A capability investment is the only kind of asset purchase where a large share of what you bought can resign.
A bank approves a delivery capability uplift. Four million dollars over two years: a documented method, a tooling refresh, certification for eighty project managers, and an uplift in the organisation's assessed maturity. The business case promises fewer overruns, better quality, lower risk exposure and improved productivity.
Two years later the money is spent. The method exists. The tooling is installed. Seventy-one of the eighty are certified. The maturity assessment has moved. And the executive who approved it cannot answer a simple question from the board: what do we now own that we did not own before?
The question is not rhetorical, and it is not hostile to the investment. It is the question that should have been answered at approval, and it almost never is — because capability spending is appraised as a cost with a promised benefit rather than as an acquisition with a retained asset.
A note on the reasoning in this article. The claims below about portfolio behaviour are ERANORTH's argument, not findings drawn from evidence. The source material behind this piece is project-level and contains no portfolio data. Treat what follows as a way of framing a decision, and test it against your own numbers.
The Strategic Context
Every large organisation funds delivery capability. Methods, offices, tooling, training, certification, maturity programs. The spending is rarely enormous relative to the portfolio it supports, which is precisely why it receives less scrutiny than a comparable capital item — a four-million-dollar plant investment would be interrogated far harder than a four-million-dollar capability program.
The scrutiny gap has a specific cause. Plant is obviously an asset; capability is felt to be an improvement. Assets get depreciation schedules, ownership, and a residual value question. Improvements get a benefits statement and a project closure report.
But some of what a capability investment buys genuinely is an asset — retained, transferable, and available to the organisation after the people who created it have gone. And some of it is not. The distinction is knowable in advance, and drawing it changes what an organisation should be willing to fund.
What Leaders Commonly Misread
The first misreading sits in the standard benefits claim, and it is worth quoting precisely. The conventional formulation is that when applied effectively, project management can ensure projects are delivered on schedule and within budget, improve the quality of deliverables, minimise exposure to risk, and provide greater productivity and therefore greater profitability.
Three problems, in ascending order of importance.
Ensure is too strong. No management discipline ensures a schedule outcome; it improves the probability and improves the information available when the probability turns bad.
The conditional does more damage than the overstatement. When applied effectively makes the claim close to unfalsifiable — every disappointing result can be attributed to ineffective application rather than to a limitation of the approach. A proposition that cannot fail cannot be tested, and a business case built on one cannot be appraised.
And greater productivity and therefore greater profitability does not follow. Productivity gains convert to profit only under conditions: where freed capacity can be redeployed to work that earns, where demand exists to absorb additional output, or where cost can actually be removed rather than merely made idle. In an organisation whose constraint is demand rather than delivery capacity, a productivity improvement in delivery produces a better-run portfolio and no additional profit at all. That may still be worth buying. It is not what was promised.
The second misreading is that certification is capability. Certification attaches to a person. It travels with them, including out the door. This is not an argument against training people — a workforce that has been developed is more effective and more likely to stay. It is an argument for knowing which part of the spend produces something the enterprise holds and which part produces something an individual holds, and for being clear-eyed that the second part is a recruitment and retention investment rather than a capital one.
The third misreading is that a maturity score is the answer. The instruments in common use — and the one that appears in the source material behind this article is explicitly labelled an example model rather than a standard — grade a project manager, not an organisation. The five levels run from a technical manager with a superficial grasp of project management principles, through basic awareness, to a project-focused manager with formal study, to an integrated manager defined by formal qualification, consistent use of a common methodology, proactive management of all aspects of the project, and consistent application of general management skills across the internal and external environments, and finally to a level defined by mentoring, ongoing professional development, and contribution to the organisation's continuous improvement process.
Read that list again with the investment question in mind. It is a description of individuals. Four of the five levels describe what a person has learned and how they behave. Only the fifth mentions anything the organisation retains — and it retains it only while that person is present.
Using an aggregate of individual maturity as evidence for an organisational capability investment is a level-of-analysis error, and it is extremely common. Whether conformance to a method is a good proxy for capability at all is a separate argument, taken up in [Related article: Why "The Principles Apply to Any Project" Is Only Half True]. The point here is narrower: even a perfectly good individual instrument does not measure the thing being purchased.
Reframing the Issue
The reframing is to appraise capability spending the way you would appraise an acquisition: what is retained, by whom, and for how long?
Four categories, and the proportions differ sharply between them.
Codified assets. Method documentation, templates, decision instruments, tooling and configuration. Fully retained. Also, on their own, close to worthless — a method nobody follows is a folder. Their value is entirely a function of whether the third category exists.
Embedded practice. The habits, defaults and shared expectations of teams. Retained as long as the teams persist, and highly resistant to individual departure. This is the most valuable category and the least often funded explicitly, because it is produced by doing the work differently under supervision, not by attending a course.
Individual credentials and expertise. Not retained. Valuable, worth buying, and correctly classified as a people investment rather than a capability asset. The most consequential instance is the transition of specialists into delivery leadership, which most organisations fund poorly and which is examined in [Related article: From Specialist to Delivery Leader: The Promotion That Is Actually a Career Change].
Institutional memory of decisions. The record of what was tried, what failed, and why — the only category that compounds over time, and the one most reliably destroyed by reorganisation and by the dissolution of project teams, as discussed in [Related article: The Hidden Cost of Putting Work Into Project Form].
An investment weighted toward the first and third categories buys documentation and trained individuals. An investment weighted toward the second and fourth buys something the enterprise keeps. Most capability business cases are written for the former and justified as the latter.
