Casefleet vs CounselPro: how they compare in 2026

The AI layer is substantial and prominently positioned, and the vendor's own pricing table shows it is severable. The homepage is labeled AI case management software, and the features menu opens with a group of seven AI entries: Casey AI Assistant, Agentic Workflows, Document Intelligence, Summarization, Entity Extraction, Document Organization and Proposed Facts. The published plan comparison shows the Starter plan at 30 dollars carrying none of it. Summarization, semantic indexing, entity recognition, fact extraction, natural language search, AI issue generation and AI credits are all marked unavailable on Starter and appear only on Advanced AI at 140 dollars. A buyer can therefore purchase a fully working version of the product with no machine learning in it. What remains on Starter is the platform the company has built its name on since 2016: chronologies, timelines, outlines, document review, transcripts and reporting. The AI is a real and heavily marketed layer on a product that stands without it.

The models are the product. CounselPro's AI reads each statement page, extracts and categorizes transactions, re-reads pages that do not reconcile, answers questions about the records in Daystrom chat and drafts a forensic report. Without the models there is little left to sell.

Grounding is built into the product and documented, but nothing measures it. Every fact in the chronology links to its source, and evidence citations come on every plan, including the plan without AI. The assistant returns answers with clickable citations to the user's source documents, and Proposed Facts arrive with a citation to the source of each drafted fact. A user can open and check that linked material, and the retrieval target is the case record rather than published legal authority. The agreement is candid about limits. It states that the AI Features may not be 100 percent accurate and may contain errors, omissions or misinterpretations, and that Casefleet makes no representations or warranties on the accuracy, completeness or reliability of AI output. Beta or preview features are to be used with heightened caution. What is absent is measurement: no accuracy figure, test set, evaluation or error taxonomy, and no retrieval method described beyond the citation link. Grounding to primary legal authority and citation status checking do not apply to a fact management product and are counted neither way.

Grounding and a checking method are described; accuracy is not measured. Every figure links to the statement page it came from, each statement cycle is added up against the beginning and ending balances the bank printed, a cycle that misses is re-read against the page image, and one that still cannot be proved is flagged rather than closed. Daystrom shows the arithmetic behind each answer and leaves unsupported figures open. No error rate, test set or independent evaluation is published.

Proposed Facts are drafted by Casey and staged for the user, who reviews, edits and approves every one before it joins the chronology. The assistant's product page describes Casey setting up cases, proposing facts and organizing files while the user approves every step. The agreement states that AI content and analyses supplement, not replace, professional legal work and human oversight, and that the user is solely responsible for reviewing, validating and verifying any output before relying on it. An activity feed records every change across a case, and the platform applies audit logging. Agentic Workflows, which have their own feature page, are marketed as handing Casey multi step case tasks. The assistant page and the agreement do not state what an agent may do unattended, when it stops, what a user sees while it runs, or what happens after it acts wrongly. No confidence signal is described as shown to the reviewer.

Flagged gaps and a written review duty, short of stated limits. The product flags statement cycles it cannot reconcile, missing months and unsupported figures for the lawyer to weigh, and reports are editable after they are written. The Terms of Service say output is not legal advice and must be reviewed by a qualified lawyer. Nothing states what the forensic report may not be used for without review.

Six customer quotations are published. Three carry a person's name and role but no organization: Jason Shinn and Paul Fleming, both described only as Attorney, and Risa Beck, Paralegal Manager. The other three are attributed by state and role alone: a South Carolina consumer protection lawyer, an Arkansas investigative paralegal and an Illinois criminal law attorney. No firm or legal department is named, no customer logos appear, no case study is published, and no operational figure is offered: no customer count, no adoption number, no time saved, no outcome measured. The company says it has been helping to win cases since 2016. One quotation praises the chat feature as a legal AI tool.

No named customer, testimonial, case study or measured result is published; the site's examples are illustrations.

