Fictitious composite
Every name, figure and finding on this page is invented to show the form of the work. Acme Industrial Group does not exist. No real client, person or result appears here.
01 · The decision and the plan
Fictitious composite
One division out of two, live by January.
Every forecast opens with the decision it serves, in the client's own terms. If this page is wrong, every page after it is wrong, so it is agreed with the accountable leader before any evidence is gathered.
- The client
- Acme Industrial Group, a fictitious European manufacturer of around 6,000 people, invented for this sample.
- The decision
- Combine the Components and Systems divisions into one operations-led division under a new operating model.
- What is financed
- A board-approved integration budget covering one-off integration costs, the consolidation of shared services, and a new divisional leadership team of seven roles.
- When it must be settled
- The leadership structure named by the end of September. The operating model live on 1 January, the start of the new financial year.
- What the plan demands
- Decisions that cross the old divisional boundary. One commercial motion where there were two. A single engineering roadmap owned by one person. Managers carrying teams from both former divisions through the first year.
- The accountable leader
- The incoming Chief Operating Officer, who owns the consequence if the integration slips. Roles are used throughout this sample; no individual is named or assessed.
What this page informs
A shared, written understanding of what the plan requires of the organisation, from whom, and by when. Every later rating is read against this page.
What it does not decide
Whether the merger is the right move. That decision was made before Atlas arrived, and the forecast does not reopen it. It asks only whether the organisation can deliver it.
Nothing on this page is recorded and nothing is sent. It is a reading copy, built for inspection.
02 · The evidence base
Fictitious composite
The evidence used, and the evidence left out.
No single source is treated as the truth. The judgement is formed by reading the layers together, against what the plan requires. Every input is listed here with its source and the consent behind it.
- Behavioural assessment
- Atlas's own psychometric assessment, taken by the leaders and managers expected to carry the plan across both former divisions. In this fictitious case, 38 of the 41 people invited completed it. Each person received their own results privately.
- Organisational survey
- Atlas's survey of both divisions, describing how the current situation is experienced: clarity of the plan, decision speed, and where work already crosses the old boundary.
- Plan and context documents
- The integration plan, the target operating model, and current role and decision-rights descriptions, supplied by the client. Every document is time-windowed; anything stale is excluded rather than quietly reused.
- Interviews
- Structured interviews with leaders from both divisions, conducted by Atlas consultants. Interviewees are told before they speak how their words will be used and at what level they will appear.
- Consent and visibility
- Every participant is told why Atlas is involved, what they will do, what they receive, and who can see what. Individual results belong to the individual. This report shows patterns at team and division level, never a person's private profile.
- Deliberately excluded
- Emails, calendars, performance ratings and personnel files are not used. They were not consented for this purpose, and the method does not need them.
What this page informs
Exactly what the forecast stands on, so the reader can weigh every finding against its source. Where the evidence is thin, the later pages say blind rather than guess.
What it does not decide
Nothing. Evidence never decides anything on its own. It informs the judgement on pages three and four, and people make the decisions.
Nothing on this page is recorded and nothing is sent. It is a reading copy, built for inspection.
03 · What can help and hinder execution
Fictitious composite
Four patterns, read against the plan.
This page rates the arrangement, never the people. Each pattern describes how the organisation is set up against what the plan demands, with the evidence for and against it. Where the signal is insufficient, the rating says blind.
Delivery discipline holds under pressure in both divisions
The behavioural evidence shows a strong, shared pattern of finishing what is committed, and the survey confirms both divisions recognise it in each other. The plan leans heavily on this in the first two quarters, and the organisation can carry it. No counter-evidence appeared in the approved sources.
Cross-boundary decisions concentrate in one role
In the target model, the three decisions that cross the old divisional boundary all escalate to the same role. The interview evidence shows both divisions already route disagreements upward rather than settling them sideways. Under integration pressure, that single desk becomes a queue. Workable, if decision rights are made explicit before January.
