People Manager Operating Licence: One Month of DecisionsSend everything to HR and you are not managing. Settle everything yourself and neither are you.
A licence with a pass mark, taken over one continuous month rather than thirty unrelated scenarios. Five threads run through four weeks and every one of them comes back, so what you decided in week one is the position you are standing in by week four.
In brief
People Manager Operating Licence: One Month of Decisions is an online assessment in AssessAll's Applied Judgment Certification line, with 36 scored exercises taking about 45 minutes. It measures calling it fairly, keeping a conversation a conversation, standing between two people and what leaves your desk, plus 1 more.
| Format | Online, one sitting |
|---|---|
| Duration | About 45 minutes |
| Scored exercises | 36 |
| What it measures | Calling It Fairly, Keeping a Conversation a Conversation, Standing Between Two People, What Leaves Your Desk, Holding the Line Upward |
| Built for | HR and L&D teams using a single gate in front of an annual manager cohort, where a pass mark has to survive being questioned; People managers already in the chair who want an honest, private read on the calls they made this month rather than a competency profile; Heads of function deciding which of their managers can be handed a team with a live performance case in it |
| Scoring | Evidence-keyed |
| Report | Available immediately after submission |
| Currencies | INR and USD |
One month, five threads, and a pass mark that is not a rounding
Most manager assessments hand you a scatter of unconnected situations and average your answers into a competency profile. This one hands you a month. The two analysts whose mid-year notes you write in week one are the two you rate in week three and the two you speak to in week four. The person who goes quiet in week one is the person HR asks about in week two and the person who asks in week four whether any of this is a warning. The situations recur, the position is authored and identical for everybody, and the report says so plainly rather than letting you assume the month reacted to you.
The pass mark is not taken on a point estimate. Every score carries a standard error, and a licence granted on a number that a second sitting would move is a rounding dressed as a decision. So the gate here requires the bottom of the narrow error band to clear the cut, and a sitting whose point estimate clears it while the band does not is reported as provisional and told exactly why. The cut is published on the same axis the reader looks at, alongside both error spans, and the score itself is expressed against what a respondent choosing at authored per-option rates would earn, so zero means typical rather than nothing right.
Alongside the licence, four manager patterns are read one-tailed, because each of them only matters at one end: handing on what you could have settled, keeping it pleasant instead of clear, settling things privately instead of by a rule anybody could hear, and taking the instruction from above over the process owed to the person below. Each is reported as a bounded, conditional flag with the situations that produced it, never as a percentile or a type, and each is printed beside the strength it is the shadow of. A risk without its strength is a diagnosis. A strength without its risk is a horoscope.
What you walk away with
Granted, granted with a named condition, provisional on the cut line, or not yet — with the cut, both error spans and your own position drawn on one axis, and a sentence saying which of those four it is and why the band decided it rather than the point.
Four week columns, five thread lanes, one lettered card for every decision. You can see whether a thread held together across the month or came apart in week three, which is the reading a single percentage destroys and the reason the threads recur at all.
Bounded, conditional, and tied to the decisions that produced them, with the number of openings each pattern actually had and how many you took. Every one is printed in two columns: what the reflex gets you, and what it costs when the pressure is on.
Corrected against authored per-option rates rather than a uniform guess, with the raw figure, the chance figure, the points earned and the range of the declared rates all printed. The reliability behind every band is stated as an assumption, because no sitting has yet measured it.
A single boxed intention naming a situation you will recognise and a behaviour small enough to do, drawn from your own highest reading. Underneath it, the number your index would have to move at a later sitting before the movement is larger than the measurement error.
Inside your report
Illustrative sample — your report is generated from your own responses.
A pass mark taken on a point estimate is taken on a number a second sitting would move. Here the band has to clear the cut, and a sitting that sits on the line is told so.
The threads recur, so a lane shows whether a decision held across the month or came apart in week three. The month is authored and identical for everybody, and the report says so.
Read one-tailed, with the threshold published rather than tuned. No percentile, no type, and no risk printed without the strength it belongs to.
Built for
- HR and L&D teams using a single gate in front of an annual manager cohort, where a pass mark has to survive being questioned
- People managers already in the chair who want an honest, private read on the calls they made this month rather than a competency profile
- Heads of function deciding which of their managers can be handed a team with a live performance case in it
- Organisations that want the risks in a manager's judgement named and bounded rather than buried in a percentile nobody can interpret
Your team already knows how you decide. This is the first time you will.
Forty-five minutes, one sitting, the licence decision and the full report immediately — including the cut, both error spans and the rule behind the gate.
₹1,999 (incl. GST) · assessment and full report, nothing further to pay
Frequently asked questions
It measures the decisions a manager in a seat actually makes over a month: writing a fair record when two people did unequal amounts of work, holding a performance conversation that does not become a formal process by accident, allocating something scarce between two reports and saying how, routing a disclosure you are not qualified to resolve, knowing what must not be routed, and testing an ask from above. It is built for HR and L&D as a cohort gate, and for managers already in the chair.