Substantive published commitments, several of them AI specific, without a statement of how the vendor treats privileged material. The agreement grants Casefleet rights over customer data only to provide the service and expressly for no other purpose. A confidentiality section binds both parties and carries an AI clause. Under it, Casefleet applies technical and organizational measures when data passes through AI Features and guards against commingling confidential data between users and organizations. In the product, administrators control which data each user can see, which the security page describes as keeping privileged and confidential information that way. On account deletion, Casefleet will delete or destroy all copies of customer data unless legally prohibited. Privilege and work product are named in the agreement, but as the user's compliance duty rather than as a statement of how Casefleet handles privileged material. One provision cuts the other way. The warranty section puts sole responsibility for data security and backup on the user, does not guarantee that transmissions to third party vendors will be encrypted, and disclaims liability for unauthorized access to or loss of customer data.

Strong commitments are published, partly in the agreement, with a gap on the vendor's own use of data. The Terms of Service treat customer data as confidential information and bar sharing it with other customers or third parties beyond named subprocessors. They state that, to CounselPro's knowledge, data sent to third party language models is not used to train them. The security page says each matter is processed in isolation, even from other matters in the same firm, that records never train any model, that data stays in the United States and can be deleted at any time, and names every AI provider. The terms also reserve by default the processing of customer data to calibrate CounselPro's internal AI systems for general product improvement, and privilege is addressed only in marketing.

The agreement has a dedicated section on legal advice and the attorney client relationship. It states that Casefleet is not a law firm, that no attorney client relationship arises from use of the platform or its AI Features, and that the software is not a substitute for an attorney's advice. Two of the five numbered acknowledgments that follow address users who are not lawyers: such a user should consult an attorney before making decisions based on the platform, and bears sole responsibility for obtaining counsel. The AI Features section requires users to keep their use within their professional ethical obligations, naming attorney client privilege, the work product doctrine and confidentiality. Casefleet also publishes use case pages for people handling their own matters, such as VA disability claims, small claims, custody documentation and estate settlement. On jurisdiction, the agreement acknowledges only that AI output may not apply in the user's jurisdiction.

A clear advice line in the agreement, short of jurisdiction limits. The Terms of Service say output is not legal advice, that CounselPro does not provide legal advice, and that customers who are not lawyers must have output reviewed by a qualified lawyer; the warranty section repeats the review requirement. The security page says the product is built around ABA Model Rule 1.6. The jurisdictions and uses the product covers or excludes are not stated.

No governance disclosure is published, and the absence should not be mistaken for general silence. The agreement does substantial AI work: it defines AI Features as a contractual term, devotes a section to their limitations, allocates ownership of AI output, and commits on training. All of that is disclosure about the product and about data handling, and none of it is governance. Nothing published describes an AI policy, a responsible AI or ethics position, a bias or fairness assessment, or model evaluation or testing of any kind. No internal owner for AI decisions is named, and no governance framework such as ISO 42001 or the NIST AI Risk Management Framework is cited. Nothing addresses how models are selected, assessed before release or monitored in use, on a product whose agentic features act on litigation case files. Surfaces read: the homepage and full features and use case navigation, the Terms of Service in full, the pricing page and its FAQ, and the security page.

No AI governance framework, accountable owner or testing before release is published beyond the stated principle of flagging what cannot be verified.

Stewardship is contractual and AI specific, with two gaps. The agreement commits that Casefleet will not use customer data to train its AI systems without the customer's explicit consent, and separately describes a closed loop system in which customer data is not used to train AI models. The data license is granted only to provide the service and for no other purpose. AI output derived from customer data belongs to the customer. The confidentiality section adds AI specific technical and organizational measures and a safeguard against commingling data between users and organizations. Security controls are described in operational detail. They include encryption in transit and at rest, a virtual private cloud, virus and malware scanning of all uploads, daily backups, least privilege employee access revoked on termination, and code review before release. The first gap is that AI Features process data through third party AI services that are not named, so the buyer cannot see who is in the loop. The second is that the warranty section does not guarantee that transmissions to third party vendors and hosting partners will be encrypted, and disclaims liability for unauthorized access to customer data.

