Changelog

Released and examined from 26 September 2026. Every sitting started on or after that day draws from the v3.0 question bank and its credential records framework version 3.0. From 27 October 2026 a sitting also draws its questions by the per-competency weights in its Exam Content Outline.

Level III renamed, 5 November 2026

The third level, formerly EME-III, Engineering Management Expert, is now EML-III, Engineering Management Leader. Indicators that read "A certified Expert can …" now read "A certified Leader can …", and the level is called Leader wherever the standard names it. The reason is a name conflict found in a trademark search, not a change of content.

Nothing else changed. The twenty-nine competencies, every indicator's meaning, the outline's weights, its questions and every anchor are as they were, and the framework and outlines stay at v3.0. The level had not opened, so no sitting was examined under the old name and no credential carries it. The old pages, /certifications/eme-iii and its outline, forward to the new ones.

Body of Knowledge v3.1, 7 October 2026

A minor version under the governance rules: clarified prose, and the corrections of 8 October listed below.

The prose of every knowledge unit, the five domain hubs, the index, the glossary definitions, the technique cards and the chapters on how judgment scales, distributed work, governance and citation was edited for plainness. Sentences are shorter and more direct, and the habits that made the text read as if generated (stacked dashes, reflexive contrasts, slogan endings) are gone. Every claim, qualification, example, number, named model, evidence grade and source was meant to stay unchanged in meaning; the corrections below put back the few places where it did not. Nothing was added and nothing substantive was removed.

What did not change:

Every anchor. Each section, principle, model, tension and self-check question keeps its number, each principle and tension keeps its bold lead-in word for word, and each model keeps its name. The sentences after a lead-in were reworded with the same meaning, so a citation such as EMA Body of Knowledge v3.0, PL-1.P3 points at the same principle, saying the same thing, in v3.1.
The Competency Framework and the Exam Content Outlines. They are what is examined, a minor version may not change them, and they stay at v3.0. What a credential records is the framework version, so credentials are unaffected.
The bibliography, the glossary terms and the technique cards' names and references. One evidence grade changed, as a correction listed below.

The v3.0 text of the Body of Knowledge is not kept as a separate archive: v3.1 says the same things in plainer words.

Corrections to v3.1, 8 October 2026

A review of v3.1 against v3.0 found sentences where the plainer wording had moved the meaning. They now say what v3.0 said:

TJ-2 Technical debt: after a major incident the executive commits to the board a remediation programme. The rewrite had the executive committing the board.
PL-1 Delegation, ownership, and decision rights: in the worked scenario the failure lies in the delegation, not in the delegate; and task-relevant maturity depends on experience with the task rather than on general seniority.
PL-4 Performance management: a leader's bad quarter with an outside cause is judged on their decisions, not on the raw outcome. PL-6 Leading through leaders: leading leaders is a different job, not a larger version of leading engineers. PL-2 Feedback and difficult conversations: the upward-feedback question asks about the manager's own behaviour.
TJ-6 Security, privacy, and compliance as engineering responsibilities principle 4: the fix goes into what produces the pattern, not into today's instances. DE-3 Reliability, incident response, and operational ownership: a rising on-call load is not the team's private problem. DE-2 Estimation, planning, and honest commitments: the competency is not producing accurate estimates.
Firm statements the rewrite had softened, restored in DE-1 Prioritisation under constraint, DE-2 Estimation, planning, and honest commitments, DE-5 Designing how work flows, CC-1 Psychological safety and productive conflict, CC-2 Coaching versus directing, CC-3 Inclusive practices and equitable opportunity distribution, CC-4 Motivation, recognition, and sustainable pace, the People Leadership hub, the glossary, three technique cards and Distributed and Hybrid Work (where the 2015 study's finding is again self-reported satisfaction).

