Discover
Find product capabilities, plan facts and stated trial or contract limits.
SaaS
Make SaaS discovery and self-service agent-ready.
Coming later: test whether agents can find accurate product facts, answer from your docs and reach signup. Agents stop before login or account creation.
For self-service SaaS, product-led growth, developer tools, support and billing teams.
Test product understanding and account workflows together.
The audit checks stated commercial facts without requiring every SaaS company to publish pricing.
Answer from docs
Find a known answer and preserve the supporting source.
Start signup
Reach the public submission boundary without creating an unwanted account.
Onboard
Coming later: complete signup and onboarding on an approved staging site with known fixtures.
Self-serve
Find plan-change or cancellation instructions on public pages.
Recover
Coming later: handle validation and expired sessions on a staging site without entering an auth loop.
Keep website claims, docs and product UI consistent.
The run follows one question from discovery into the controlled product environment.
Sourced answer
The agent finds a current claim and cites the canonical supporting page.
Valid entry
The signup entry point keeps the chosen plan and locale. The agent stops before the form is sent.
Self-service assertion
The expected setting or instruction is present and independently verified.
SaaS friction often crosses team boundaries.
Split commercial facts
Marketing pages, help content and product UI describe different limits.
Ambiguous CTA
Trial, demo and account actions share labels without clear outcomes.
Signup dead end
The signup entry point loops back or hides the next step, so the agent can't find where to start.
Hidden self-service
Upgrade, downgrade or cancellation guidance cannot be found or verified.
Trace the cost of finding an answer and acting on it.
Evidence separates content discovery, authentication and product interaction time.
Step screenshots
Follow cited sources, the signup stop point and self-service steps, with masked screenshots.
Completion time
See how long discovery and docs navigation took, step by step.
Layout shifts
Measure layout shifts on product, docs and signup pages.
Source consistency
Compare claims across marketing, documentation, help and product surfaces.
Public mode stops before account creation or a commercial commitment.
AI agents stop before login, account creation and any form that sends data. Onboarding tests with approved test accounts are coming later.
The audit maps both public content and agent-native interfaces.
Missing optional technology is reported as state, not treated as a blanket failure.
- Canonical product pages, docs and llms-full.txt
- SoftwareApplication structured data when appropriate
- Coming later: accessible forms and validation
- Coming later: session persistence and account recovery
- Declared WebMCP forms when present: detected, not run
- Prompt injection exposure and data boundaries
Teams receive one evidence chain across public and product surfaces.
- Results per discovery and docs task for each agent that ran
- Step screenshots and source evidence
- Cross-surface content inconsistencies
- Signup entry point findings with the failed step
- Engineering backlog with acceptance criteria
SaaS audit questions
Must a SaaS website publish pricing?
No. The audit checks the accuracy and discoverability of commercial information the company chooses to state.
Can authenticated product flows be tested?
Coming later, with approved test tenants, accounts and scoped permissions. Agents stop before login today.
Are documentation answers evaluated only by an LLM?
No. Literal checks on source presence and final state decide the result. An AI check can confirm or flag it, never turn a failed check into a pass.
Practical testing guide
Keep one SaaS account intent intact.
Build a repeatable scenario across product facts, signup, authentication, onboarding, permissions and self-service recovery.
Read the SaaS testing guideRelated
- GuideTest one account journey end to end.Test SaaS websites with AI agents across product facts, signup, authentication, onboarding, permissions, self-service and safe account boundaries.
- MethodologyHow a task is defined, run, checked and attributed.
- Use caseE-commerceAvailable today: AI agent task tests for product discovery, price, stock, delivery and returns, plus the automatic site check. Cart and checkout journeys are coming later.
- Use caseBooking and travelComing later: AI agent task tests for search, availability, prices, policies and timezones. Today every website gets the automatic site check.
- Use caseMarketplacesComing later: AI agent task tests for seller discovery, offer comparison, trust signals, fees and availability. Today every website gets the automatic site check.
- Use caseDigital agenciesRun automatic site checks across client sites with ranked fixes and retests. AI agent task tests for client journeys are coming later.
Test one SaaS journey across marketing, docs and product.
Apply with a discovery, onboarding or self-service workflow for a private partner review.