Strategic Analysis
The opportunity cost is a portfolio question
Four million dollars of capability spending is four million dollars not allocated to a revenue initiative, a resilience investment, or debt reduction. That is obvious. What is less obvious is the second cost: a capability program consumes the attention of exactly the people the portfolio most needs — senior delivery leaders, sponsors and the operational managers whose participation makes practice change real. Capability programs that succeed are expensive in attention; ones that are cheap in attention do not change practice.
An organisation running near its capacity limit should therefore be more, not less, sceptical of a large simultaneous capability program, because the attention it requires is drawn from the same constrained pool as delivery itself.
What a good capability case looks like
It states which of the four categories the money is buying, in proportion. It names the practice that will be different and how anyone would know. It identifies the retained asset. And it makes a falsifiable claim — not improved delivery performance but something with a number, a date and a comparison to the current record. The discipline of testing such a claim before approving it is examined in [Related article: Enough Technical Depth to Test the Answer], and capability cases deserve the same interrogation as any other forward-looking claim.
What it does not need to do is prove that the money will change specific decisions. Whether a governance activity produces a decision or merely produces evidence of activity is an important question with its own treatment in [Related article: What a Stage Gate Is Actually For], and it is not the test being applied here.
The most under-priced capability is usually the narrowest
There is a consistent asymmetry worth noticing. Broad capability programs — method, training, certification across a whole population — are expensive, slow, and produce results that are hard to attribute. Narrow ones — a specific instrument taught to the twenty people who make continuation decisions, a single new artefact required at initiation, one measure added at a handover — are cheap, fast, and produce attributable changes.
There is also a prior question that capability cases rarely ask: whether the method being funded suits the work the organisation actually does, or was designed for a different kind of work altogether. That inheritance is traced in [Related article: What Kind of Work Were These Instruments Built For?].
The narrow interventions are less often funded because they are unimpressive in a paper. They also tend to be the ones that survive, because they change what the organisation requires rather than what it knows.
Decision Framework
Six questions for any capability investment above a material threshold.
1. What is retained? Answer in the four categories, with rough proportions. If the honest answer is that most of the spend is codified assets and individual credentials, that is a legitimate investment — appraise it as documentation and training, and price it accordingly.
2. What practice will be different, and who would notice? Name the behaviour, the people, and the observer. Answers that describe awareness, understanding or alignment are not answers.
3. What is the falsifiable claim? One sentence, with a number, a date and a comparison to current performance.
4. Does the productivity chain close in our circumstances? If the investment promises profitability through productivity, establish whether the freed capacity can be redeployed to earning work or removed as cost. Where neither is true, the benefit is a better-run portfolio — worth having, but a different case.
5. What attention does this require, from whom, and what does that displace? Capability programs are paid for twice, in money and in the scarce attention of the people the portfolio needs.
6. What is the residual value in three years? Under a reorganisation, a change of delivery model, or thirty per cent turnover in the trained population, what survives? The answer separates a capability asset from a training expense.
From Strategy to Execution
Immediate. Take the largest capability investment currently in flight and answer question one. Most organisations find the spend is heavily weighted toward codified assets and individual credentials, with embedded practice funded implicitly if at all. That finding is actionable within the existing budget: shifting even a modest proportion toward supervised practice change alters what remains at the end.
Medium term. Change the appraisal template. A capability case should be required to state its retained asset, its falsifiable claim, and its attention cost. This is three additional fields and it does more to improve the quality of these investments than any amount of governance.
Long term. Treat institutional memory as an asset with an owner. The record of what was tried and why is the only category that compounds, and in most organisations it is nobody's responsibility and disappears with every restructure. An enterprise that keeps it makes better allocation decisions for a decade; one that does not relearns the same lessons at full price.
Signals to Monitor
- Practice reverting after the program ends. Track use of a small number of specific artefacts twelve and twenty-four months after implementation. Decay is the clearest evidence that codified assets were funded and embedded practice was not.
- Maturity scores rising while delivery outcomes do not. A widening gap between the assessment and the portfolio's actual performance indicates the instrument is measuring conformance rather than capability.
- Certification rates as the headline metric. When the reported measure is how many people hold a credential, the program is reporting on the least retained category.
- Turnover among the trained population. Departures in the first eighteen months after certification directly erode the individual-credential portion of the investment, and should be tracked against it explicitly.
- Capability spend rising while portfolio throughput is flat. Worth watching as a simple ratio over several years. It is not proof of anything, but a sustained divergence is a question that deserves an answer.
Questions for the Leadership Team
- For our largest capability investment, what do we now own that we did not own before — stated as an asset, not as an improvement?
- What proportion of that spend produced something an individual can take with them when they leave?
- Does our productivity-to-profitability chain actually close in this business, and has anyone tested it rather than asserted it?
- Are we using an instrument that grades individuals as evidence for an organisational capability position?
- What attention did our last capability program consume, and what did the portfolio not get as a result?
- If we reorganised next year, how much of what we bought would survive?
Closing Perspective
Capability investment is one of the few areas of enterprise spending where the standard business case does not have to say what is being acquired. A plant purchase names the asset. A software licence names the term. A capability program names a benefit and leaves the asset unspecified — and so the organisation ends up two years later with a method nobody follows, a cohort of certified individuals, some of whom have left, and a maturity score describing people rather than the enterprise.
None of that is fraudulent, and none of it means the investment was wrong. It means it was never appraised as an acquisition, and so nobody drew the line between the part the enterprise keeps and the part that walks out the door.
Draw that line before the money is committed. It is a single question, it takes an afternoon, and it changes what capability programs are designed to do.
About the author
Kevin Jogin is Founder & Principal Advisor at EraNorth. Meet the Founder.