Databento

Company sizes or stages where this product is unnecessarily complex or insufficient

• Solo traders or early-stage users who only need standard OHLCV bars and limited historical depth may find lower-cost alternatives more appropriate.

Industries or use cases where competitors are more appropriate

• For workflows that are primarily charting and indicator-based, cheaper bar-centric sources are typically more suitable.

Technical requirements that this product may not meet

• Buyers requiring full coverage across specific APAC venues immediately may need specialist providers or direct exchange arrangements; coverage varies by venue and is expanding over time.

Budget constraints that make alternatives more appropriate

• Organizations that cannot accommodate exchange license fees for real-time equities/options should consider broker-derived data, charting platforms, or limited-scope APIs.

Specific workflows or integrations where the product may be a poor fit

• Stacks built primarily in languages outside Python/C++/Rust where teams do not want to implement a socket protocol may face integration friction.

Red flags in the buying process

• If the vendor cannot provide written total-cost estimates for the exact symbol universe, schema, and retention period (including license fees), cost uncertainty is a material risk.

Current limitations acknowledged by the vendor

• Live gateways are restarted weekly on Sundays, causing client disconnections (documented).
• Mid-week restarts may occur without notice for urgent fixes (documented).
• Connection/session limits per dataset/team are defined and can constrain some architectures.
• Usage-based live pricing has been deprecated for parts of the offering, per pricing documentation.

Features on the roadmap (not yet available)

• The vendor mentions Spot FX as an upcoming data offering; timing is not specified in public pages.

Scale limitations (at what volume/size does it break down?)

• Practical scale constraints include the buyer’s compute and storage costs, exchange license fees, connection limits, and the need for robust ingestion backpressure handling.

Security or compliance gaps

• The absence of publicly documented SOC 2 / ISO 27001 certification pages may be a procurement blocker for some enterprise buyers.

API limitations or integration constraints

• The proprietary Raw API protocol for live data may require additional engineering effort compared with standard HTTP/websocket-based ecosystems.

• Community posts frequently praise responsiveness from support staff in resolving issues despite reports of intermittent outages.
• Users often note the clarity of the schemas for downloading order-book data.

• DBN: a binary encoding and associated tooling for market data intended to streamline ingestion and storage workflows.
• Raw PCAP delivery with hardware timestamps for some feeds, useful for event ordering and latency analysis.
• A self-service exchange licensing workflow that can reduce procurement friction compared with legacy vendors, subject to exchange rules and licensing constraints.

• Community comparisons suggest buyers select Databento when fidelity and detailed market structure access are higher priorities than billing simplicity.
• Competitors are favored when simplicity, flat-rate pricing, or broad packaged datasets are the primary requirements.

Important availability caveat on formal review platforms

• Gartner Peer Insights
  • A product-specific Gartner Peer Insights page for Databento was not found in the accessible sources during this research; therefore no Gartner-verified reviews are available in this report.
• G2
  • The vendor has a presence on G2 with zero user reviews listed on its product/seller page at the time of research.
• TrustRadius
  • No TrustRadius product page with Databento reviews was found via accessible search results.

Community and social feedback (examples)

• Positive feedback
  • Some community posts on Reddit report positive experiences with customer support and successful data downloads for order-book research.
  • Community commenters have noted favorable impressions of the vendor’s schemas and the ability to obtain tick-level and order-book data.
• Critical or mixed feedback
  • Community threads include reports of live streaming reliability issues and intermittent outages.
  • Users have reported data interpretation differences versus other platforms, highlighting the need to select the correct schema and understand normalization conventions.
  • Cost-related comments indicate some users encounter non-trivial data costs for large runs or for multi-instrument L2/L3 workloads.

Why customers chose Databento (themes from community and vendor materials)

• Need for L2/L3 depth and/or longer history than lower-cost aggregated offerings.
• Desire for a unified schema, batch downloads, and a self-service licensing workflow.
• Requirement for tick-level and order book research for microstructure analysis.

Why some customers pick alternatives instead (themes)

• For simple bar-based strategies (hourly/daily), some users find lower-cost providers more appropriate.
• For low-cost level-1 data and limited history, alternatives such as Sierra Chart are often cited as more cost-effective.
• For predictable flat-rate real-time equities needs, other providers (e.g., Polygon/Massive) are commonly considered.

Public case studies (vendor-published)

• The vendor’s blog contains engineering write-ups and integration announcements (for example, a QuantConnect LEAN integration). These materials are vendor-published and describe integration scenarios rather than independently audited ROI case studies.
• The vendor’s strategic investment announcement includes growth metrics and a claim of a large customer selection; the metrics and customer identity are vendor-published and were not independently verified in the sources reviewed.

Quantified outcomes and methodology gaps

• Growth metrics presented by the vendor (e.g., revenue growth multiples and user growth percentages) were provided without auditable methodology, baseline figures, or third-party verification.
• Performance claims for normalization latency and gap rates were not accompanied by third-party benchmarking results in the accessible sources.

Mixed results / projects that struggled

• Community reports of live stream issues and data discrepancies show that real-world integrations can encounter reliability and interpretation challenges. Buyers should plan for engineering effort to validate correctness for their symbol universe and schemas.

• Small quant teams and fintechs: approximately 1–50 employees that require institutional-grade datasets without enterprise procurement.
• Mid-market trading/research organizations: approximately 50–500 employees where multiple researchers require consistent normalized datasets.
• Larger institutions: teams that require exchange-direct feeds but prefer a developer-oriented delivery layer.

