Home  /  Insights

PRINCE2 Agile Version 2 was rebuilt, not refreshed — and the hexagon has quietly gone

March 21, 2026 · PRINCE2 Agile

If you learned PRINCE2 Agile from the 2015 guidance, there are three things you can probably still draw from memory: the hexagon, fix and flex, and the five targets. They were the parts that made the method feel distinctive — the visual answer to the question every project manager gets asked, which is what gives when the date will not move?

Open the Version 2 syllabi and they are not there.

I want to be careful about how I say this, because the distinction matters. I have read the PeopleCert Foundation and Practitioner syllabi for PRINCE2 Agile (Version 2), both marked version 2.0 and dated May 2025, and I have cross-checked them against an independent accredited course datasheet published in August 2025. Fix-and-flex, the hexagon and the five targets appear in none of the three. What I can say is that they are not examinable in Version 2. I have not read the official Version 2 book cover to cover, so I will not claim they have been struck from the method itself. Syllabus absence is strong evidence of non-examinability. It is not proof of textual deletion.

But there is a structural reason to think this was deliberate, and once you see it you cannot unsee it.

The geometry gave way

PRINCE2 Agile Version 1 built fix-and-flex on six performance targets: time, cost, quality, scope, risk and benefits. Six targets, six sides, one hexagon. The visual worked because the arithmetic worked.

PRINCE2 7, released in September 2023, added a seventh performance aspect: sustainability. The full set is now benefits, cost, time, quality, scope, sustainability and risk.

Version 2 of PRINCE2 Agile is built on PRINCE2 7. PRINCE2.com describes it as “a deliberate and thoughtful alignment with PRINCE2 7 — the latest evolution of the core PRINCE2 project management methodology.” The German accredited training organisation Serview puts it more bluntly: Version 2 is “based entirely on the modernized PRINCE2 7.”

A hexagon cannot hold seven aspects. You either redraw the model as a heptagon — which nobody wants to say out loud, let alone print — or you retire it.

I think they retired it, and I think that was the right call. But it leaves a gap, because fix-and-flex was doing real work. It was the mechanism by which a governance framework told a project board in advance which variable it had agreed to sacrifice. Version 2’s answer to that question is distributed across tolerance, prioritisation and tailoring rather than concentrated in one memorable diagram. That is more honest and considerably less teachable.

What Version 2 actually changed

Six things, from the syllabi and PeopleCert’s own material.

It runs on PRINCE2 7’s architecture. The same seven principles, the same seven practices — note practices, not “themes”; that word was retired in September 2023 — and a role model derived from PRINCE2 7 rather than from the 2015 guidance.

Terminology was aligned. PRINCE2.com: “As PRINCE2 7 introduced a more accessible and streamlined language … this new version of PRINCE2 Agile has followed suit.”

Mindset became a first-class topic. Version 2 leans hard on the distinction between “doing agile” and “being agile” — the latter defined as “adopting the underlying mindset and values that make agility work.” The Agile Onion appears in the syllabus: mindset, values, principles, practices and, at the outer layer, processes.

New content on agile leadership and behaviours. This is where the “people” element of PRINCE2 7 flows through.

Performance measurement in agile projects, and expanded tailoring and scaling guidance.

AI. The Practitioner syllabus includes “AI support for PRINCE2 Agile” in its wider-context section — a small allocation, but its presence tells you where the next revision will go.

What survived from Version 1 is the Agilometer, and I am glad. It is in the Version 2 Practitioner syllabus, and I will come back to it in a separate article because it is the most under-used instrument in the whole method.

The eleven roles, which nobody is talking about

The Version 2 role model is worth reading carefully, because it is where the method commits itself: Project Executive, Chief Product Owner, Senior Supplier, Project Assurance, Agile Coach, Project Manager, Project Support, Product Owner, Team Coach, Developer, Tester.

Eleven roles. Two coaching roles. A Chief Product Owner sitting above Product Owners. Project Assurance kept as a distinct accountability.

