Daloopa: Initiation of Coverage

GAII Intelligence defines the AI Research & Intelligence market as software that applies generative and agentic AI to the research process of professional investors: locating candidate securities, assembling and interrogating evidence, producing written analysis with traceable sources, and monitoring positions once held. Buyers are hedge funds, long-only managers, multi-manager platforms, family offices and allocators; the economic buyer is typically the chief technology officer, the head of research, or a portfolio manager with a technology budget.

Capabilities we treat as mandatory

• Evidence retrieval at scale: the ability to search and reason across regulatory filings, transcripts and news, and across documents the client supplies.
• Source traceability: every material claim in an output must be reachable back to the passage that supports it.
• Institutional data handling: a stated position on where client data goes, how long it is held, and whether it is used for model training.

Capabilities we treat as differentiating

• Persistent state: the system remembers a thesis, a coverage list or a position between sessions and acts on it without being re-prompted.
• Workflow surfaces: delivery inside the applications analysts already use — the spreadsheet, the research management system, email and messaging.
• Owned or exclusive content: material that a competitor cannot obtain by signing the same third-party contract.
• Published evaluation: a measured, disclosed accuracy methodology rather than an assertion of quality.

Daloopa is a data supplier rather than a research platform, and several companies in our coverage — Hebbia, Rogo and Samaya AI among them — consume its data. Its direct competitors are Canalyst, now within AlphaSense, and the fundamentals offerings of FactSet and S&P Global, which sit alongside it as data partners in Claude for Financial Services.

The following are the capabilities and characteristics that, in our assessment, most distinguish Daloopa within the AI Research & Intelligence market. Each is drawn from publicly documented product capability, published interviews with management, or independent hands-on review.

• Source auditability: Every data point is hyperlinked to the filing, presentation, release or transcript it came from, which suits model audit, compliance review and grounding AI agents.
• Distribution through AI platforms: Launch data partner for Claude for Financial Services, with MCP connectors for ChatGPT, Microsoft 365 Copilot, Gemini Enterprise and Perplexity, and data inside Hebbia and Rogo.
• Excel-native model updates: The add-in updates the analyst's own model at earnings while preserving format, and Retrofitter audits historical inputs against source.
• Coverage and depth: More than 6,000 companies with 10–14 years of history and deep KPI and segment detail, refreshed within about 90 minutes of earnings by the company's account.
• Rerunnable benchmark: The retrieval benchmark dataset is published on Hugging Face, so a buyer can test the claims directly.
• Fresh capital and growth: A $47m Series C led by Brighton Park in May 2026 brought total funding to about $100m, and the company reports revenue doubling year on year.
• Accessible commercial model: A free tier, per-company subscriptions and AWS private offers make it straightforward to trial on a firm's own universe.

The following are factors that prospective clients should weigh, and areas where our public review did not find sufficient evidence to reach a settled view. Several are common across vendors at this stage of the market and are not specific to Daloopa. We encourage prospective clients to raise each of them directly with the company; we have done the same, and will update this note as answers arrive.

• Security attestation: We did not locate a SOC 2 or ISO 27001 attestation; the trust centre does not name a framework. Prospective clients should request the report, its type and period, and the subprocessor list.
• Identity integration: Single sign-on supports OpenID Connect through Microsoft Entra only; SAML 2.0 is not supported. Firms using Okta or Ping should confirm a path.
• Hosting and retention: Hosting is multi-tenant AWS with no dedicated-tenant option. Scout retains prompts, responses and workbook context for fourteen days, sends them to Anthropic, and offers no zero-retention option.
• Accuracy evidence: Accuracy is stated as above 99% in one place and 99.99% in another, without method or third-party audit, and the benchmarks are vendor-run. We encourage buyers to test a sample against source.
• Coverage claims: Company counts have moved from about 4,000 to 6,000+ across materials, and benchmark accuracy was lower for non-US companies. Buyers should validate coverage of their own universe.
• Operational model: A large share of staff appears to be in India, where analysts historically validated data. The current human-verification process is not documented publicly; buyers should ask how quality is maintained at peak earnings.
• Pricing: Paid tiers carry no published prices, entitlements vary by company count and credits, and the free tier was tightened in September 2026.
• Named clients: Preqin is the only client named publicly; bank and fund case studies are anonymised.

Daloopa Inc. was founded in New York in 2019 by Thomas Li, formerly a TMT analyst at Point72 and KCL Capital, Jeremy Huang, formerly an engineer at Airbnb, and Daniel Chen, formerly a distributed-systems researcher at Microsoft. It is headquartered at 122 East 42nd Street, New York, and operates a large data and engineering team through Daloopa Private Limited in Noida, India.

Daloopa has raised about $100m: seed funding led by Nexus Venture Partners, a $20m Series A led by Credit Suisse NEXT Investors (2021), an $18m Series B led by Touring Capital with Morgan Stanley (2024), a $13m strategic round (2025) and a $47m Series C led by Brighton Park Capital (May 2026). Phil Hadley, former CEO of FactSet, advises Brighton Park.

Company-reported figures are shown as such and are not independently verified. Verification is carried into the vendor questionnaire.

Daloopa delivers one dataset through several surfaces: Excel for analysts, MCP and API for AI platforms, and cloud delivery for data teams. Figure 2 sets out our assessment of how it maps onto the investment workflow.

Data Sheets and Hub

Historical spreads downloadable into Excel, with every figure linked to source.