Most of the ground is published and specific; incident handling is not. The security page describes AES-256 encryption at rest, TLS 1.3 in transit, per-matter isolation, one-time-code sign-in with no passwords, US-only hosting and deletion of any document or matter at any time, and the subprocessor page names every vendor with its role and location and is updated when the list changes. No incident notification commitment is published, and the Privacy Policy, from July 2024, does not state retention periods.

A published agreement with a stated cap and a real indemnity in the customer's favor, and no recourse at all for AI output. The liability cap is quantified: fees actually paid in the six months before the claim, or five hundred dollars where no fees apply, with indirect and consequential damages excluded. Casefleet gives an intellectual property indemnity, defending infringement actions at its own expense, with a right to procure, replace or modify and a pro rata refund if it cannot. Data can be downloaded for 30 days after termination. The AI position is a complete disclaimer. The agreement makes no representations or warranties on the accuracy, completeness or reliability of any output the AI generates, makes the user solely responsible for reviewing and verifying it, and states that no output is legal advice. The service is provided as is with all warranties disclaimed, with no uptime commitment, service credit or cure period, and Casefleet disclaims liability for corruption, deletion or loss of customer data. A firm whose chronology rests on a wrongly extracted fact has no contractual remedy for it, and any claim is capped at six months of fees.

Addressed through a standard limitation clause. The Terms of Service cap CounselPro's liability at the fees paid in the prior twelve months, exclude indirect damages with carve-outs, disclaim warranties of output accuracy with a requirement for qualified lawyer review, and require the customer to indemnify CounselPro for claims arising from customer data and misconduct. Nothing stands behind a figure that proves wrong.

The site has no integrations page, partner listing, or developer or API documentation anywhere in its navigation of features, use cases, pricing, resources and support. No document management, practice management, case management, billing, ediscovery or court filing system is named. The agreement allows for third party applications a customer may use with the platform, whether or not Casefleet designates them official integrations, but names none. It also prohibits users from using any Casefleet APIs that are not public. Single sign on via SAML2 is offered on Enterprise plans, which is authentication rather than practice system integration. The product is built as a self contained workspace.

Statements are uploaded as PDFs and results export as CSV, XLSX, PNG or PDF; no connection to a case management, document management or accounting system is published.

Neither data location nor tenancy is stated. The hosting provider is named plainly: all Casefleet infrastructure and customer data sit on Amazon Web Services, and the security page links to AWS's own security and data center documentation. The architecture is described in some detail. Infrastructure handling user data runs inside a virtual private network that exposes only required components to the public internet, internal and external communication must be encrypted, and daily backups are kept for an extended period. None of that answers either question. No region or data center location is stated, no residency option is offered, and nothing addresses data for customers outside the United States. Nothing says whether the service is single tenant, pooled or configurable, and the anonymization of data by UUID between components is an internal design detail rather than a tenancy model. No self hosted or private deployment option is published. Naming the hosting provider identifies who runs the infrastructure, not where the data rests.

Region and isolation are stated for a single hosted service. The security and subprocessor pages say all client data is hosted in US data centers and never leaves the country, with every subprocessor located in the United States, and that each matter runs in its own isolated space. No other deployment option is offered, and where each model provider processes data beyond the country is not broken down further.

Controls are described in detail and no certification of any kind exists. The security page covers five areas. Application security includes XSS and CSRF protection, signed cookie sessions, rate limiting, audit logging, password hashing and malware scanning. Network and storage security requires TLS internally and externally and encrypts data at rest. Operations covers automated security testing, mandatory peer code review, staging before release, daily backups, and encrypted, firewalled laptops for engineers with production access. Account security offers two factor authentication, SAML2 single sign on for Enterprise, granular permissions and alerts on concurrent logins, and physical security relies on AWS and key card office access. A vulnerability disclosure route is published with an address. No third party attestation exists anywhere: no SOC 2, no ISO 27001, no auditor or certification body named, no penetration test disclosed, and no trust center. The claim that all code, infrastructure and operations are continually audited for security impact is an internal practice statement, not an external audit. HIPAA is handled commercially, with the vendor stating it is HIPAA compatible and will sign a business associate agreement on annual accounts for an extra fee. Nothing unsupported is claimed.

