On this page
The Digital Academy launched with a town hall, a logo and a landing page. Eighteen months later it had 4,000 enrolments, a 71% completion rate, and no discernible effect on anything.
Nobody shut it down. It just stopped being mentioned. The operations director who asked for it had moved on, and the two L&D managers running it were producing completion reports nobody read.
Josh Bersin, who named the capability academy in 2019, put the diagnosis in one line: an academy is a capability-centric organisation that should be led by a business leader, not an L&D professional. This one never was.
The content was good. The platform worked. What went wrong was decided long before launch, in a meeting where the business handed the problem to L&D and called that sponsorship.
TL;DR. A capability academy is a governance structure for building one business-critical capability, owned and funded by the business leader who needs it. The term came from Josh Bersin in 2019, and his defining constraint is that an academy led by L&D rather than by a business leader is a course catalogue with better branding. Most capability gaps do not justify an academy. Knowing which ones do is the whole decision.
Key highlights
- A capability academy is organised around a business capability, not a job family or a department.
- Business ownership is the definition, not a success factor. Without it the structure reverts to a catalogue.
- Most gaps are better answered by a course, a hire, or a process change. Academies are expensive and should be rare.
- The trigger for building one is a capability that is strategic, cross-functional, and absent from every existing role.
- Sponsorship means holding the budget and the target, not appearing at the launch.
What is Capability Academy?
A capability academy is a structure for building one specific business capability, bringing together content, practice, experts, assessment and credentials around it, and owned by the business leader accountable for that capability.
Short answer. A capability academy organises learning around a capability the business needs rather than around a job title or a department. Josh Bersin, who introduced the term in 2019, describes it as a capability-centric organisation that should be led by a business leader rather than an L&D professional.
The distinction that matters is not the content model. It is who holds the budget and the target.
Simple. A training programme asks what people should learn. A capability academy asks what the business needs to be able to do that it currently cannot, then organises everything around closing that.
Three things separate an academy from a well-branded curriculum.
- It is organised around a capability, such as underwriting quality, digital selling or plant automation, not around a function or a grade.
- It is sponsored by the executive who owns the business outcome, who also owns the budget.
- It mixes formal content with practice, expert access and assessment, because capabilities are demonstrated rather than completed.
None of that requires a platform, a brand or a launch event. All three are what organisations reach for first, which is why so many academies are catalogues wearing a logo.
How is Capability Academy connected to your workplace?
The connection is not through the learning function. It runs through whatever the business is trying to change.
Consider a manufacturing group moving from scheduled maintenance to condition-based maintenance across eleven plants. The capability required does not sit in any existing job description. Maintenance technicians need to read sensor data. Planners need to schedule against predicted failure rather than a calendar. Plant managers need to defend a maintenance decision to a finance team that has never seen one justified this way. Procurement needs to buy differently.
Four functions, one capability, no single owner. That is the shape of a problem an academy is built for, and it is also why a departmental training plan cannot solve it. Each function would train its own people correctly and the capability would still not exist, because the handoffs between them are where it lives.
| Element | A departmental training plan | A capability academy |
|---|---|---|
| Organising unit | Function or job family | One business capability |
| Owner | L&D, with a functional stakeholder | The executive accountable for the outcome |
| Budget source | The learning budget | The business or transformation budget |
| Scope | People inside one function | Everyone who touches the capability |
| Success measure | Completion and satisfaction | Movement in the business metric the capability exists to shift |
| Lifespan | Annual cycle | Until the capability is embedded, then it closes |
The last row is the one most organisations never plan for. An academy that never closes has become a department.
Mercer's Global Talent Trends 2026 gives a sense of how much of this work is now queued up. Across nearly 12,000 respondents, 98% of executives said they are planning organisational design changes within two years, and 65% expect between 11% and 30% of their workforce to be redeployed or reskilled because of AI in that window. Redeployment at that scale is not a training problem attached to existing roles. It is capability construction, and it lands on the CHRO whether or not the structure exists to absorb it.
Why do industries need to map Capability Academies?
Mapping matters because the alternative is building one per request, and academies are expensive enough that a portfolio assembled by accident becomes a cost problem within two years.
The mapping question is not "where are our skill gaps". It is narrower and harder: which capabilities are strategic enough, cross-functional enough and durable enough to justify a standing structure rather than a project.
