Yearly running cost Assuming 6,000 contracts a year
By hand $240K / year 3,000 hours of work
Subscription $1.8K+ / year
This workflow $1.1K / year Machine usage only; setup, hosting and review are extra. Contract review What do you want back from the review?Check contracts against your legal team's playbook. Get cited issues, or add proposed edits using your approved clauses.
Find deviations Cited issues, preferred positions and review questions. $1,152 – 3,360 setup Also draft changes Add proposed edits using your approved fallback language. $1,872 – 5,280 setup
How should proposed edits be delivered?DOCX with tracked changes Structured edit suggestions for my own editor Your legal team decides. Nothing is accepted, signed or sent automatically.
What is included Guide prices assume reusable document and review components adapted with AI-assisted development. Your legal team supplies the playbook, fallback language and labelled examples.
Check the contract against your approved playbook and return findings with clause references. Do not modify the contract.
The starting scope covers an agreed agreement family, such as NDAs, with up to 20 playbook checks. Use English DOCX or text PDFs. A Word add-in, scanned documents and contract lifecycle integrations are separate requirements.
Accuracy How thoroughly should it check your playbook?Choose how many deviations it should catch and how many flags your reviewer should confirm.
Playbook checks Find deviations, link evidence and suggest your approved position. ≥95% deviations caught · ≥90% flags confirmed Included More thorough checking Add another verification pass and more difficult-contract testing. ≥98% deviations caught · ≥95% flags confirmed Setup +$432 – 1,200 My own requirements Describe your contract types, playbooks or review targets. Quote separately
How accuracy is measured These are acceptance targets, not measured results. Your legal team labels held-out contracts using the selected playbook. Count missed issues and false flags separately, and check the evidence behind each finding.
Targets for your selected checks What is checked Target Playbook deviations caughtDetected labelled deviations divided by all deviations the reviewer marked under the chosen playbook; missed or unanswered checks count as misses. ≥95% Flags confirmed by reviewersFlags the reviewer confirms divided by all flags; duplicates and unsupported allegations count as false flags. ≥90% Correct clause referencesFindings whose cited clause supports the finding, or whose missing-clause claim is correct, divided by all findings requiring evidence. Missing evidence fails. ≥99% Playbook-aligned recommendationsExpected issue responses using the correct preferred or fallback position divided by all labelled responses needed. Omitted recommendations fail; suggested redlines must preserve the intended meaning. ≥95%
Your legal team reviews every result. Unclear terms, missing evidence and conflicting rules must be escalated. High scores do not establish enforceability or replace legal judgment.
Speed How quickly do you need the contract review?Choose how quickly findings and any proposed edits should be ready. Your legal team's review time is separate.
Within 3 minutes Included Within 90 seconds Setup +$144 – 480 Within 30 seconds Setup +$288 – 960
Timing details From file received to findings and the selected output, including parsing, queueing and retries. At least 95% of test contracts should meet the target, up to 20 pages and 20 checks, with 1 in progress at a time. Confirm hardware and document limits with your provider.
Running-cost target How much do you want to spend per contract?Select what your workflow can spend reviewing each contract. This may affect the model and number of verification passes.
Up to $0.35 Included Up to $0.25 Setup +$144 – 480 Up to $0.15 Setup +$288 – 960
Cost details Reference machine cost: $0.22 – 0.30 per contract at 500 contracts a month. Includes parsing, model calls, retries and allocated hosting, not legal review, maintenance or software subscriptions.
Your selected budget is a target, not a measured cost. Test it together with your accuracy and speed requirements; a lower budget may require different components.
Your environment Where do you want it to run?Choose where it runs and whether your contract and playbook text can be sent to an approved AI model.
Run it onMy cloud account My own computer or server AI model accessUse an approved AI model Keep all processing inside my environment
Data and access details Runs in your cloud account with your access and retention settings. Model access separately controls whether contract text leaves it.
Only authorised contract and playbook text goes to the named external model. Confirm access, retention and permitted use with your legal team.
Confirm contract confidentiality, document permissions and retention with your legal team. Keeping processing local does not make an untested model reliable.
Interface How do you want to use it?Use your existing AI agent, a dedicated review agent or a web page. You can choose more than one.
My existing AI agent Let your agent submit contracts and retrieve findings through MCP tools. Setup +$72 – 360 A dedicated review agent Use a separate agent to run playbook checks and explain findings. Setup +$144 – 480 A review web page Upload contracts, inspect cited issues and download proposed edits. Included
Estimated setup price $1,152 – 3,360
This guide price is a starting point for your conversation with a provider. Agree the final scope and price directly with your provider.
Your requirements
Workflow Find deviations
Accuracy Playbook checks
Speed target 3 minutes / contract
Cost target $0.35 / contract
Environment My cloud account
Interface A review web page Per review run, including contracts requiring legal attention.
Prepare my brief → Generate a draft prompt to share with your connected RenX agent.
Enable JavaScript to configure a brief, or prepare your task in RenX .