Contract Review AI: Comparing LegalOn, Spellbook, and Luminance
A feature-by-feature analysis of three leading platforms—and where each excels or falls short.
The contract review AI market has matured past the hype phase. Tools that once claimed to "do it all" now occupy distinct niches with specific strengths and limitations. For legal teams evaluating these platforms, understanding the differences matters more than marketing claims.
This comparison examines LegalOn, Spellbook, and Luminance across features, use cases, and practical limitations based on published benchmarks and user feedback.
Market Context
The 2025 legal AI landscape shows clear specialisation. Corporate legal departments and law firms are actively evaluating and deploying AI for contract review at an accelerating rate. The average in-house attorney still spends significant hours daily on contract review -- a number that suggests substantial room for efficiency gains.
However, adoption comes with documented risks. General AI tools hallucinate on legal queries 69--88% of the time according to Stanford research. Success requires selecting tools with genuine legal expertise built into the AI, not generic models with legal branding.
LegalOn
Best for: In-house legal teams and law firms handling standard contract types at volume.
Core Approach
LegalOn differentiates on immediate usability. The platform ships with pre-built, attorney-crafted playbooks that work from day one. Users report reviewing contracts within one hour of installation -- no training period, no custom configuration required.
The platform serves over 6,500 companies globally and offers 50+ pre-built playbooks. The "My Playbooks" feature allows customisation of the AI review system according to internal standards.
Strengths
- Zero ramp-up time: The pre-built playbook architecture means teams can start reviewing contracts immediately
- Playbook customization: Organizations can modify rules to match internal standards and preferred language
- Consistency at scale: The playbook approach ensures consistent review across team members and time
Limitations
- Contract type constraints: The platform works best with contracts that match available playbook types. Novel or highly bespoke agreements may not benefit as much.
- Customization ceiling: While playbooks can be modified, teams with unique review requirements may find the framework limiting.
- Pricing opacity: Custom pricing model makes direct comparison difficult. Market reports suggest small teams pay $3,000-$8,000 annually.
Best Use Cases
- NDA review at volume
- Standard commercial agreements (MSAs, SaaS terms)
- Procurement contracts with consistent structure
- Teams needing immediate deployment without training investment
Spellbook
Best for: Solo practitioners and small firms drafting contracts in Microsoft Word.
Core Approach
Spellbook integrates GPT-4 directly into Microsoft Word, making contract drafting as frictionless as possible. The platform is designed for small businesses and lean legal teams that work primarily in Word and need real-time assistance during drafting rather than separate review workflows.
The company reports reviewing over 10 million contracts across 4,000+ legal teams in 80+ countries, and in October 2025 raised $50 million in Series B funding led by Khosla Ventures.
Strengths
- Microsoft Word integration: Work within your existing environment without context-switching
- Real-time drafting assistance: Suggestions appear as you draft rather than in a separate review phase
- Clause identification: Excels at spotting risky clauses, suggesting alternative language, and flagging missing provisions
- SOC 2 Type II compliance: Security certification addresses enterprise data handling concerns
Limitations
- Not for large-scale review: The drafting-focused design means it is not optimized for reviewing large document sets or due diligence projects
- Individual-focused: Better suited for individual attorneys than coordinated team review workflows
- Price point: Starting around $589/user/month (annual payment), the cost adds up quickly for teams
Best Use Cases
- Solo attorneys drafting agreements from scratch
- Small firms that work primarily in Microsoft Word
- First-pass clause review during drafting
- Teams that prefer in-document assistance over separate review platforms
Luminance
Best for: Large law firms and enterprises handling M&A due diligence and large document portfolios.
Core Approach
Luminance is built on a proprietary legal LLM trained on over 150 million verified legal documents. The platform uses unsupervised machine learning to detect patterns and anomalies in large document sets -- a fundamentally different approach from playbook-based review.
The Legal Inference Transformation Engine (LITE) combines pattern recognition with supervised and unsupervised machine learning. In February 2025, Luminance raised $75 million to expand its enterprise capabilities.
Strengths
- Scale: Purpose-built for reviewing thousands of documents in due diligence scenarios
- Anomaly detection: Pattern recognition identifies unusual clauses across large portfolios
- Multi-language support: Works across languages without extensive data training
- M&A specialization: Provides instant insight across over 1,000 legal concepts for deal room analysis
Limitations
- Learning curve: More sophisticated capabilities require more sophisticated users
- Enterprise pricing: Quote-based model with estimates ranging $10-$100 per user; not accessible for small teams
- Overkill for routine work: The platform's strengths are wasted on standard contract review that doesn't involve large document sets
Best Use Cases
- M&A due diligence with large data rooms
- Lease portfolio analysis
- Regulatory compliance across document sets
- Enterprise contract analytics and risk visualization
Feature Comparison Matrix
| Feature | LegalOn | Spellbook | Luminance |
|---|---|---|---|
| Implementation Time | Hours | Hours | Days/Weeks |
| Primary Interface | Web platform | Microsoft Word | Web platform |
| Pre-built Playbooks | 600+ | Limited | Via LITE |
| Custom Playbooks | Yes | Limited | Yes |
| Due Diligence Scale | Moderate | Low | High |
| Anomaly Detection | Rule-based | GPT-based | Pattern ML |
| Multi-language | Limited | Limited | Strong |
| Team Collaboration | Good | Individual | Enterprise |
| Approximate Cost | $3K-$8K/year | $7K+/user/year | Enterprise quote |
Performance Metrics
Published efficiency claims (note: these are vendor-reported figures):
LegalOn: Users report reviewing contracts within one hour of installation. NDA review reduced from 1-2 hours to 15-30 minutes.
Spellbook: Real-time suggestions during drafting; no separate review step required.
Luminance: Contract review time savings up to 90% (vendor-reported). A&O Shearman's ContractMatrix (built on Luminance) saves around seven hours from average contract review -- approximately 30% efficiency gain.
Limitations Common to All
These tools share certain constraints that buyers should understand:
-
Human review remains essential: AI flags issues and suggests changes but cannot replace professional judgment on business terms and risk allocation.
-
Training data matters: All tools perform better on contract types well-represented in their training data. Novel structures may produce unreliable results.
-
Integration complexity: Getting AI tools to work with existing document management and CLM systems requires planning.
-
Version control: Coordinating AI suggestions with human edits across multiple reviewers introduces workflow complexity.
Selection Framework
Choose LegalOn if:
- You need immediate deployment without configuration
- Your contracts fit standard types (NDAs, MSAs, procurement)
- Consistency across reviewers is a priority
- Budget allows $3,000-$8,000 annually
Choose Spellbook if:
- You work primarily in Microsoft Word
- You need real-time drafting assistance
- You are a solo practitioner or small team
- Per-seat cost at ~$589/month is acceptable
Choose Luminance if:
- You handle M&A due diligence or large document reviews
- Pattern detection across portfolios adds value
- You need multi-language support
- Enterprise budget and implementation timeline available
Key Takeaways
- LegalOn excels at immediate deployment with pre-built attorney playbooks—best for standard contract types at volume
- Spellbook integrates directly into Microsoft Word for real-time drafting assistance—best for solo and small-firm practitioners
- Luminance's pattern recognition and scale capabilities make it the specialist for M&A due diligence and enterprise document analysis
- All three tools require human oversight—AI identifies issues but professional judgment determines responses
- Selection should match specific use cases rather than seeking a general-purpose solution
For practical playbook implementation guidance, see Dioptra's guide to building contract playbooks that work with AI.