A penetration test is a scoped, time-boxed assessment in which a contracted tester examines agreed systems and reports on everything they cover. A bug bounty pays outside researchers for each valid vulnerability they find, for as long as the programme runs. A penetration test gives you planned coverage and evidence; a bounty gives you ongoing, varied attention.
Both involve skilled people trying to break your systems, and both find real vulnerabilities, so it is easy to treat them as alternatives. They are not. They differ in who does the testing, what gets covered, how you pay and what you can show a customer or auditor afterwards. For most organisations, a penetration test comes first.
Executive Summary & Key Takeaways
- A penetration test is a contract for coverage: a defined scope, known testers, a fixed timeframe and a report on what was tested, including what held up.
- A bug bounty is a contract for results: you pay per valid finding, coverage is whatever researchers choose to look at, and nothing tells you what was not examined.
- Both sit on top of a vulnerability disclosure process. UK codes of practice expect software vendors to publish one, and the law requires manufacturers of consumer connectable products to publish a way to report security issues.
- Bounties work best for mature teams that can triage and fix quickly. Without that capacity, a bounty produces a queue rather than security.
- Only 13% of UK businesses reported carrying out penetration testing in the previous 12 months, so for many organisations independent testing is the gap to close first.
How each one works
Penetration test
You agree the scope, rules of engagement and timing with a testing provider. Testers combine tools with manual techniques for a set number of days, then report findings with evidence, severity and fixes, usually followed by a retest. Because scope is agreed in advance, the report can say what was covered as well as what was found. Our guide to what happens during a penetration test walks through each stage.
Bug bounty
You publish rules setting out which systems are in scope, what testing is allowed and what you pay for each class of valid vulnerability. Independent researchers test when and how they choose within those rules, and submit reports. You validate each report, pay for valid and unique findings, and fix them. Programmes can be private (invited researchers only) or public, and are often run through a specialist platform that manages researchers and payments.
Vulnerability disclosure: the foundation for both
A vulnerability disclosure policy is a published route for anyone to report a security issue, with no reward attached. The NCSC’s Vulnerability Disclosure Toolkit sets out three components: a contact route, a policy explaining what reporters can expect and what is in scope, and a security.txt file (RFC 9116) telling finders where to report (NCSC, retrieved 10 November 2026). ISO/IEC 29147 is the international standard for vulnerability disclosure (ISO, retrieved 10 November 2026).
A bug bounty is, in effect, a disclosure policy with payments and active encouragement. If you cannot yet handle unsolicited reports well, you are not ready to invite more of them.
Side by side
| model | Penetration test | Bug bounty |
|---|---|---|
| Who tests | Contracted testers you have selected | Independent researchers, often unknown to you |
| Scope | Agreed in writing; can include internal and authenticated systems | Published rules; usually internet-facing assets |
| Coverage | Systematic within the agreed scope | Uneven: researchers go where rewards and interest lie |
| Timing | Fixed window, scheduled around change | Continuous for as long as the programme runs |
| Payment model | Fee for the engagement | Rewards per valid finding, plus any platform fees |
| Output | One report with evidence, severity, fixes and retest results | Individual reports, one vulnerability at a time |
| Evidence for customers and auditors | Strong: shows what was tested and when | Limited: shows what was found, not what was covered |
| Internal effort | Scoping, support during testing, fixing | Ongoing triage, validation, researcher communication, payments, fixing |
Which one do you need?
For most organisations the sensible order is a vulnerability disclosure process, then regular penetration testing, then a bug bounty once the first two are working well.
A disclosure process is increasingly expected
The Software Security Code of Practice, published by DSIT and the NCSC on 7 May 2025, asks software vendors to implement and publish an effective vulnerability disclosure process (principle 3.2) (GOV.UK, retrieved 10 November 2026). The Code of Practice for the Cyber Security of AI sets the same expectation for AI developers and operators (provision 6.4) (GOV.UK, retrieved 10 November 2026). Both codes are voluntary, but the software code is explicitly designed for customers to use in negotiations and contracts with their suppliers.
For manufacturers of consumer connectable products sold in the UK, it is law. Since 29 April 2024, the product security regime has required them to publish at least one point of contact for reporting security issues, and when reporters can expect an acknowledgement and status updates (legislation.gov.uk, retrieved 10 November 2026).