The same review checked three sources, cited in four places, against the original publications. The attributions predate v3.1:

Sabine Sonnentag and Charlotte Fritz, The Recovery Experience Questionnaire: Development and Validation of a Measure for Assessing Recuperation and Unwinding from Work (2007) is a cross-sectional study validating the recovery-experience questionnaire. It found detachment most strongly related to well-being and reports nothing on next-day performance, which CC-6 Self-management had credited to it.
Christina Maslach and Michael P. Leiter, Understanding the Burnout Experience: Recent Research and Its Implications for Psychiatry (2016) locates burnout's causes in six areas of the job and notes that interventions have favoured individual coping. It does not use the phrase "treating the person for the job's disease", which CC-4 Motivation, recognition, and sustainable pace had presented as the research's term; the unit now makes the point in its own words.
Abi Noda, Margaret-Anne Storey, Nicole Forsgren and Michaela Greiler, DevEx: What Actually Drives Productivity (2023) proposes the developer-experience framework from its authors' earlier research and reports no new study. One author, not all of them, had led DORA. DE-5 Designing how work flows and the technique card now say so, and the model's evidence grade, on the unit and on its technique card, is corrected from research to practice.

A second pass the same day found the Sonnentag and Fritz claim still in the Recovery and detachment technique card, the glossary entry Recovery and a CC-6 Self-management self-check answer; all three now state the association with well-being only. It also restored v3.0's meaning in a few more places: the risk-list ordering in a DE-4 Managing scope, risk, and dependencies across workstreams self-check (uncertainty, not severity), a DE-5 Designing how work flows self-check (the flow, not the people), the CC-6 Self-management self-care tension (the test is managerial leverage), and firm statements the rewrite had softened in DE-1 Prioritisation under constraint, DE-4 Managing scope, risk, and dependencies across workstreams, DE-5 Designing how work flows, SV-1 Engineering strategy to SV-6 Organisational design and leading change, TJ-6 Security, privacy, and compliance as engineering responsibilities, PL-2 Feedback and difficult conversations and the Technical Judgment hub.

Erratum, 8 October: the framework, the index and How Judgment Scales said all three certifications "assess the same twenty-nine competencies". They draw on the same twenty-nine; the EMA-I outline gives PL-6 Leading through leaders and SV-6 Organisational design and leading change no questions. The framework itself is unchanged.

One version line

Before this release the framework was v1.0 (June 2026) and the Body of Knowledge was v2.0 (July 2026), each numbered on its own. They are one standard: the framework says what is examined, the Body of Knowledge teaches it, and the Exam Content Outlines say how each examination samples it. From v3.0 all three carry the same version, and a major version revises them together. A minor version may revise one of them and carries its number only there, as v3.1 does for the Body of Knowledge. Framework v2.0 does not exist; the number was skipped so that a version means one release everywhere it appears.

The framework: twenty-five competencies become twenty-nine

Five domains, equal weights and the five domain URLs are unchanged.

