Casepoint

What it is

Casepoint is a cloud eDiscovery and data-discovery platform for organizations where the security review is the hard part of the purchase. It covers the full lifecycle — legal hold, in-place preservation and collection, processing, AI-assisted review, and production — and adds FOIA and investigations workflows that most eDiscovery vendors leave alone. The platform runs at FedRAMP High with DoD Impact Level 5 and Impact Level 6 authorizations. Casepoint states it is one of only seven SaaS providers worldwide cleared at IL6, which is the claim the entire buying case rests on for defense and intelligence work.

It is also a materially different company than older reviews describe. In January 2025 Casepoint merged with OPEXUS — the government process-management vendor behind FOIAXpress, ATIPXpress, and eCASE — under a majority investment from Thoma Bravo, and it added the mLINQS compliance product in July 2025. The combined business now sells under the Casepoint name, with the former OPEXUS products listed alongside Casepoint eDiscovery, Legal Hold, FOIA, Filestore, and Investigator. Casepoint reports serving more than 200 corporates and enterprises and, on the public-sector side, over 150 federal civilian, defense, and SLED customers.

Why it shows up in legal-ops stacks

• The authorization is the differentiator. For a federal agency or a contractor handling classified-adjacent data, FedRAMP High plus IL5 and IL6 eliminates most of the market before features are discussed. Relativity and Everlaw both serve government, but the IL6 ceiling is where the shortlist gets very short.
• FOIA and congressional inquiry sit in the same platform as litigation. This is the scoped use case Casepoint wins: an agency or regulated enterprise consolidating litigation eDiscovery, FOIA response, and internal investigations onto one system of record rather than running a separate FOIA tool with its own custodian list. The May 2026 release added an AI Reading Room Assistant, an AI Intake Assistant, and a patent-pending AI Request Writing Assistant to that side of the product.
• CaseAssist active learning with no seed set. Reviewers can train one or several models without assembling a sample set first, which matters on matters that start with no known-relevant exemplars. Natural-language search, document summarization, and automatic document classification round out the GenAI layer, over 600 file types, with connectors for Microsoft 365, Teams, and Slack.
• An MCP server that inherits the user’s permissions. Shipped 30 July 2026, covering eDiscovery, Legal Hold, and FOIA. It exposes review batch status, reviewer productivity, custodian reports, legal hold status, and FOIA request and annual-report data, and every request executes under the authenticated user’s existing role-based permissions. That is the right design for regulated work, and it is deliberately positioned against platforms that bind customers to one AI vendor.

Pricing

Casepoint publishes no list price. The pricing page markets an all-inclusive posture — “If you don’t use a product, you don’t pay for it” — with AI and analytics bundled rather than metered as add-ons, which is the opposite of Relativity’s approach of charging for aiR consumption on top of hosting. Useful, but not a number.

Federal award records are the honest anchor, because Casepoint sells heavily through GSA and those obligations are public. Recent examples from USAspending: the Department of Veterans Affairs awarded $3,828,000 for an explicit 12-month eDiscovery SaaS and maintenance contract running August 2026 to August 2027; the SEC’s Electronic Discovery Application and Storage Environment ran $46,060,473 across two years; DHS/ICE obligated $6,792,363.73 over roughly three years. At the small end, the Department of Agriculture has bought the same product on consecutive one-year awards at $95,660, $125,660, $161,660, and $169,430.

Read that as a working band: a small agency or single-department deployment lands near $100,000–$170,000 a year, a mid-size one in the $400,000–$2.5M range, and only the largest litigation programs reach eight figures. Corporate buyers should expect the low end of that spread, since federal awards bundle support and compliance obligations a commercial contract does not carry.

Best for

Government legal teams and heavily regulated enterprises — agency counsel, defense contractors, and financial-services litigation groups — that need one platform for litigation eDiscovery, FOIA, and investigations and cannot clear a vendor below FedRAMP High.

Watch-outs

• Renewal escalation is visible in the public record. That USDA series is the same product at the same agency rising from $95,660 to $169,430 across four consecutive annual awards. Some of that is data growth rather than rate increases, but you cannot tell which from the outside. Guard: fix per-GB rates in the base contract, negotiate priced option years up front, and cap the annual escalator before signing — not at first renewal.
• The independent review base is thin. G2 shows 4.7 out of 5 from roughly 25 reviews, which is too small a sample to triangulate against Relativity’s or Everlaw’s review volume. Guard: require reference customers at your actual data volume and in your sector, and run a paid pilot on one live matter before committing the portfolio.
• The merger pulled center of gravity toward government. The OPEXUS combination and the mLINQS purchase both bought public-sector workflow, and the flagship 2026 release was FOIA. That is good news for agencies and a question for corporate buyers. Guard: ask for the commercial roadmap specifically, and count how many of the last four quarterly releases shipped capability for corporate legal rather than FOIA.
• The MCP server is read-oriented at launch. It returns status and reporting data; it is not a write path into review decisions, and there is no public developer portal for the underlying APIs. Guard: scope agent workflows to reporting and triage, and get the write-capability timeline in writing before designing around it.

Alternatives worth pricing against: Relativity when outside counsel already runs your matters there and being on the same platform removes a handoff; Everlaw when you want stronger cloud-native review and trial preparation and do not need authorization above FedRAMP Moderate; DISCO when per-matter economics and speed of setup matter more than the security ceiling; Logikcull for self-serve mid-market matters that never justify an enterprise deployment; and Microsoft Purview eDiscovery when the data never leaves Microsoft 365 and the matters are internal.