Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
BSIMM is a descriptive model of software security built from observed practice in real firms. It measures what your programme does against what peers do, rather than against a list of requirements.

What Is BSIMM?
BSIMM, the Building Security In Maturity Model, is a measuring stick rather than a rulebook. It began in 2008 when Gary McGraw, Sammy Migues, and Brian Chess interviewed nine firms about what their software security teams actually did, wrote down the activities they observed, and published the result. Every edition since has repeated that exercise with a larger pool of participants, and the model changes when observed practice changes.
That is the distinction worth holding on to. Most frameworks are prescriptive: here is what you should do. BSIMM is descriptive: here is what firms like yours are doing, and here is where you sit among them.
Activities are organised into four domains, each containing three practices, giving twelve practices in total.
| Domain | Practices |
|---|---|
| Governance | Strategy and metrics, compliance and policy, training |
| Intelligence | Attack models, security features and design, standards and requirements |
| SSDL touchpoints | Architecture analysis, code review, security testing |
| Deployment | Penetration testing, software environment, configuration and vulnerability management |
Within each practice, individual activities are graded at three levels. Level 1 activities are common, level 3 activities are rare and usually appear only in mature programmes. An assessment records which activities an organisation performs, producing a scorecard that can be compared with the pool average and with a chosen vertical such as financial services, healthcare, or technology.
A BSIMM assessment is an interview exercise, not a technical audit. An assessor speaks to the software security group, engineering leadership, and development teams, then marks each observed activity. There is no pass mark and no certificate; the deliverable is a spider chart and a set of observations.
Two outputs tend to be useful. The first is the gap between what the governance function believes happens and what engineers describe, which is often wide. The second is the comparison with peers, which changes budget conversations: it is easier to request investment in architecture analysis when the data shows most comparable firms already do it.
The two are frequently confused. OWASP SAMM is prescriptive and open: it sets out five business functions with maturity levels you can work towards, and the materials are free to download and self-assess. BSIMM reflects what a measured community already does, with data collected by the model’s owners.
In practice, teams use SAMM to plan and BSIMM to benchmark. A security programme that needs a roadmap gains more from SAMM. One that needs to justify its position to an executive committee or a prospective client gains more from a BSIMM scorecard. Neither replaces hands-on verification, which is why both models include penetration testing and code review as distinct practices.
BSIMM carries no regulatory status in the UK, so nobody will ask for it in a tender the way they ask for Cyber Essentials or ISO 27001. Its value is internal, and it maps neatly onto obligations that do carry weight. The NCSC’s secure development and deployment guidance covers much of the same ground in principle form, and software supply chain expectations arriving through the EU Cyber Resilience Act will push UK manufacturers selling into Europe towards exactly the activities BSIMM records: threat modelling, vulnerability handling, and coordinated disclosure.
Watch for two failure modes. Treating the scorecard as a target invites activity theatre, where teams perform a level 3 activity badly to tick a box. And benchmarking against the pool average is reassurance, not a risk statement: the average firm in any pool still ships exploitable defects. Use the model to find the practices you have genuinely skipped, then fix those.
You do not need a commercial engagement to benefit. Take the twelve practices, walk them with your engineering leads, and mark honestly which activities happen on every release rather than occasionally. Most organisations discover the same three gaps: nobody owns attack modelling, architecture review happens only for new systems, and security testing stops at the scanner. Those are tractable problems with real returns.
If you want independent evidence of how your software actually stands up, our web application security services provide the testing side of the picture that a maturity model can only describe.
No. BSIMM produces a descriptive scorecard showing which activities an organisation performs and how that compares with other participants. There is no pass mark, no certificate, and no accredited body issuing one. Organisations sometimes share assessment results with clients as evidence of software security maturity, but it carries no formal status in procurement the way ISO 27001 certification does.
BSIMM is descriptive and tells you what observed firms actually do, so it suits benchmarking. OWASP SAMM is prescriptive and open source, setting out maturity levels you can work towards, so it suits planning and self-assessment at no cost. Many teams run a SAMM self-assessment to build a roadmap and commission BSIMM when they need external comparison.
Typically a few weeks of elapsed time for a single business unit, built around interviews with the software security group, engineering managers, architects, and developers, plus documentation review. Large organisations assessing multiple divisions take longer. The scoring itself is quick; the useful work is reconciling what different parts of the organisation believe is happening on every release.
The full model assumes a dedicated software security group, which a ten-person team will not have. The practice list is still useful as a checklist: ask which of the twelve practices has a named owner and happens consistently. Smaller teams usually get faster value from OWASP SAMM or from fixing a specific gap such as dependency management or threat modelling.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.