Who Is Ready in the Management Trainee Pool? Grounding the Promotion Decision in Observation

In most organisations a management trainee programme begins with a well-run selection stage and ends with an uncertain decision stage. The pool is formed, training is delivered, rotations are completed; then at the promotion table the question is asked: which of them is ready? The answer usually rests on three sources, and all three share one blind spot. Gallup's State of the American Manager research reports that organisations fail to choose the candidate with the right talent for the manager job 82% of the time, and notes that promotions are mostly based on seniority and performance in the previous role. In this article we examine how SimAna decision simulations fill that gap, and — just as importantly — what they do not fill.

The data we have and the data we need

The three sources on the promotion table measure a candidate's past and their ability to describe themselves. Management, however, is a future behaviour, and until it is observed it is guessed at.

The data that is needed gathers into four questions, and all four become visible only at the moment of decision: What do they sacrifice when resources are scarce? Where do they lean when safety and cost pull against each other? On a full agenda, what do they take first and what do they leave waiting? In a crisis, whom do they contact, and how quickly?

The cost of a wrongly placed management candidate is not only that person's performance; it is the renewed search, the time to reach full effectiveness, lost productivity and rising turnover in the team, and operational decisions that are delayed or made wrongly. Seeing that risk before the appointment is markedly cheaper than learning it afterwards.

A work sample, not a personality test

A decision simulation puts the candidate not into an inventory of tendencies but into the work itself. The candidate makes an organisation's real decisions and every choice writes points to the measured competencies. In industrial psychology the method is called a work sample: the candidate is asked not to describe the job but to do a scaled-down sample of it.

In practice this distinction changes two things. First, what is measured is behaviour rather than assertion; the candidate does not say "I stay calm under pressure", they do something under pressure. Second, every candidate sees the same case with the same information in the same amount of time, so the difference between them comes from the decision rather than from the case. The standard condition is what makes the comparison meaningful.

Six competencies

The competencies measured are defined before the session and mapped to the organisation's own competency model. The common framework of decision simulations has six headings:

• Decision under pressure — being able to decide while time and information are limited.
• Multi-criteria balance — holding conflicting metrics in view at the same time.
• Prioritisation — establishing an order among tasks that all look urgent.
• Resource and crisis management — allocating a scarce resource, managing the unexpected.
• Ethical judgement — deciding in line with principle and transparently in a hard situation.
• Operational awareness — being able to follow how a decision shows up in the numbers.

The six are not measured in a single session, because no single session format is broad enough to cover them all. A weekly decision cycle measures multi-criteria balance well but does not measure prioritisation at all; an inbox does exactly the opposite.

Two lenses, one pool

Assessment therefore rests not on one scenario but on two complementary lenses. The Field Operations lens measures what the candidate chooses: a weekly decision cycle, several business metrics moving at once with every decision, and the choice showing up in the balance sheet. The Decision Box lens measures what the candidate takes first: simultaneous demands, choices that cannot be taken back, and whether the chance to verify a claim was used.

The two lenses look from inside the same organisation and together they cover all six competencies. In a programme they are used in sequence; which lens enters at which stage is set against the organisation's own calendar, and starting with a narrow scope is possible too.

The scoring key is fixed in advance

The most fragile point of any assessment tool is where the score comes from. In decision simulations, how many points each option writes to each competency is determined while the scenario is authored, and no human judgement enters the session result. Two candidates who make the same decision receive the same score; the assessor's mood on the day does not enter the outcome.

The key is also not a black box. A scenario arrives with a ready scoring key, but if an organisation wants to define its own competency weights the key is built together and fixed before the session. A key that can be changed afterwards stops the measurement from being a measurement.

How the raw score is read is a separate choice. If the score is calculated only against the other candidates in the same session, comparison across periods loses its meaning, because the scale is rebuilt in every pool. When expected and target levels are defined in advance the scale stays fixed and the same number means the same thing in every period. Duration and consistency across rounds are recorded as separate signals as well.

The output: a competency map and a decision trail

Two different readings are produced at the end of a session. At the individual level the candidate sees their profile across the six competencies and how it compares with the group median. That profile is more useful than a single number: a candidate who is strong in a crisis but struggles to order the agenda in calm periods looks "average" under one overall score, and no development plan can be built from it.

Beside the profile stands the decision trail: which decision was made how, which data was consulted, how long a warning took to reach the level above. This is what can be discussed in a promotion conversation; a score ends the discussion, a decision trail starts it.

At the pool level the reading changes. A large number of candidates clustering in the same band on the same competency is an organisational finding, not an individual one. Its decision consequence shows up in three places: the development budget is directed at one shared gap rather than at the same training for everyone, promotion order is discussed against a competency profile rather than a single score, and the same framework stays comparable with future pools.

Comparison with the alternatives

A personality inventory, an assessment centre and a decision simulation are not rivals; they answer different questions. Which one to choose depends on which question is being asked.

In practice the soundest arrangement is to add a behaviour layer to the existing assessment data rather than to replace it. How leadership simulations develop decision-making ability is treated separately.

The limits of the method

The most reassuring thing about a measurement tool is that it says what it does not measure. A decision simulation has four known limits, and all four are taken into account when the programme is designed.

1. It does not measure face-to-face interaction. The ability to run a meeting, to persuade and to carry a difficult conversation is invisible to this tool; that requires observation.
2. In a small pool the within-group percentile is noisy. In a pool of a few people a ranking carries no information. In the first period scores are read within their own comparison group; the norm strengthens only as the pool grows.
3. It does not decide a promotion on its own. It complements existing performance data, the manager's view and the interview; it does not replace them.
4. It is not a personality test. What it measures is not a trait but behaviour in a particular context. It assumes that behaviour can change when the context changes — which is, after all, the basis of a development plan.

Content validity depends on how the scenario is built: the cases are authored and reviewed together with managers from that sector. A generic case not adapted to the sector does not measure the thing it believes it measures.

Employee assessment data and data protection

A candidate's decision behaviour is personal data and must be processed as such. The framework that has to be settled before the programme starts gathers under four headings.

The data processed is name, corporate email address, session decisions and scores; no special-category personal data is processed. Access is separated: the participant sees only their own report, the programme manager sees individual and group reports. Notice is given before the session and it states plainly that the data is processed for the purpose of employee assessment. Roles are fixed in writing: the organisation is the data controller and the platform provider the data processor; the processing terms are contracted before the programme begins.

Setting this framework up front is not a compliance formality. Assessment data is among the most sensitive records an organisation keeps about an employee, and when the question of who sees what starts being argued after the fact, it is the programme itself that suffers.

Conclusion

In management trainee assessment the real question is not which tool is used but what the decision is grounded in. Past performance, the interview and seniority carry real information about a candidate; what they do not carry is what that candidate will do under pressure in a manager's seat. A decision simulation aims to fill that gap, and it can do so only under three conditions: the scoring key must be fixed in advance, every candidate must be measured under the same conditions, and the tool's limits must be stated openly. When those three hold, what emerges is not a verdict but an observation that can be discussed at the promotion table. The mistake has then been paid for in a simulation rather than in a career.