Before you commission a maturity assessment, answer one question: who reads the number? The honest answer determines what the exercise will produce, and it is rarely the answer given in the proposal.
An engineering services firm commissions a project management maturity assessment. The stated purpose is continuous improvement. The actual trigger was a tender document from a major client requesting evidence of assessed capability, and the real deadline is the bid submission date.
Nothing improper has occurred. The firm does want to improve, the client's request is reasonable, and an assessment will serve both purposes to some degree. But the two purposes make different demands, and where they conflict the deadline wins. An assessment run for a bid will be scoped to what can be evidenced, timed to the submission, and reported in a form a client can read. An assessment run for improvement would be scoped to where the firm suspects it is weakest, timed to the improvement cycle, and reported in a form that is uncomfortable.
The first produces a rating. The second produces a work programme. Very few organisations get both from one exercise, and almost none decide in advance which they are commissioning.
The Strategic Context
Rad and Levin, writing on project management maturity assessment in the 2006 AACE International Transactions, do something unusual for the genre: they set out plainly why organisations actually seek a rating. Their list is organised in the paper under a figure titled "Motivational Pressures", and it is commercially frank in a way most treatments are not.
The motivations they name include the proactive internal desire for improvement; reactive pressure arising from past disasters, a very recent crisis, or an ongoing crisis; a client's wish to assure the success of a project being outsourced; a client requirement to screen a certain number of prospective contractors; a client's interest in minimising the effort required to monitor the project; competitive pressure arising when a competitor has just announced that it has reached a particular level of maturity; merger due diligence, where the two parties must verify certain capabilities; evaluation of a subcontractor or partner organisation; entry into new lines of business; assurance that teams can perform in virtual mode; and use as an enterprise credential for highlighting one contractor among the members of a bidder pool.
Read that list as an executive rather than as a practitioner. Roughly half of those motivations have an external audience. The assessment is being performed so that a number can be shown to somebody — a client, a bidder panel, an acquirer, a partner.
The authors are equally direct about what happens next. Reactive motivations, they observe, tend to fade once the crisis is over. And they close with a warning that ought to be quoted more often than it is: assessments should not become an end unto themselves.
What Leaders Commonly Misread
The first misreading is that the purposes are compatible. They overlap and are not the same. An improvement assessment is most valuable where it is most unflattering, because that is where the work is. A credential assessment is most valuable where it is defensible, because that is what will survive scrutiny. Point the same instrument at both and the scoping conversation will resolve, quietly and reasonably, toward what can be evidenced — which is not where the problems are.
The second misreading is that an external audience makes the exercise dishonest. It does not. Producing a credential for a client or a bidder panel is a legitimate commercial activity, and a firm that does it well is answering a real market requirement. The error is not in doing it; the error is in calling it improvement, because that mislabelling means the improvement work is recorded as done when it has not been commissioned.
The third misreading is that a rating that rises is an organisation that improved. A rating can rise because practice changed, because evidence improved, because the assessor changed, or because the scope was drawn around the stronger parts of the business. Only one of those is improvement, and the rating alone does not distinguish them.
A fourth misreading concerns durability. Rad and Levin's observation about reactive motivation fading is easy to skim past and is the most practically useful sentence in their paper. An assessment commissioned in the wake of a failure carries organisational energy for perhaps two quarters. If the improvement programme it generates is designed to run for eighteen months, most of it will be abandoned — not by decision, but by the quiet withdrawal of attention once the crisis recedes.
Reframing the Issue
The reframing is to name the audience before commissioning, and to accept that this determines almost everything else about the exercise.
Three audiences, three different instruments.
For an external counterparty — client, bidder panel, acquirer, partner — the requirement is a defensible, comparable, evidenced statement of capability at a point in time. Scope it to what is genuinely assessable, use a recognised instrument if the counterparty has named one, and treat it as a compliance and evidence exercise. It should be run by whoever runs bids or assurance, and it should be budgeted as a cost of sale.
For internal improvement — the requirement is a diagnosis. It should be scoped to the areas the leadership team already suspects are weak, should not produce a single headline number, and should be run by or for the people who will act on it. Its output is a prioritised work programme with owners, not a level.
