BriefCatch

What it is

BriefCatch is an editing engine for legal writing. It runs inside Microsoft Word and Outlook, and since September 2026 as a standalone web app. It reads a brief, motion, memo, or letter and returns suggestions with a plain-English explanation and a before-and-after example. It does not write the brief for you. Ross Guberman built it in 2017. He is the author of Point Made and has trained new federal judges on writing, and the rule set comes from his curriculum. Third-party listings cite more than 10,000 rules and algorithms.

The September 9, 2026 launch turned it into a three-part suite:

• Writing Assistant — suggestions scored across five dimensions: engagement, concision, flow, punch, and clarity.
• CiteCheck — Bluebook citation correction, now included on every plan.
• RealityCheck — reads the cases you cite and checks whether each opinion supports the quotation, holding, disposition, or proposition you attach to it. It flags altered quotes, misstated holdings, reversals, concurrences cited as the majority, and wrong pincites. It also scans the file for hidden text: white-on-white type, hidden formatting, and zero-width characters. BriefCatch says it checks against three case-law sources and marks the cases it could not confirm for manual review.

On February 12, 2026 BriefCatch bought WordRake’s core product and its 12 US editing patents in an asset purchase. That adds WordRake’s sentence-level concision engine to BriefCatch’s document-level analysis. WordRake still runs as a separate product, and existing licenses are honored until the two are merged. The deal followed a $6M Series A led by Full In, announced December 29, 2025. In June 2026 BriefCatch named Josephine Petrick, a former Fifth Circuit clerk and appellate partner, as Chief of AI Products. The company reports 300+ law firms, 45+ AmLaw 200 firms, 80+ courts, and SOC 2 Type II. It states that document text is not retained or used for training.

Why legal ops teams look at it

The question it answers is “BriefCatch vs Clearbrief vs Definely: which one makes the brief better, not just checks it?” Clearbrief ties your factual assertions to the record. Definely checks definitions and cross-references in contracts. BriefCatch is the one that edits the argument itself: structure, emphasis, and sentence length. Since RealityCheck launched, it also catches the fabricated or misread authorities that have led courts to sanction lawyers who filed AI-drafted briefs.

Specific use case: a 40-lawyer litigation group makes BriefCatch a filing gate. Before any brief goes to a partner, the associate runs Writing Assistant, then CiteCheck, then RealityCheck. The associate clears every RealityCheck flag or documents why it can stay. Partner review time then goes to the argument, not to typos and citation form.

Pricing reality

BriefCatch publishes prices, which is rare in legal AI. All plans are billed annually:

• Individual — $60 per seat per month, one-seat minimum, 650 AI credits per month. Includes the web app, the Word add-in, RealityCheck, and Google Drive and OneDrive import.
• Pro — $75 per seat per month, five-seat minimum, 700 credits. Adds the Outlook add-in and the full Learning Hub (90 self-paced modules).
• Enterprise — custom pricing, 20+ seats. Adds SSO, SCIM, DMS integrations, LMS integration, a 99.9% uptime SLA, and a dedicated CSM.

At list price, a solo pays $720 a year, and the smallest Pro team pays $4,500 a year. A 40-lawyer group on Pro would pay $36,000 at list. At that size, ask for an Enterprise quote, because it adds SSO and DMS integration. Unused monthly credits do not roll over; extra credit packs are sold separately. There is a 7-day free trial with no card required. For reference, PerfectIt lists at $70 per user per year, which shows how much of BriefCatch’s price is for RealityCheck and the style engine rather than proofreading.

Best for

Litigation and appellate practice leaders, and the legal ops or KM staff who support them, at firms and in-house teams that file substantive briefs every week and want one tool that improves the writing and screens for AI-hallucinated authority before filing. Courts and chambers are a natural second audience.

Alternatives and when to pick them

• PerfectIt — the other long-standing standalone editor. Pick it for consistency and proofreading checks (defined terms, capitalization, house style) across mixed document types, at a fraction of the price.
• Clearbrief — the fastest-growing entrant in brief verification. Pick it when the risk is facts that are unsupported by the record, not weak prose. Clearbrief links each assertion to the deposition or exhibit behind it.
• Definely — pick it for transactional drafting, where the problem is definitions and cross-references, not persuasion.
• Harvey or Legora — pick one of these when you want AI to draft the first version. BriefCatch still sits well downstream of either one as the editor and verifier.
• Your citator, Thomson Reuters CoCounsel or LexisNexis Protégé — keep KeyCite or Shepard’s for whether a case is still good law. RealityCheck checks whether the case says what you claim it says. You need both checks.

Watch-outs

• RealityCheck is a second check, not a citator. BriefCatch does not publish the full jurisdiction coverage of its three case-law sources, and it marks some cases for manual review. Guard: keep KeyCite or Shepard’s in the filing checklist, get the written coverage list before rollout, and have the filing attorney sign off on every flagged or unverified authority.
• Credits cap the verification you can run. The 650-700 monthly credits do not roll over, and BriefCatch does not publish how many credits one RealityCheck run uses. Guard: during the trial, record how many credits each RealityCheck run on a real brief uses, then size the plan or credit packs to your monthly filing volume before you sign an annual contract.
• WordRake integration is not finished. The two products still run separately. Guard: if you already license WordRake, ask for a co-termed migration or credit in writing so you do not pay for both editors after the merge.
• The style is opinionated. The rule set reflects one expert’s view of good legal writing, and junior lawyers tend to accept suggestions without judging them. Guard: have a senior writer mark which rule categories the group follows, and teach associates the reasons through the Learning Hub instead of letting them accept suggestions in bulk.
• Thin independent review record. Capterra showed zero user reviews in September 2026. Guard: run the 7-day trial on three recently filed briefs and compare what it flags with what partner review actually caught.

Background: retrieval-augmented generation.