ID	Change	What changed and why
PL-1	Expanded	Adds decision rights as an explicit object: who may decide what, and how that authority is made visible. v1.0 treated delegation as handing off work; v3.0 treats it as designing who decides.
PL-2	Expanded	Names the difficult conversation as its own skill. The delayed, avoided or badly run conversation is the failure the examinations most often test.
PL-3	Expanded	Adds assessing capability rather than tool access: interviews that measure what a candidate can do, not which tools (including AI assistants) they were allowed in the room.
PL-4	Kept	Unchanged in scope. Fair process is stated rather than implied.
PL-5	Expanded	Widened from the one-on-one to the whole growth path: promotion decisions, the senior individual-contributor track, and the choice between it and management.
PL-6	New	Built from the v2.0 Professional-altitude material of PL-1, PL-4 and PL-5, which described leading managers as a variant of leading engineers. It is a different job with its own failure modes, and now has its own competency.
DE-1	Kept	Unchanged.
DE-2	Sharpened	Re-stated around uncertainty. The competency is not producing accurate estimates; it is making commitments that are honest about what is not known.
DE-3	Expanded	Reliability as an ongoing responsibility joins incident response: service-level objectives, the on-call system, and what an organisation learns from failure.
DE-4	Kept	Unchanged.
DE-5	Expanded	Widened from choosing a process to designing the flow of work: developer experience and AI-assisted workflows are part of what a manager designs.
SV-1	Expanded	Adds writing it down. A strategy that exists only in conversation cannot be argued with, delegated against or checked.
SV-2	Expanded	From communicating decisions outward to partnering: the relationship with product, design and executives is two-way.
SV-3	Kept	Unchanged.
SV-4	Expanded	Renamed from resource allocation and build-vs-buy to engineering economics: cost of delivery, budgets and headcount, vendors, and a third option beside build and buy.
SV-5	Expanded	Adds interpretation and the measurement of AI impact, where the pressure to show a number is strongest and the numbers are least settled.
SV-6	New	Built from the v2.0 Expert-altitude material of PL-1, which treated organisational structure as the executive form of delegation. Designing the organisation and leading it through change is a strategic competency and sits in Strategy & Vision.
TJ-1	Kept	Unchanged.
TJ-2	Kept	Unchanged.
TJ-3	Expanded	Adds standards for AI-generated code: what review must establish when the author did not write every line.
TJ-4	Merged	v1.0 TJ-4 (staying technically credible) and TJ-5 (judging when to intervene) are one judgment: you can only decide when to step in if you are close enough to see, and staying close is only worth it for the decisions it enables.
TJ-5	New	AI-assisted engineering: how an organisation adopts AI assistance, governs its use, builds the capability to use it well, and manages the risks. No v1.0 competency covered it. The ID TJ-5 is reused for a different competency; the v1.0 TJ-5 lives on inside v3.0 TJ-4.
TJ-6	New	Security, privacy and compliance as engineering responsibilities. v1.0 left them implicit inside quality and risk.
CC-1	Kept	Unchanged.
CC-2	Kept	Unchanged.
CC-3	Kept	Unchanged.
CC-4	Expanded	Adds distributed and hybrid teams, where recognition and pace are hardest to see.
CC-5	Expanded	From modelling values to exercising ethical judgment: the situations where stated values and the pressure of the moment point in different directions.
CC-6	New	Self-management: the manager's own time, energy, boundaries and resilience. No v1.0 competency addressed the manager as the object.

The Body of Knowledge: one text per competency

v2.0 wrote each competency as core concepts followed by three altitude bands (At Associate, At Professional, At Expert). v3.0 replaces the seventy-five bands with one unified text per competency, structured identically across all twenty-nine, and moves the treatment of altitude into a single chapter, How Judgment Scales. Nothing substantive from v2.0 is dropped; band material is de-duplicated and re-homed, most visibly the "developing managers" content into PL-6 and "designing ownership through structure" into SV-6.

Every unit section and every principle has a permanent anchor (PL-1.P3), so the standard can be cited at the level of a single claim. Every named model carries an evidence grade ([research], [practice] or [contested]), and every source is a real, checkable work in one bibliography.

The v2.0 text remains readable, unchanged, in the archive.

Exam Content Outlines: new

Published for the first time, one per certification. Each states the examination's specification, the number of questions each competency carries, the proficiency indicators a certified holder meets, the scenario contexts questions draw on, the cognitive mix, what the examination does not assess, and the traceability of every competency to the Body of Knowledge. The EMA-I outline is complete; the EMP-II and EML-III outlines are complete drafts and say so until those examinations open.

Governance

A governance chapter states the separation of study material from examination development, the release cadence (annual major, quarterly minor), the public-comment process for major versions, the review statement, and the licence.

2026-10-05, within v3.0: the Exam Content Outlines' proficiency indicators are released under CC BY 4.0, joining the competency list; the rest of the standard stays CC BY-NC-ND 4.0. A How to Cite chapter is added.

Credentials

Credentials record the framework version they were examined against. Every credential issued before the v3.0 effective date carries 1.0; the verification page and the verification API show it. Existing credentials are unaffected in every other respect.

Retired

The document's former four-letter acronym, which collided with another body's long-standing one. The document is the Engineering Management Academy Body of Knowledge, short form EMA Body of Knowledge.
The Preparing for the Exams chapter. Its guidance is carried by each Exam Content Outline's opening section and by the per-certification reading paths generated from the outlines.
The altitude-band structure, as above.