Excel add-in and Scout

Updates the analyst's model at earnings and audits inputs; Scout, generally available since September 2026, builds and compares models with custom skills.

MCP server and API

Ten MCP tools for fundamentals, documents, prices and skills across about twenty AI clients; Fundamentals and Partner APIs.

Cloud delivery

Snowflake, Databricks and AWS S3 delivery, and AWS Marketplace private offers.

Scope of the product today

Daloopa does not offer research discovery, screening, broker research or consensus estimates. Its role in an AI research stack is the verified data layer that other tools call.

Daloopa extracts data with a collection of small, specialised models rather than general-purpose language models, and links every point to source. Scout, its Excel assistant, runs on Anthropic Claude but maps reported values with fixed rules rather than letting a language model place each figure.

Daloopa publishes its retrieval benchmark openly: 91.6% exact-match accuracy for Claude with the Daloopa MCP against 31.4% without it. The dataset can be rerun, although the page contains small internal inconsistencies, and the current human-verification process is not described.

Coverage spans more than 6,000 companies with 10–14 years of history, extracted from filings, presentations, press releases and transcripts, with KPI and segment depth the company puts at four to ten times other providers. Expert-call transcripts and Heron AI interviews are available through the MCP server.

Our Data & Content dimension weighs expert-network, broker and consensus content that a data vendor does not sell. We score Daloopa on what its data enables in the research workflow; the internal framework note on this point is under review for the next cycle.

Daloopa fits the most time-consuming part of fundamental equity work: updating models at earnings. The add-in updates the analyst's own spreadsheet in place, and MCP brings the same data into whichever AI assistant the firm uses.

It is not a research-discovery or monitoring tool. Idea generation and thesis tracking sit with the platforms that consume its data.

Daloopa documents more of its security architecture than many peers — TLS 1.2 and AES-256, logical tenant isolation on AWS, SBOMs on request, SSO-only mode and per-company entitlements — but we did not locate a named security framework or attestation. Figure 3 sets out what our public review located.

We would encourage any firm evaluating Daloopa to request the SOC 2 report or equivalent, the subprocessor list and penetration-test summary, a SAML path for non-Entra identity providers, and a zero-retention option for Scout.

The offsetting factors are real. Most usage is structured public-company data rather than client content, entitlements are enforced per company, and fresh Series C capital supports continued investment in controls.

Daloopa publishes its tiers — Free, Core, Premium and Fundamentals API — but not prices. Pricing combines per-company subscriptions with credit-metered Scout usage and monthly MCP limits, and AWS private offers run on twelve-month contracts.

The value case is time saved on model updates — 70% less time to build new models and about two hours per ticker at earnings, by the company's account. The free tier makes a first test straightforward.

Daloopa sits in the Established Solutions quadrant of the GAII Intelligence Matrix: high investment utility from data and workflow, with market strength held back by enterprise evidence. It ranks third in the inaugural cohort.

GAII initiation universe, 20 October 2026. Scores as currently published on the GAII Intelligence page; preliminary and based on public information.

11.1 Where Daloopa is strongest relative to peers

• Against incumbent data vendors, on source links for every data point and distribution into AI assistants.
• Against research platforms, on the accuracy and depth of the underlying fundamentals they consume.
• Against manual model maintenance, on in-place Excel updates within hours of earnings.

11.2 Areas of narrower functional focus

Daloopa has built depth in one data layer rather than breadth across the research workflow.

• Workflow: no research discovery, screening or monitoring surface.
• Content: no broker research or consensus estimates.
• Enterprise: no attestation located; SSO limited to Entra; multi-tenant only.
• Evidence: one named client; accuracy self-reported.

Scores are on a 0–10 scale. Each dimension score is the mean of its sub-criteria, and the GAII Overall Score is the weighted mean of the six dimensions. The detailed sub-criteria scores that build each dimension are set out in Appendix A.

These are the areas where our public review did not find sufficient evidence to form a settled view. We have put them to the company and will update this note when they are answered. Prospective clients may find them a useful agenda for their own evaluation call.

• Whether a SOC 2 or ISO 27001 attestation exists, with type, period and auditor.
• The subprocessor list and penetration-test summary.
• Plans for SAML 2.0 support.
• Whether a zero-retention option for Scout is planned.
• Whether a dedicated-tenant option is available.
• The current human-verification process and staffing at peak earnings.
• The method behind the stated accuracy figures.
• Coverage by country and depth outside the US.
• Pricing for Core, Premium and API tiers, and credit allowances.
• Named client references among hedge funds, mutual funds and banks.
• Company-stated ARR or revenue scale.
• Commercial terms for using Daloopa data inside third-party AI platforms.

This note is an assessment of a vendor against a defined framework, not a recommendation to buy or to decline. The following are the considerations we would put at the centre of an evaluation of Daloopa specifically.

If you are a fundamental equity manager

• Test the add-in on your own models at the next earnings cycle.
• Validate coverage and accuracy on your universe, especially non-US names.
• Use the MCP connector inside your existing AI assistant before buying another platform.

If you are responsible for technology or vendor risk

• Open with Figure 3 and request the trust-centre documents.
• Confirm an SSO path if you do not use Microsoft Entra.
• Assess Scout's fourteen-day retention against MNPI policies.

If you are assessing the category rather than the company

• Verified, source-linked data is becoming the grounding layer for every AI research tool.
• Data vendors now distribute through AI assistants rather than terminals.
• Frameworks built for research platforms need adjusting for data vendors.

Each dimension score in Section 12 is the mean of the sub-criteria below.