If you sell products with digital elements in the EU, the Cyber Resilience Act has required manufacturers to report actively exploited vulnerabilities since 11 September 2026, with an early warning due within 24 hours of becoming aware (European Commission, retrieved 10 November 2026). A working intake process is how many such issues first reach you.
Penetration testing gives you planned assurance
Choose a penetration test when you need to know that a specific system, release or environment has been examined properly: before launch, after major change, ahead of an audit or for a customer’s security review. It can include internal networks, cloud configuration and authenticated areas that bounty researchers rarely see. Yet only 13% of UK businesses reported carrying out penetration testing in the previous 12 months (Cyber Security Breaches Survey 2025/2026, DSIT and Home Office, retrieved 10 November 2026). Our guide to which type of penetration test you need can help with scope.
A bounty adds breadth once you are ready
A bug bounty makes sense when you have a large or fast-changing internet-facing estate, have already fixed the findings from recent testing, and have engineers who can triage and fix reports quickly. It brings many different perspectives and keeps attention on your systems between tests. Starting with a private programme keeps volume manageable while you learn.
What affects cost and effort
Penetration test cost is driven mainly by scope: the number of systems, the depth of testing and the number of testing days needed. Bug bounty cost is less predictable. You pay rewards per valid finding, usually scaled by severity, and many programmes also pay platform or management fees.
The larger hidden cost is internal. Every report needs validating, deduplicating, answering and fixing, and a public programme can produce a steady flow of low-quality submissions alongside the valuable ones. Budget for the people, not only the rewards.
Legal and practical points
Under section 1 of the Computer Misuse Act 1990, a person commits an offence if they cause a computer to perform a function intending to secure access they know is unauthorised (legislation.gov.uk, retrieved 10 November 2026). A penetration test is authorised by contract. Bounty researchers rely on your published rules, so scope, permitted techniques and your commitment not to pursue good-faith researchers need to be written clearly.
The government has said its National Security Bill will update the Computer Misuse Act and “unlock the power of cyber security professionals to better enable them to secure computer systems” (The King’s Speech 2026: background briefing notes, retrieved 10 November 2026). As at 10 November 2026 that reform has not been enacted, so current law still applies.
Your rules should also tell researchers what to do if they encounter personal data: stop, do not keep it, and report it.
Common mistakes
- Launching a public bounty before fixing known issues, and then paying for vulnerabilities a penetration test would already have found.
- Offering a bounty as evidence of coverage when a customer or auditor has asked for an independent test.
- Publishing vague scope, which invites testing of systems you do not own or did not mean to expose.
- Underestimating triage effort, so reports go unanswered and researchers lose patience.
- Asking people who report issues unprompted to sign non-disclosure agreements, which the NCSC advises against.
Frequently asked questions
Can a bug bounty replace our annual penetration test?
Rarely. A bounty cannot show what was covered, usually excludes internal and authenticated areas, and gives no assurance that anyone looked at a particular system. Customers and frameworks that ask for independent testing generally expect a scoped test and a report. A bounty works best alongside testing, not instead of it.
Do we need a bug bounty to comply with anything?
The UK requirements described above call for a way to receive and handle vulnerability reports, not for paid rewards. A disclosure policy meets that expectation. A bounty is a choice you can make once disclosure and testing are working.
Should we run a private or public programme?
Start private. Inviting a small group of researchers keeps report volume manageable and lets you refine scope, rewards and triage before opening up. Move to a public programme only when your team can handle a steady flow of reports.
What should we do if someone reports a vulnerability we didn’t ask for?
Respond promptly and thank them, pass the report to whoever owns the affected system, ask for more detail if you need it, and keep the reporter updated until it is fixed. The NCSC stresses that an unsolicited report is not in itself an attack or an incident (NCSC, retrieved 10 November 2026). Afterwards, publish a disclosure policy so the next person knows exactly what to do.
Getting started
If you have never had an independent test, start there: it tells you where you stand, and clears the obvious issues before you pay anyone else to find them. Our penetration testing service runs from scoping to retest, and our product security work covers setting up vulnerability disclosure. Talk to our team about the right mix for you.

Jared Henderson
Writing about cyber security and technology at Falx.