Industry shapes that answer more than most frameworks admit, and the differences are not cosmetic. They change what the academy is for, how it is funded, and what eventually kills it.
| Sector | Capability that typically qualifies | What is different about the academy | What usually kills it |
|---|---|---|---|
| Banking and insurance | Advisory selling, underwriting judgement, financial crime detection | Built around a product generation or a regulatory regime, so it has a natural expiry and must be designed to close | Funding it as permanent when the capability was tied to a product cycle |
| Manufacturing | Plant automation, condition-based maintenance, quality systems | Spans shop floor, planning, engineering and finance, so it lives or dies on the handoffs rather than on any one group | Training operators well and leaving planners and finance out |
| Healthcare | Clinical protocol adherence, patient safety, credential currency | The requirement never ends and the evidence trail has to be continuous, so it is a standing structure by necessity | Collapsing it into compliance reporting until it stops building capability at all |
| Retail and hospitality | Service standards, in-store selling, shift leadership | Turnover means the output is a reproducible path to proficiency, not a trained cohort | Measuring the cohort instead of the time it takes a replacement to reach the bar |
| IT services | Emerging platform skills, solution architecture, delivery leadership | Demand moves with the project pipeline, so the mapping refreshes against bid and demand signals rather than an annual plan | An annual planning cycle attached to a pipeline that turns over quarterly |
Two patterns are worth pulling out of that table, because they cut against how most academies are budgeted.
- In healthcare and regulated manufacturing, the capability requirement does not expire, so the academy is permanent and the risk is drift into pure compliance training reporting. Guard the capability-building half explicitly or it erodes.
- In banking, insurance and IT services, the capability is tied to a product generation or a pipeline, so the academy should be funded as a fixed-term structure. Funding it as permanent is how organisations end up with three academies nobody can close.
- In retail and hospitality, the economics invert the usual measure. Success is how fast a new joiner reaches the bar, not how many people completed the programme, because the population turns over faster than the programme runs.
EY's 2026 CEO Outlook, a survey of 1,200 chief executives conducted in March and April 2026, found that when CEOs identify what is holding back value from their AI investments, they name limited skills, leadership capability gaps, weak learning infrastructure and cultural resistance. Three of those four are capability problems rather than technology problems, and none of them resolve through a course catalogue.
Mistakes. The most common mapping error is starting from the org chart. Capabilities that need an academy almost never map to a single function, and the ones that do map neatly are usually well served by existing departmental training. If your proposed academy fits inside one reporting line, it probably does not need to be an academy.
what way to find and sort Capability gap in workplace?
Finding gaps is the easy half, and most organisations already have the inputs: performance data, exit interviews, project post-mortems, audit findings, unfilled vacancies, manager escalations. The difficulty is sorting, because every gap that surfaces arrives with somebody arguing it is urgent.
A gap earns an academy only when it clears four tests at once.
- Strategic. The capability sits on the path to a stated business objective, and someone on the executive team can name the objective without checking.
- Cross-functional. More than one function has to be good at it for the capability to exist. If one team can carry it alone, train that team.
- Absent. The organisation cannot currently buy the capability in at reasonable cost or speed. If the labour market will sell it to you in ninety days, hiring is cheaper.
- Durable. The capability stays relevant long enough to repay a standing structure, generally more than two years.
A gap that clears all four is rare. Most enterprises running this test honestly find two or three, not a dozen.
| What the gap looks like | Best answer | Why not an academy |
|---|---|---|
| One team missing a defined skill | Targeted training for that team | No cross-functional dependency to coordinate |
| Capability available in the market at acceptable cost | Hire or contract | Building is slower and more expensive than buying |
| Failures traced to a broken process rather than to people | Fix the process | Training people to compensate for a bad process entrenches it |
| Short-lived requirement tied to one project | Project-based enablement that ends with the project | A standing structure outlives its purpose |
| Strategic, cross-functional, absent, durable | Capability academy | This is the case the structure exists for |
Market. The same Mercer research found that C-suite confidence in organisational readiness for the human-machine era fell to 51% in 2026 from 65% in 2024, and that while 75% of leaders say their organisation must become more digital to compete, only 30% rate their digital agility as high. A gap the executive team already knows it has is a far easier academy to fund than one L&D has to prove exists.
One practical note on sequencing. Run the sort before the discovery exercise if you can. Organisations that map every gap first and prioritise second end up defending a list rather than making a decision, and the list itself becomes the deliverable.
What are the posible steps to industry have do Capability Academy?
Six steps. None require a platform, and the first three are where the outcome is actually decided.
- Name the capability and the business metric it moves Write one sentence: the capability, and the number it changes. If the sentence needs a subordinate clause to stay true, the capability is too broad to build around. Condition-based maintenance reducing unplanned downtime is buildable. Operational excellence is not.
- Get a business leader to own it, with the budget This is the step organisations skip, and skipping it is the reason most academies drift. Sponsorship means holding the budget and being accountable for the metric in step one. An executive who attends the launch and approves a business case is not a sponsor. If nobody outside HR will take that, you have learned something useful and should stop here.
- Define what good looks like at three levels Describe observable behaviour for someone starting, someone competent and someone others learn from. Write these with practitioners, not with a competency library. The middle level is the one that matters, because that is what the academy is actually producing.