Whether your organisation has eleven people available for eleven roles is not the point — PRINCE2’s tailoring principle expects roles to be combined. The point is that the method has taken a position on which accountabilities must exist somewhere, even when one person holds three of them. In my experience the two that go missing first are Project Assurance and the Chief Product Owner, and their absence produces the same failure twice: nobody independent is testing what the project team reports, and nobody is reconciling competing product priorities above the level of individual teams.

What the method still cannot do for you

Now the uncomfortable part, and it belongs in any honest article about a project management framework.

In 2019 Birmingham City Council — the largest local authority in Europe — began replacing a heavily customised back-office system with Oracle Fusion Cloud, covering finance, HR, payroll and procurement.

The design principle was sound, and it is the correct principle for a cloud SaaS implementation: “adopt not adapt.” Configure the organisation to the software’s standard processes rather than bend the software to the organisation’s existing ones. Every experienced implementer will tell you this is the single decision that most determines whether a cloud ERP succeeds.

The council’s external auditor, Grant Thornton, found that the council did not hold to it. It reversed into adapting the ERP to existing council processes.

The go-live slipped from December 2020 to a big-bang cutover in April 2022. The solution design was never fully resolved beforehand. Bank reconciliation could not be tested end to end — in the auditor’s words, “Council staff testing BRS were unable to fully test the solution because the first step in the process, loading the bank files, and key consolidation reports were failing.” For roughly eighteen months after go-live, from April 2022 to late 2023, there was no usable audit trail, which removed the council’s ability to detect fraud. The auditor concluded that the system “effectively crippled the council’s ability to manage and report on finances.”

The budget history is the part I would put in front of a project board:

StageFigure
Original budget (2021)£19.965m over three years
Revised after go-live slip (2022)£40m
Revised (2024)£131m
Reported total including 2026/27 provision (Sept 2025)£170m
Additional Oracle contract (June/July 2024)£10m to December 2026

The council issued a section 114 notice in September 2023. In fairness — and this matters for accuracy — the section 114 was driven principally by the council’s equal pay liability, with the ERP failure as a major compounding factor. The auditor and the government commissioners treat them as linked but distinct. Full re-implementation is now targeted for April 2026, roughly five years after the original go-live date.

The governance findings are the transferable part

Read these as a checklist, because every one of them is available to you today.

Officers’ understanding of the solution was “limited”, and the digital directorate lacked the Oracle expertise to act as an “intelligent customer” capable of challenging the systems integrator.

The Financial Impact Assessment — which explicitly warned of “major problems transacting leading to late payment or collection of debt” — was not discussed at the decisive Steering Committee meeting.

Supplier caveats to go-live readiness opinions appeared only “in the supporting detail to Steering Committee reports,” not in the main decision papers.

Decision-makers “failed to adequately scrutinise available information as part of their decision-making process” before go-live.

And on elected member oversight: “members had indirect sources of information that could have enabled them to challenge officers but this never occurred.”

Not one of those is a methodology failure. PRINCE2’s practices cover every one of them — business case, quality, risk, issues, progress. The information existed. It was in the pack. It was in the supporting detail. Nobody read it out loud in the room where the decision was made.

That is the recurring lesson of forty years for me: frameworks are good at requiring that information be produced, and useless at requiring that anyone act on it.

What to do about Version 2

If you hold PRINCE2 Agile Version 1, your material is now a version behind and rests on a hexagon that is no longer examinable. That does not make your knowledge wrong — fix-and-flex is still a perfectly good way to think — but do not teach it to a team preparing for a Version 2 exam.

If you are buying training, validate the course outline against the PeopleCert syllabus PDFs before you sign. Several accredited training organisation pages badged “Version 2” still contain Version 1 content. The quickest tell is the word “Themes”, which was retired in September 2023. If a page selling you Version 2 says “themes”, it has not been updated, whatever the heading says.

If you are adopting the method, start with the eleven roles rather than the processes. Write down which of your people holds each accountability. The two you cannot fill are the two that will hurt you.

And put the Birmingham findings on the first page of your project board induction pack. Not as a horror story — as five questions the board is expected to ask at every stage boundary.