Two planning rhythms, and the merged division inherits both
Components plans annually; Systems replans quarterly. The integration plan assumes a single rhythm but does not say which, and no role in the target model owns the choice. Every dependency between the old divisions crosses this gap. Left unresolved, it slows the merged division from the first month.
Second-line management in Systems
Survey participation among Systems' second-line managers was too low to read a pattern, and the interview round did not reach them. Rather than extrapolate from their leaders, this area is marked blind. It is the first evidence gap the follow-up closes.
What this page informs
Where the leader should look first, and why. Patterns in how the organisation is arranged against the plan, each with its evidence and its uncertainty stated.
What it does not decide
Who is at fault, or who should stay or go. No individual is rated, ranked or named. The colours describe the work and the arrangement around people, never a person's worth.
Nothing on this page is recorded and nothing is sent. It is a reading copy, built for inspection.
04 · The recommendation
Fictitious composite
Settle the rhythm before naming the structure.
A recommendation is a proposed first move with its reasoning, not an instruction. It says what to address first, in what order, and what the evidence cannot support concluding. The client decides.
Address the planning rhythm first. It is the red pattern, it sits upstream of everything else, and it is cheap to settle now and expensive to settle in February. Choose one rhythm for the merged division, name the role that owns it, and write it into the operating model before the leadership structure is announced.
Second, make the three cross-boundary decision rights explicit, in writing, before January. The yellow pattern becomes red the moment integration pressure meets an implicit escalation path.
Third, sequence the announcements. Naming the leadership team while the rhythm and decision rights are still open invites both divisions to read the structure as the answer. Announce the way of working first, then the people who will carry it.
- Deliberately not concluded
- No view is offered on which individuals should hold the seven leadership roles. The evidence was neither gathered nor consented for individual selection, and the forecast does not stretch beyond what its evidence supports.
- Also not concluded
- No judgement is made about second-line management in Systems. That area is blind, and a blind area is a reason to gather evidence, never a licence to assume the worst or the best.
- Evidence against
- The case for moving slower: both divisions are mid-delivery on committed customer work, and every pattern above could be revisited after the year closes. The reason it is not recommended: the January date is fixed and public, the plan behind it is funded, so the risk moves with the plan either way.
What this page informs
The first move, its order, and the reasoning the leader can defend in front of a board: this evidence, read this way, points here.
What it does not decide
The decision itself. Atlas informs it; the accountable leader makes it, and remains free to weigh the evidence differently.
Nothing on this page is recorded and nothing is sent. It is a reading copy, built for inspection.
05 · The follow-up
Fictitious composite
This report is a baseline, not a verdict.
A judgement that is never checked teaches nothing. The final page fixes the review date, records what must be watched, and defines how the next reading will use this one.
- The review date
- Ninety days after the operating model goes live. Set now, in writing, while the judgement is fresh and before anyone knows how it turned out.
- What the next reading does
- It asks the same questions against the same baseline. Did the red pattern move once the rhythm was settled? Did the yellow one hold? What does the previously blind area show now that the evidence exists? Each original rating is compared with what actually happened, and the comparison is recorded.
- What must not worsen
- The human outcome is stated up front: trust between the two former divisions, and the participation rate that made the evidence readable. The outcome is weighed together with the other pages, so hitting the date is not allowed to hide what happened to trust and participation.
- If the plan changes
- A materially changed plan does not get a patched forecast. It gets a new judgement against the new plan, with this baseline preserved so the two can be compared honestly.
- What accumulates
- Every forecast and its later outcome joins a recorded audit trail. Atlas makes no accuracy claim until enough reviewed outcomes exist to test one.
What this page informs
Whether the intervention worked, on a date agreed before the outcome was known, and what the next judgement inherits from this one.
What it does not decide
Anything about future plans. The judgement covers this plan, this arrangement and this horizon. When those change, the work is redone, not extended.
Nothing on this page is recorded and nothing is sent. It is a reading copy, built for inspection.