For an internal argument — securing funding, settling a dispute between functions, justifying a restructure — the requirement is evidence for a decision that is already in play. This is a legitimate use and it is the one least often admitted, which matters because an assessment commissioned to win an argument will be read as such by everyone whose function it examines, and their cooperation will reflect that.
Two consequences follow.
If the audience is external, do not expect improvement as a by-product. Budget the improvement work separately or it will not happen.
If the audience is internal, do not produce a number. A single level invites comparison, and comparison invites the credential use, and the credential use will take the instrument over.
Strategic Analysis
What the model itself is measuring
The five-level organisational model in the teaching material behind this dataset runs: Common Language, Common Processes, Integrated Methodology, Continuous Improvement, Benchmarking. It is described in its source as an example model, carries no attribution, and its level names correspond closely to a widely published model [SOURCE DETAILS REQUIRED]. It should never be presented as a standard, and this article does not.
Two features of it are worth an executive's attention because they are counterintuitive and rarely remarked on.
A project office appears only at Level 5. In this model, establishing a project office belongs at the top of the scale, alongside benchmarking against processes, methodologies and cultures across industries. That placement sits oddly with practice, where a project office is frequently among the first things an organisation establishes — often at the point corresponding to Level 2. An organisation reading its position against this model would conclude either that it is more mature than it is, or that it built its office prematurely. Neither conclusion is safe, and the divergence is a good illustration of why a scale should be understood before it is scored against.
"Informal project management" is a Level 3 characteristic. It appears in the same list as integrated processes and cultural support. Read carelessly this looks like a defect in the model; read carefully it is the most interesting entry in it. What it appears to describe is an organisation whose practice has been internalised sufficiently that it no longer requires formal enforcement — the point at which method becomes habit. If that reading is right, the model is asserting that heavy formality is a mid-scale condition and that maturity involves shedding it. That is a claim most maturity programmes act against.
A third feature is a straightforward inconsistency worth recording: the decks place "Cultural support" at Level 3, and a later capture of the same material lists it at Level 4 [SOURCE DETAILS REQUIRED]. Small, but a reminder that an unattributed example model is not a fixed instrument.
What the scale does not reach
No maturity instrument in common use examines how finely an organisation divides its specialists across concurrent commitments, which is among the most reliable constraints on delivery — the arithmetic is set out in [Related article: There Is No Such Thing as Half a Project Manager]. An assessment can return a high rating for an organisation whose real capacity problem it never looked at.
The commercial logic of the credential, and its limit
The bidder-pool motivation deserves examining on its own terms, because it is the one with the clearest commercial return and the least discussed downside.
Rad and Levin's argument is that a maturity rating provides an objective and methodical means by which a prospective contractor can claim superiority over another, and that the higher-rated bidder can claim greater credibility in its assertions during proposal dialogue. In markets where bids are received globally and prior experience with a given contractor may be rare, that is a real advantage.
The limit is that the same logic applies to every bidder, and a credential that all serious bidders hold selects for nobody. Ratings that begin as differentiators become entry conditions, and the cost of maintaining them becomes a cost of doing business rather than a source of advantage. An organisation should therefore be clear about which phase it is in: pursuing a rating early, while it differentiates, is an investment; pursuing it late, when everyone has one, is overhead — and both are legitimate as long as the enterprise knows which it is buying.
Assessment is not the same as control
A maturity rating describes the state of an organisation's practice. It does not govern any particular initiative, and it is a poor substitute for the instruments that do — bounded, expiring delegations of the kind examined in [Related article: Authority With an Expiry Date]. Organisations sometimes pursue a rating in place of building those instruments, which is an expensive way to avoid a cheaper decision.
The half-life of a crisis
Rad and Levin's observation that reactive motivation fades has a direct design consequence.
If an assessment is commissioned in the wake of a failure, the improvement programme it produces should be designed to complete its highest-value work inside two quarters. Not sequenced by logical dependency, which is the natural instinct, but sequenced by what will still command attention when the memory of the failure has gone. In practice this means front-loading the two or three changes that will outlive the energy, and treating everything else as conditional.
This is unglamorous and it is the difference between an assessment that changes something and one that produces a report. Most reactive improvement programmes fail not because the diagnosis was wrong but because they were scheduled as though attention were a constant.
Decision Framework
Five steps, taken before the assessment is commissioned rather than after.