No attestation of its own. The security page says CounselPro runs on providers audited to SOC 2 Type II and states that this is the providers' certification, not CounselPro's. The security page and subprocessor list are public and security questionnaires are welcomed, but no audit report or certification of CounselPro itself is published.

The shape of the supply chain is disclosed and nothing in it is named. The agreement states that AI Features use multiple AI models and technologies from various providers and that specific features may run on different underlying systems. It also states that data may be processed through various AI systems, including third party AI services. A buyer therefore learns that the architecture uses several external providers and that the model behind one feature may differ from the model behind another. What none of it gives is a name. No model, version, family, provider or inference route is identified on any surface read, and there is no subprocessor list, DPA or trust center where one might sit. Amazon Web Services is named as the host for all Casefleet infrastructure, which says where the application runs rather than whose model reads a client's case file. A firm can tell its client only that several unnamed third parties process case material through AI, a disclosure that raises the question it does not answer.

Providers are named; models are not. The subprocessor page names Anthropic, Google, Cohere and DeepInfra as AI model providers, says what each does and that each is in the United States, and says every provider is bound by an agreement that forbids training on customer data; the page is updated when the list changes. Which models are used, and advance notice of a change, are not published.

Pricing is published in fine detail without any sales contact, and the top of the range is withheld. Starter is 30 dollars per user per month and Advanced AI is 140 dollars, with a 22 percent discount for annual payment. A plan comparison table sets out what each tier includes. Every metered component has a stated unit and overage rate. Optical character recognition is one dollar per hundred pages after 3,000 free pages per billed user per month, and transcription is six cents per minute after three free hours. Document intelligence is one cent per page after 12,000 free pages, and storage is ten dollars per gigabyte per month after five included. AI credits cost one tenth of a cent each after 100,000 free credits per billed user per month. The agreement says credits are allocated monthly, do not roll over, vary in consumption by feature, cannot be refunded and can be capped by the customer. Enterprise, for twenty or more users, is a custom quote. The HIPAA business associate agreement carries an unstated fee, bulk uploads above 1,000 files a day are quoted upfront, and Enterprise data migration is unpriced.

A buyer can price it without a sales call. The pricing page publishes a Single Project at $499 per matter and firm-wide plans at $100, $250 and $500 a month, lists what each includes, prices usage in credits per page with top-up packets from 10 to 40 cents a credit, charges no per-seat fees, offers a free trial and links a refund policy.

Practice coverage is broad and set out page by page. Practice area pages cover employment law, criminal law, personal injury, workers' compensation, family law, complex litigation and medical chronologies. Workflow pages cover case management, litigation management, investigations, cross examination preparation, summary judgment and early case assessment, which is coverage by litigation stage as well as by subject. The vendor says it serves both plaintiff and defense firms, and two free tools support practice work: a litigation deadline calculator and a statute of limitations lookup by state and claim type. Further pages address individuals handling their own matters, such as VA disability claims, disability appeals, small claims, custody documentation, insurance disputes and estate settlement. The limits are a single jurisdiction and a thin firm dimension. Everything is United States practice, the agreement is governed by Georgia law with exclusive jurisdiction in Fulton County courts, and no international coverage is claimed. On firm size, the only published signal is that Enterprise starts at twenty users, and no segment statement, firm size breakdown or named practice is published.

Coverage and limits are both stated. The site names family law, bankruptcy, probate, estate planning, business litigation and accounting, with use cases for paralegals, high-net-worth divorce, mediation and bankruptcy trustees; it reads bank, credit card and check statements from PDFs, says paper must be scanned first, gives typical matter sizes, and states that brokerage and tax documents are not yet supported and are planned for the fourth quarter of 2026.