No, and that was tested rather than assumed. Sending everything on is the strongest available answer on three of the thirty decisions and worth nothing at all on several others, so answering the whole month that way scores just below the typical respondent and never reaches the cut. Every fixed habit was pushed through the real scorer, including always waiting, always documenting, always keeping the peace and always selecting everything, and none of them passes. The value of every move is printed on your report.
About forty-five minutes in one sitting, thirty-six exercises, no writing and nothing to install. You receive the full report immediately: the licence decision with the cut and both error spans drawn on one axis, the month as a board of thirty lettered cards, the five thread standings, four named risks each beside its strength, the reliability table, one if-then action and a change threshold. There is nothing further to unlock.
As a gate and as a development conversation, which is why the licence has four outcomes rather than two. A provisional result says the score sits on the cut line and the error band decided it, so the honest next step is a second sitting rather than an argument. The named risks are written to be readable without a coach in the room, bounded and conditional, so a manager can hold their own downside without somebody trained to interpret it for them.
It is relevant anywhere, and it is not a legal check. No decision in the month requires the employment law, statutory leave rules, notice periods or disciplinary code of any country, and every keyed answer is defensible on process reasoning alone: evidence, consistency, proportionality, stated limits of confidentiality and whether the person can answer what is said about them. The report says this at normal size rather than in a footnote. It is priced in Indian rupees and in US dollars.
Certifications carry a pass mark and a credential; the rest of the catalogue profiles a capability without gating it. All of them score real situations against published evidence rather than testing recall, and each has a report built for its own subject.
Browse the catalogue →Methodology: Measures applied people-management judgment across one authored month of thirty decisions, with a pass mark. Construct statement: it measures the decisions a manager in a seat actually makes over a month - writing a fair record when two people did unequal amounts of work, holding a performance conversation that does not become a formal process by accident, allocating something scarce between two reports and saying how it was allocated, receiving a disclosure the manager is not qualified to resolve and routing it correctly, knowing what must NOT be routed, and standing between a team and an ask from above while accepting the ask from above that is reasonable. It does NOT measure readiness for a first management role, does not produce a competency battery, does not measure what a person's own team thinks of them, and is NOT a legal or employment-law compliance check. No item requires the employment law, statutory leave rules, notice periods or disciplinary code of any country, and every keyed answer is defensible on process reasoning alone: evidence, consistency, proportionality, the stated limits of confidentiality, and dignity. Declared response instruction: behavioural tendency throughout. Every situational item asks what the respondent is most likely to do, never what should be done, so knowledge framing and behavioural-tendency framing are never mixed within the scored bank. The six self-check statements are self-report and are excluded from the score entirely; they feed a calibration reading only. Item format: a fixed-state consequence cascade over one month (playbook B1 as corrected by field note 6.14). A static item bank cannot branch, so the position at every point in the month is authored, is identical for every respondent, and does not react to anything chosen earlier. Recurrence is carried by a thread tag rather than by an index, so it survives option permutation. A secondary balanced revision mechanic (playbook B11) runs on six of the thirty decisions: on three of them new information arrives that warrants a change of course, and on three of them what arrives is not new information and the keyed answer is to hold. The two halves are scored against each other, so always-revise and never-revise both score zero. Scoring design: a one-tailed derailer risk scale (playbook C12) operated as a licence gate. The graded composite is chance-corrected against authored per-option base rates rather than a uniform null, because a graded partial-credit percentage has a floor near half by construction (field note 6.7); the declared marginals are printed on the report with their range and mean, and they will be replaced by observed marginals once enough sittings exist. The licence is granted only when the lower bound of the 68 per cent band clears the cut, so a sitting on the line is reported as provisional rather than rounded up. Four risks are computed one-tailed as Risk = max(0, theta - tau) / (100 - tau) over the opportunities each risk actually had, reported as a flag with a described conditional behaviour and never as a number, a percentile or a type. Every risk is printed beside the strength it is the shadow of. Reliability is stated as an assumption rather than a measurement until live data exists: no competency carries a number, because six items at an assumed average inter-item correlation of .25 give an omega below the point at which a number is defensible, and the table that justifies the refusal is printed rather than the refusal being made in silence. Construct grounding draws on published work in: procedural and interactional justice, and voice and consistency as the determinants of whether a decision is experienced as fair; performance feedback specificity and the finding that feedback directed at the person rather than the task can depress performance; goal setting and review dates; the documented drift by which supportive conversations become formal processes without anybody deciding that they should; conditional confidentiality and the practice of stating the limits of a confidence before it is relied on; proportionality in workplace responses; upward influence tactics, in which consultation and rational persuasion outperform pressure and compliance; the distinction between routing with a named owner and escalation as an exit; attribution and credit allocation in team work; belief updating and the difference between openness to evidence and suggestibility; the derailer literature's finding that a risk is the shadow of a strength and must be reported bounded, conditional and paired; situational judgment test validity and behavioural-tendency framing; balanced keying and acquiescence control on self-report scales; and set-overlap partial credit, which penalises over-selection and under-selection alike. All items are original works. No trademarked instrument, scale name, derailer set or report layout is reproduced, and no affiliation with any source is claimed. AssessAll original design.
Last updated