- Map who touches the capability, across every function List every role involved in the handoffs, not just the obvious ones. The maintenance planner and the finance analyst belong in the condition-based maintenance academy even though neither will touch a sensor. Capabilities fail at the seams.
- Build practice and expert access before content Content is the cheapest component and the one most academies over-invest in first. Identify who internally already has the capability, and design how their time gets used, because that access is the constraint that determines scale.
- Set the closing condition at the start Decide in advance what would prove the capability is embedded, and what happens to the structure then. An academy without a defined end becomes permanent overhead, and permanent overhead is what gets cut in the next budget round regardless of whether it is working.
Simple. If you can only do one of these properly, do step two. An academy with a real business owner and mediocre content outperforms a well-built academy with a nominal sponsor, every time.
The uncomfortable part of step two deserves saying directly. If no business leader will hold the budget, that is a decision about the capability, not a failure of persuasion. It usually means the capability is not as strategic as the proposal claims, and running the academy anyway from the learning budget produces the eighteen-month death described at the start.
How disprz easy to map Capability gap?
Everything above is platform-independent, and a first academy should usually be run that way. The test of whether tooling earns its place is whether the business sponsor can see capability movement without asking L&D to produce a report, because a sponsor who has to request evidence stops requesting it.
Disprz addresses three parts of that specifically.
Capability measured as an index rather than as completions. Journey scorecards, capability indexes and readiness scores give a view of where a population stands against a defined capability bar. For an academy sponsor, that is the artefact that replaces the completion report, and it is the difference between reporting activity and reporting capability.
A cross-functional view, which is where academies actually break. Skill heat maps roll up by skill family, category and individual skill across departments, and org-wide readiness reporting shows the same capability across the functions that touch it. Step four in the sequence above is the one that is hardest to maintain by hand, because it spans reporting lines that do not share a system.
Assessment that does not depend on self-reporting. Skill benchmarking uses adaptive assessments, and 360 assessments bring in observed input rather than only self-rating. Capability claims built on self-assessment are the fastest way to lose a sponsor's confidence once the business metric fails to move.
Where the capability is expected to open career routes, mapping it to career paths informed by employee aspiration keeps the academy connected to retention rather than sitting beside it.
None of it substitutes for step two. Tooling makes a sponsored academy measurable. It does not make an unsponsored one work.
The Digital Academy in the opening had everything except an answer to one question: who loses their bonus if this does not work?
That question is unpleasant, and asking it early is the cheapest diagnostic available. An academy that cannot answer it is a course catalogue that has not found out yet. An academy that can has already cleared the only barrier that reliably predicts whether these structures survive contact with a budget cycle.
So before the platform conversation, the content plan or the launch, find out whether anyone outside HR is willing to be accountable for the capability. The answer tells you whether to build.
Frequently asked questions
What L&D and business leaders ask most often about capability academies.
What is a capability academy?
A capability academy is a structure for building one business-critical capability, combining content, practice, expert access and assessment, and owned by the business leader accountable for the outcome. Josh Bersin introduced the term in 2019 and defines it as capability-centric rather than course-centric, led by a business leader rather than by L&D.
How is a capability academy different from a corporate university?
Corporate universities were organised around functions and delivery, often with a physical campus and a broad curriculum. Capability academies are organised around a single business capability, sponsored by the executive who owns that outcome, and designed to close once the capability is embedded rather than to operate indefinitely.
Who should own a capability academy?
Ownership belongs to the business leader accountable for the capability and the metric it moves, with L&D acting as architect and operator. Bersin's example is Capital One assigning its CIO to lead the Digital Academy. An academy owned by L&D alone tends to revert to a course catalogue within about two years.
Which capability gaps justify building an academy?
Gaps justify an academy when the capability is strategic, spans more than one function, cannot be bought in at acceptable cost or speed, and stays relevant beyond about two years. Gaps failing any of those four are better served by targeted training, hiring, a process fix, or project-based enablement that ends.
How many capability academies should an organisation run?
Most enterprises running the four tests honestly find two or three capabilities that qualify, not a dozen. Academies carry real standing cost in sponsor attention and expert time, so a large portfolio usually signals that the criteria were applied loosely rather than that the organisation has unusual needs.
How do you measure whether a capability academy is working?
Measurement tracks the business metric named at the start, supported by capability indexes showing movement against a defined proficiency bar. Completion rates indicate consumption rather than capability. A sponsor who cannot see capability movement without requesting a report will disengage well before the programme formally ends.
When should a capability academy be closed?
Closure should be defined before launch, tied to evidence that the capability is embedded in normal operations and reproducible without the structure. Academies with no defined end become permanent overhead, which makes them vulnerable in budget cycles regardless of performance, and obscures whether the original capability was ever built.
