Use casesFour workflows. One methodology.
Celadon doesn't replace your research stack. It adds the layer that scores the sources, stress-tests the thesis, and documents the reasoning — the layer that doesn't exist in any tool you currently own.
Your analyst has a view. Test it before the IC does.
Morningstar says Salesforce is a buy. Your analyst agrees. Before the IC presentation, run the thesis through Celadon. The pipeline scores the evidence base, searches for the strongest counter-evidence, rates conviction across four dimensions, and identifies the specific conditions under which the thesis breaks. The IC gets a memo with the thesis and its stress test. The analyst looks thorough. The PM makes a better decision.
“Is Morningstar’s wide-moat rating on CrowdStrike sustainable given the Falcon platform’s pricing pressure from Microsoft Defender?”
Source hierarchy table
Counter-thesis rated Decisive/Material/Moderate/Weak
4-dimension confidence assessment
Monitoring variables with quantified triggers
4,000+ public companies have no Morningstar coverage. Celadon analyzes any entity with public information.
Your coverage universe has gaps. The mid-cap industrial your PM is interested in has no sell-side coverage and no Morningstar rating. Celadon produces a structured analysis from public filings, industry data, and institutional research available online — scored, contradiction-tested, and uncertainty-decomposed — in thirty minutes. It’s not a Morningstar report. It’s the analysis that doesn’t exist anywhere else.
“What is the competitive position and regulatory risk profile of Axon Enterprise in the law enforcement technology market?”
Full pipeline on any public entity
Source scoring from SEC filings and public data
Comparable format across entities
Document the analytical process, not just the conclusion.
Investment committees, compliance functions, and LP reports require documentation of how research conclusions were reached — not just what the conclusions are. Celadon’s output is an auditable artifact: every claim traced to a source, every source scored, every contradiction surfaced, every confidence dimension rated. Export the evidence trail as a PDF appendix or JSON payload. When the auditor asks ‘how did you arrive at this conclusion,’ the answer is the report itself.
“What are the material risks in acquiring a SaaS company with 80% revenue concentration in financial services?”
Full evidence provenance
Exportable audit trail (PDF/JSON)
Source scoring transparent in report
30 structurally comparable analyses. Same methodology. Same scoring. Cross-entity contradiction detection.
A PE firm screening 15 public safety tech companies for a sector thesis needs comparable analyses — same structure, same evidence standards, same adversarial rigor. No analyst team produces 15 comparable reports. No AI chat tool produces structurally consistent output across entities. Celadon’s portfolio-scale mode runs the full pipeline on every entity, shares evidence across reports, and surfaces cross-entity contradictions. The output is a sector package, not 15 independent documents.
“Compare the competitive positioning, regulatory exposure, and financial trajectory of the top 15 public safety technology companies.”
Batch analysis with shared evidence pool
Cross-entity contradiction detection
Structurally comparable output
Sector-level synthesis
Report types by tier
| Free | Professional | Team | Enterprise | |
|---|---|---|---|---|
| Market Research | ✓ | ✓ | ✓ | ✓ |
| Company Analysis | ✓ | ✓ | ✓ | ✓ |
| Topic Deep-Dive | ✓ | ✓ | ✓ | ✓ |
| Investment Thesis | — | ✓ | ✓ | ✓ |
| Competitive Intel | — | ✓ | ✓ | ✓ |
| Due Diligence Pack | — | — | — | ✓ |
| Regulatory Pack | — | — | — | ✓ |
Consumer AI tools can waive your privilege
In February 2026, Judge Rakoff ruled in United States v. Heppner that 31 documents generated using a consumer AI tool were not protected by attorney-client privilege. The reasoning: no attorney-client relationship exists with an AI provider, and the provider's privacy policy destroyed confidentiality expectations.
For deal teams running due diligence on privileged materials, this creates three tiers of risk. Consumer AI tools offer no privilege protection. Enterprise vendor tools used at the direction of counsel offer partial protection, subject to the vendor's data handling practices. In-house deployment — where data never leaves the firm's own infrastructure — offers the strongest privilege position.
Celadon's Enterprise tier deploys in your own environment. Research questions, source materials, and generated reports never leave your network. No third-party provider processes your privileged data. The pipeline runs entirely within your infrastructure, and reports are stored in your own database. This is not a privacy policy promise. It is an architectural fact.
“Courts will find that using consumer AI tools may waive the attorney-client privilege, while enterprise AI tools used at the direction of counsel offer more protection.”
— Morgan Lewis, February 2026
This matters beyond legal teams. Any professional handling confidential client materials — financial advisors with portfolio details, consultants with strategic plans, analysts with pre-announcement data — faces the same privilege and confidentiality calculus. The architecture is the protection, not the privacy policy.