Expert services
Research, judgment, human review, content execution, program management, and review interpretation.
Available nowRESPONSIBILITY BEFORE PRICE
First choose who owns execution,
then compare prices
Service fees buy judgment, execution, and review responsibility; subscriptions buy self-serve tools for your team. Authorized external collection and data calls are metered separately. Ranges below are validation pricing — published test ranges, revised openly as real delivery costs are measured.
Three fee classes, three different responsibilities
Product subscription
Agent, Codex Plugin, project collaboration, files, and the evidence workbench.
Private preview / plannedAuthorized usage credits
External providers, official data surfaces, remote execution, and tasks that need authoritative receipts.
Estimated and confirmed per taskAVAILABLE SERVICES
Four engagement depths, from baseline to enterprise systems
Final quotes depend on brand count, question scope, surfaces, cadence, and collaboration depth. We publish pricing mechanics, test ranges, and responsibility boundaries — never an unvalidated one-size-fits-all low price.
01Available now
Baseline diagnosis & strategy start
A first systematic view of how your brand performs across AI entry points
USD 1,500–2,500 · fixed scope- GEO / AEO / SEO baseline
- Brand, competitor, and citation analysis
- Company facts and evidence boundaries
- Question priorities and strategy roadmap
- 14-day verification or focused diagnosis
02Available now
Technical, content & source sprints
Confirmed gaps that need to become publishable assets
USD 3,500–8,000 · per sprint- Technical SEO and crawl readiness
- Entities, structured data, page specs
- Trusted content and content clusters
- Third-party sources and citation ecosystem
- Publishing acceptance and retest plan
03Available now
Ongoing GEO operations & retesting
Monthly progress on questions, content, sources, and reviews
USD 1,000–3,000 / month- Periodic answer and competitor monitoring
- Technical, content, and source task queue
- Fact changes and source-conflict upkeep
- Publishing records and like-for-like retests
- Human review and business recaps
04Assessed per program
Enterprise workflow & agent programs
Multiple brands, multiple teams, or internal agent environments
4–8 weeks · quoted per program- Question bank, fact, and evidence standards
- Multi-team review roles and report templates
- Agent / Plugin workflow design
- Permissions, data regions, and SOPs
- Training, integration, and acceptance
PRODUCT ACCESS
Subscriptions tiered by how your team works
There is no public purchase entry and no committed final pricing. We show tiers, audiences, and included surfaces first — prototypes are never packaged as SaaS already on sale.
Observer
Review projects, evidence, and reports
- Product surfaces
- Web review
- Public price
- Not yet set
Operator
A single-brand operator running tasks
- Product surfaces
- macOS Agent
- Public price
- Not yet set
Team
Teams working across Codex and Agent
- Product surfaces
- Agent + Codex Plugin
- Public price
- Not yet set
Enterprise
Multi-brand, permissions, audit, support
- Product surfaces
- All surfaces
- Public price
- Not yet set
CREDIT BOUNDARY
Everyday conversation and authorized execution never share one meter
Your own agent conversations, local file work, and ordinary drafts are not double-billed into FactRelay usage credits. Only explicitly labeled authorized services produce metered usage.
- Execution entry
- Codex Plugin
- Task
- 15 prompts × 5 runs baseline
- External provider
- 1 · explicitly labeled
- FactRelay credit
- Estimated before, receipted after
- Local conversation
- Never counted as FactRelay credit
You buy the execution process, not a ranking guarantee
Acceptance covers question sets, raw snapshots, evidence, human decisions, remediation tasks, publishing records, and like-for-like retests. FactRelay does not guarantee fixed AI rankings, recommendations, or revenue.