What the Version 2 exams actually test

If you are planning study, the syllabus weightings tell you more than any course brochure.

Foundation (Version 2) — 40 questions, 60 minutes, closed book, 60% to pass:

AreaWeighting
Key concepts30%
Principles, practices and artefacts40%
Processes and workshops25%
Wider context5%

Practitioner (Version 2) — 50 questions, 150 minutes, open book (the official book only), 60% to pass:

AreaWeighting
Agile mindset and change management20%
Practices and roles44%
Processes30%
Wider context6%

Two things jump out. Nearly half the Practitioner paper sits on practices and roles — which is why the eleven-role model is worth learning properly rather than skimming. And a fifth of it is mindset and change management, a category that barely existed in the Version 1 emphasis. If your preparation is all process diagrams, you are studying for the wrong exam.

The artefacts and techniques named in the syllabi are worth listing, because they are what you will actually be asked to apply: product and project backlogs, benefit, project and team dashboards, epic user story, project canvas, release map, timebox — in its three forms of iteration, release and stage — and waterfall as an explicit artefact rather than an insult. On techniques: agile estimation including story points and t-shirt sizing, burn charts, minimum viable product, user stories, definition of ready, definition of done, retrospectives and agile workshops.

Note what is not named anywhere in either syllabus: SAFe. PRINCE2 Agile’s stated posture is tool-neutrality — the original guidance said it “will work with any established Agile approach” — and Version 2 keeps that. The approaches named as popular agile methods are Scrum, Kanban, Lean and Lean Startup. If a course or a consultant tells you PRINCE2 Agile contains a SAFe mapping, ask them to show you where.

One more caution on a widely repeated detail. MoSCoW prioritisation appears in both versions, and people routinely attach to it the 60/20/20 effort split across Must, Should and Could. That ratio originates with DSDM and the Agile Business Consortium, not with PRINCE2 Agile. Use it if it helps you — it often does — but attribute it correctly, particularly in front of an examiner or a client who knows the difference.

One thing I would not claim

I have described Version 2 as a rebuild rather than a refresh, and I stand by that. But I want to be careful not to oversell it either, because the same evidence supports a more modest reading.

The seven principles are PRINCE2 7’s principles. The seven practices are PRINCE2 7’s practices. The seven processes are unchanged from a structure that has been stable for decades. What Version 2 genuinely adds on its own account is the mindset framing, the leadership and behaviours content, expanded tailoring guidance, performance measurement, and a small allocation to AI.

That is a substantial and coherent update. It is not a new method, and anyone who tells you your existing project governance is now obsolete is selling something.

The honest summary is that Version 1 had become incoherent — aligned to an edition of PRINCE2 that had been superseded for nearly two years — and Version 2 fixes that. Coherence is worth having. It is just a smaller claim than “rebuilt from the ground up,” and I would rather make the smaller claim accurately.


Sources

  • PeopleCert, PRINCE2 Agile Foundation (Version 2) syllabus, v2.0 (May 2025) — peoplecert.jp
  • PeopleCert, PRINCE2 Agile Practitioner (Version 2) syllabus, v2.0 (May 2025) — peoplecert.jp
  • PeopleCert, PRINCE2 receives major update (PRINCE2 7 availability, 4 September 2023) — peoplecert.org
  • PRINCE2.com, Aligning PRINCE2 Agile Version 2 with PRINCE2 7 (29 July 2025) — prince2.com
  • PRINCE2.com, From doing agile to being agile (15 July 2025) — prince2.com
  • Serview, PRINCE2 Agile Version 2: what is new and what remains (19 September 2025) — serview.de
  • Purple Griffon, What’s new in PRINCE2 7 (seven performance aspects, practices, integrated elements) — purplegriffon.com
  • Knowledge Train, PRINCE2 Agile targets (Version 1 fix-and-flex and the five targets) — knowledgetrain.co.uk
  • Grant Thornton, Value for Money report on Birmingham City Council’s Oracle implementation (reported February 2025)
  • Birmingham City Council commissioners’ third report, 23 October 2025