• Quant trading and investment research
• Trading technology vendors
• Market microstructure research and execution analytics
• The vendor positions the product for a range of buyers from pre-product startups to investment firms with significant AUM.

• Decision makers
  • Head of Trading, CTO, Head of Quant Research, Data Engineering leadership
• End users
  • Quant researchers, data engineers, trading systems engineers

• Fast access to historical tick/L2/L3 data without negotiating multiple direct vendor contracts.
• Consistent normalized schemas across venues and between live/historical workflows.
• Reduced engineering effort compared with building full capture and normalization infrastructure in-house.

• The organization requires tick-level and/or order book history for backtesting that must align with production microstructure assumptions.
• The buyer has engineering capacity to implement robust streaming ingestion, reconnect logic, and schema management.
• The organization can model data usage costs and limit downloads to required subsets.

• Requirements are limited to daily/hourly bars for a modest symbol set.
• The procurement process requires publicly documented security attestations (e.g., SOC 2/ISO) prior to vendor evaluation.
• The organization requires a purely flat-rate predictable bill with minimal exchange-licensing overhead.

• Databento is a B2B market data provider offering:
  • Historical market data (API + batch downloads)
  • Live/real-time streaming market data (socket-based “Raw API” + client libraries)
  • Reference data (e.g., corporate actions/security master—described by the vendor as coverage across many venues)
  • Raw packet captures (PCAP) for certain feeds (intended for replay and market microstructure analysis)
• The product is presented as institutional-style data capture and normalization (DBN format). The vendor emphasizes tools for developers and a self-service onboarding approach intended to reduce procurement friction compared with traditional feed setups.

• Primary target users
  • Quant researchers, algorithmic traders, trading infrastructure engineers
  • Buy-side firms (from small quant teams to larger funds)
  • Fintech / trading tooling vendors that require historical and live feeds
• Positioning themes (vendor-stated)
  • The vendor describes the API as having a simple surface, a unified message format, alignment with institutional formats, and normalization designed to reduce processing latency.

• Founded: February 2019 (per company timeline).
• HQ: Salt Lake City, Utah (also reflected in third-party listings such as PeeringDB).
• CEO/co-founder mentioned in press: Christina Qi.

• Series A / early funding
  • Databento reports a Series A in November 2021 and states it raised over $27M in initial rounds of funding.
  • A 2024 press release reported an additional $10M closed on Oct 30, 2024, for a Series A total of $30M (vendor-provided announcement).
• Strategic investment (announced Nov 4, 2025)
  • The company announced a strategic investment tied to a large quant hedge fund selecting Databento as its primary options market data provider; the company stated the deal is expected to generate millions in revenue (terms confidential).
  • The announcement included vendor-published metrics (a 4.2× increase in GAAP revenue and 187% user-base growth “over the past year” as of Nov 4, 2025); these metrics were published by the vendor without an independently verifiable methodology in the announcement.
• Series B (announced July 9, 2026)
  • A press release reported a $97M Series B led by NEA, with participation from DRW Venture Capital, Redpoint Ventures, and others.
  • This is a vendor-provided press release; round size and investor participation are useful for sizing but do not validate product performance claims.
• Customer count (indicative)
  • The vendor states that "3,000+" firms and startups use the APIs. This is a vendor claim; no independently audited customer count was found in accessible public sources.

• Historical data
  • Usage-based billing priced by uncompressed binary size in Databento’s DBN encoding (metered per byte/GB).
  • Streaming versus batch download has different billing semantics:
    • Streaming: billed each time the same data is streamed.
    • Batch download: billed once; re-downloads within 30 days are not charged again by the vendor.
  • New account credits
    • Documentation indicates new accounts receive $125 in free data credits that expire after six months (one set per team).
• Live data
  • Live data involves exchange/venue license fees (passed through by the vendor). A payment-processing fee (2.9% via Stripe) is applied to license fees.
  • The vendor indicates that usage-based live data was not available for certain datasets and that usage-based live pricing would be deprecated for some datasets by March 31, 2025.
  • The vendor provides a portal table summarizing 2026 license fee examples by dataset (e.g., OPRA, Nasdaq TotalView-ITCH), which can be used for cost estimation.
• Flat-rate subscription plans
  • The vendor introduced flat-rate subscription plans to complement usage-based pricing; detailed terms and availability for these plans are typically visible only within the vendor portal.
• Practical considerations
  • Total cost depends on dataset selection (e.g., CME vs OPRA vs Nasdaq TotalView), the intended use case (personal vs commercial, display vs non-display, redistribution rights), and the number of users/devices and entitlements.

• Implementation
  • The vendor provides official client libraries for Python, C++, and Rust; other languages can integrate via HTTP API (historical) and the Raw API (live via TCP socket).
  • A Python 3.10+ runtime requirement is noted in quickstart materials.
• Cost drivers / hidden costs
  • Exchange license fees (often the dominant cost for real-time equities/options) plus processing fees.
  • Potential storage and compute costs for high-volume historical tick/L2/L3 datasets (particularly for PCAPs or full-depth feeds).
  • Engineering effort required to handle symbology, rolling futures contracts, corporate action adjustments, and exchange-specific microstructure behaviors.
• Operational overhead
  • Live gateway restarts are scheduled weekly (Sunday) and may also occur mid-week for urgent fixes; customers should implement reconnect and recovery logic.

• Polygon.io / Massive
• AlgoSeek
• Nasdaq Data Link
• Nanex
• Interactive Brokers (as a brokerage-derived data source)
• Sierra Chart