Black-box vs grey-box vs white-box penetration testing
Every penetration test is conducted with a certain amount of prior knowledge. Black-box testing gives the tester no information at all — they start where an external attacker would. White-box shares full access to source code and architecture. Grey-box, the most common model, gives partial credentials and context. Each has a different cost-to-coverage trade-off.

Black-box penetration testing
In a black-box engagement, the tester receives only the target's URL or IP range and no other information. They begin with open-source reconnaissance — searching public data sources, enumerating subdomains, mapping exposed services — before attempting any exploitation. The model closely simulates what an external attacker who has no inside knowledge would do.
Black-box is thorough for external attack surface testing. Its limitation is efficiency: the reconnaissance phase consumes a large portion of the engagement time, leaving less time for exploitation. Findings that require understanding the application's internal logic are harder to surface without code or architecture context.
- Best for: external attack surface mapping, red team exercises, testing the perimeter as an outsider would see it
- Time spent: significant proportion on reconnaissance
- Coverage: external attack surface deep, internal logic limited
Grey-box penetration testing
Grey-box provides the tester with partial context: typically a set of test user credentials, a list of application modules or endpoints, and in some cases a high-level architecture diagram. The tester knows what they are looking at without having the full source code. This is the most common engagement model.
Grey-box balances the realism of external testing with the efficiency of having a starting point. The tester can move directly to testing business logic, authentication flows and authorisation controls without spending days on reconnaissance. For web applications and APIs, grey-box delivers the highest coverage-to-cost ratio for most organisations.
- Best for: web application and API testing, mobile app testing, cloud configuration reviews, most compliance-driven engagements
- Time spent: focused on exploitation rather than reconnaissance
- Coverage: both external attack surface and internal logic
White-box penetration testing
White-box provides the tester with full access: source code, architecture diagrams, infrastructure configurations, credentials, CI/CD pipelines and documentation. The tester can review the application from both outside and inside, identifying vulnerabilities in business logic, cryptographic implementations, secure coding patterns and data-handling that would take weeks to find in a black-box or grey-box engagement.
White-box is the highest-depth option and the most expensive per engagement, but it is the most efficient for finding high-severity logic vulnerabilities in complex applications. It is commonly used for pre-launch reviews of products handling sensitive data, compliance code reviews (PCI DSS, HIPAA, DPDP), and mature organisations that have already conducted grey-box testing and want to go deeper.
- Best for: pre-launch security review, source code audit, compliance-driven code review, mature security programmes
- Time spent: analysis across full codebase and architecture
- Coverage: deepest available, including logic and cryptographic vulnerabilities
Which model is right for your engagement?
Most organisations commissioning VAPT for the first time or for annual compliance should choose grey-box. It satisfies the testing evidence requirements of RBI, SEBI, IRDAI, CERT-In and ISO 27001, and delivers the best coverage per day of engagement.
Black-box is appropriate when you specifically want to test your perimeter as an external attacker sees it — for example, before a major product launch or when assessing a recently acquired asset. White-box is appropriate when you need to assure the security of a specific application's code and logic, often as part of a secure development lifecycle or pre-certification review.
Related services
Frequently asked questions
Grey-box or black-box penetration testing by an independent, trained provider satisfies the RBI cyber resilience framework and SEBI CSCRF testing requirements. White-box may be used in addition but is not required. The key requirement is that testing is performed by a qualified independent team, not by internal staff.
In practice, attackers often acquire some prior knowledge before attacking a target — from phishing, social engineering, leaked credentials or public information. Grey-box more closely models a persistent external attacker who has done their reconnaissance than a one-shot outsider with zero knowledge. For most applications, grey-box provides better coverage of business-logic vulnerabilities.
Yes. A common progression is: grey-box for the first annual VAPT to establish a baseline; white-box in a subsequent cycle or for a specific high-value application where deeper code-level assurance is needed. We scope each engagement independently, so you can mix models across your application portfolio.