1. Name the audience in one sentence. We are doing this so that [who] can [do what]. If the sentence names more than one audience, split the exercise or choose.
2. Match the instrument to the audience. External: recognised, evidenced, comparable, point-in-time. Internal improvement: targeted, unflattering, no headline number. Internal argument: state it openly, or expect the functions being assessed to work it out and respond accordingly.
3. Decide in advance what result would change a decision. For an improvement assessment, name the finding that would cause the leadership team to reallocate money or people. If no possible finding would do that, the exercise is documentation and should be budgeted as such.
4. Check the scale before scoring against it. Read the level definitions and ask whether the sequence matches this organisation's reality — the project-office placement above is a good test case. A scale adopted without this check will produce a number whose meaning nobody can defend.
4a. Expect the assessed functions to read the motive. An assessment lands on people who may lose from what it finds, and their cooperation will reflect what they believe it is for — the asymmetry examined in [Related article: Why the Old Order Fights Harder]. Stating the audience openly is the cheapest way to keep the evidence honest.
5. If the trigger was a crisis, design for two quarters. Identify the two or three changes that must survive the fading of attention, and sequence those first regardless of dependency logic.
A supporting rule: if the answer to step one is external, commission the improvement work separately and fund it separately. The two exercises can share evidence. They should not share a scope, a timeline or a report.
From Strategy to Execution
Immediate. For any maturity assessment currently in progress or recently completed, write the audience sentence retrospectively. Where the honest answer differs from the stated purpose, the finding is actionable straight away: it tells you whether an improvement programme was ever really commissioned.
Medium term. Separate the two functions permanently. Credential assessments belong with bids and assurance and are a cost of sale. Improvement diagnostics belong with the operating leadership and are an investment. Keeping them apart is a small organisational decision that prevents a recurring confusion.
Long term. Build the habit of assessing against outcomes rather than only against a scale. A scale tells you how your practice compares to a described progression. What an executive team ultimately needs to know is whether the enterprise delivers better than it did — which requires a small number of longitudinal outcome measures maintained across years, and which no maturity instrument supplies. What the enterprise retains from a capability investment more broadly is a distinct question with its own treatment in [Related article: What Does the Enterprise Own After a Capability Investment?].
Signals to Monitor
- Assessment timing that tracks the bid calendar. If assessments consistently complete shortly before major submissions, the audience is external whatever the stated purpose.
- Improvement programmes that stall at the two-quarter mark. The signature of reactive motivation fading, and predictable enough to design around.
- Ratings that rise without practice changing. Ask what specifically is done differently since the previous assessment. If the answer is documentation, the rating moved and the organisation did not.
- Scope that narrows during the assessment. Watch for parts of the business being excluded mid-exercise. This is usually where the finding would have been.
- Level definitions nobody has read. Ask three people who commissioned the assessment to describe what the target level requires. Disagreement indicates a number being pursued without a shared understanding of what it means.
- Competitor announcements preceding your own commissioning. Not a reason to avoid the exercise, but a reason to be clear that the motivation is competitive rather than developmental.
Questions for the Leadership Team
- Who reads the number our last maturity assessment produced, and what did they do with it?
- What finding would have caused us to reallocate money or people — and would our assessment have surfaced it?
- Are we running one exercise for two audiences, and if so, which one is quietly winning the scoping decisions?
- Has anyone on this team read the level definitions we are scoring ourselves against?
- Of the improvement actions from our last assessment, how many are complete, and what happened to the rest?
- If our whole market holds the rating we are pursuing, what will we have bought?
Closing Perspective
Maturity assessment is an unusually useful instrument that is unusually often pointed at the wrong target — not through cynicism, but because two legitimate purposes have been allowed to share one exercise, and the purpose with the deadline always wins.
Naming the audience before commissioning costs nothing and settles almost everything. It determines the scope, the instrument, the reporting format, the owner and the budget line. It also makes visible the case an organisation most needs to see: that it has been running credential assessments and recording them as improvement, and that the improvement work has therefore never been commissioned at all.
Rad and Levin's own caution is the right note to end on, and it applies to every instrument of this kind. An assessment is a means of knowing where you stand. The moment the rating becomes the objective, the instrument has been captured — and the organisation is now managing its score rather than its capability.
About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.