total funding and describes the $5 billion figure as a 2021 private mark. The difference is a provenance and scope discrepancy, so the chapter retains both as reported rather than selecting a false single total. No assigned source establishes current preference stacks, liquidation preferences, secondary clearing prices, debt, cash, or dilution. Public comparables can frame diligence but cannot substitute for Cockroach Labs financials. A market-cap source places MongoDB at $29.89 billion on October 24, 2026, after a 15.73% decline in its reported 2026 market cap, while the IPO overview identifies MongoDB as the closest modern database comparable and names Oracle, Snowflake, Amazon, Microsoft, and Google Spanner as broader peers. Those market caps are not revenue multiples for Cockroach Labs. Entry discipline should therefore require a current financing or secondary price, audited or diligence-grade revenue and ARR, growth, gross margin, retention, cash burn, and the terms that determine investor downside.[CV006, CV007, CV008, CV009, CV010]

8.3 Scenarios, recommendation, and thesis-break triggers

The scenario range is deliberately evidence-limited. The bull case requires CockroachDB Cloud and the newly introduced Cockroach Continuum model to convert technical differentiation into durable, expanding, high-margin recurring revenue. The company says Continuum pools capacity, prices work rather than peak capacity, and adds database virtualization, elastic storage, agent-assisted operations, and consumption billing; these are product assertions, not proof of unit economics. The base case is continued adoption with meaningful infrastructure and migration costs, but unclear profitability and a valuation that must be underwritten from new private-market evidence. The bear case is multiple compression or a down-round if cloud costs, multi-region replication, migration complexity, incumbent bundling, or weak conversion from adoption to paid expansion outweigh the product moat. The recommendation is track, with research-more confidence, high risk, and an unknown valuation stance. The stance is not a judgment that the product lacks value; it is a judgment that the public record does not support calling the current entry price attractive, fair, stretched, or expensive. A thesis break would be evidence of materially slowing ARR growth, weak net retention, margin dilution from managed-service costs, a price below the prior private mark without compensating operating improvement, or customer losses caused by cost and complexity. A positive re-rating would require a current price, reliable financial statements, retention and margin evidence, and terms showing acceptable dilution and downside protection.[CV011, CV012, CV013, CV014, CV015]

8.4 Exhibits