Sustainable Futures Lab Ranking Task: A Worked Example

How do you approach the first Sustainable Futures Lab ranking question?

For the candidate-reported opening Sustainable Futures Lab ranking, use the brief to choose an order that serves its immediate objective, fits its constraints and respects prerequisites; McKinsey has not published a required method or answer key. Start by writing a short decision rule in your head: what must happen before the deadline, what makes another action possible, and what information would change the order? This is our preparation method, not a description of hidden scoring.

Read the whole brief before ordering anything. Underline the deliverable and time boundary, then distinguish an action that directly changes the outcome from one that only sounds important in general. A permanent solution may matter to the project but be useless for today's deadline. A communication step can be urgent without being first if it would announce an unconfirmed plan. Conversely, a people-facing action can move earlier when the brief says delay itself creates harm. There is no universal rule that “inspect first” or “communicate last.” The exact brief controls.

Candidates describe a ranking followed by later choices, so preserve a reason for each position, not just the letters. If the assessment presents a different objective or fresh facts, start over from those facts. For an overview of both reported SFL versions and a separate exercise, use the SFL practice method. The invitation-length guide owns the difference between the reported 85- and 95-minute formats, and the complete Solve guide places this game beside the other two.

Which criteria should decide the order?

For a candidate-reported SFL ranking, a useful practice grid asks whether each action protects the stated outcome, fits the deadline, unlocks another step and gives affected people a workable plan; these are our criteria, while McKinsey's scoring criteria are unpublished. Check each action against the same questions before you place it. That prevents a vivid but low-value option from winning because its description sounds decisive.

Use the following original fictional brief. A city operates 3 after-school centres and expects a heat alert in 36 hours. Only 2 coordinators are available. The objective is to reduce children's exposure during that alert. An existing cooled hall has approval for use, but staff still need to verify its capacity and the route. No building retrofit can be approved before the alert. You are ranking proposed actions, not making a claim about McKinsey's actual questions or public-health protocol.

Now read every action against the same four questions. The bars below are our marks on a scale from 0 to 3, where 3 means the action clearly does that job in this brief and 0 means it does not.

Read the picture by row and by column. B has the longest bar under the outcome, because the brief defines exposure during the imminent alert as the objective. A has the longest bar under “unlocks”, because the hall move and the shade both wait on it. C leads on a workable plan for families. E is empty on every question that this deadline asks. The bars make a judgment visible; they do not turn it into arithmetic, and no total decides the order. If the brief instead asked for a long-term building investment, E's row would fill. If the cooled hall were already verified, A's “unlocks” bar would shrink. State the condition; do not memorise a letter order.

What order works for the original heat-alert example?

For this original fictional heat-alert brief, a defensible order is A → B → C → D → E: verify capacity, open the cooled hall, send the confirmed plan, shade the waiting area, then start the roof proposal; this is a teaching answer, not a McKinsey key. The order serves the immediate objective and puts each action after what it needs. Another order could be defensible if the brief changes the exposure risk or says a route is already confirmed.

Add the first three jobs, then compare the total with the window before the alert.

So A, B and C take 11 elapsed hours and leave 25 hours of slack. The roof retrofit fails the same check on its own, even before its new approval:

One staffing trap hides in the middle of the order. B and D each need both coordinators, so they cannot run at the same time.

The close call is C versus D. D does real work, but it protects only the outdoor queue. C helps families actually reach the confirmed cooled space. After A and B, place C first. If the brief instead says people must wait outside for a long period before the hall opens, D may move earlier. This is the point of a ranking explanation: the adjacent choices depend on a stated fact, not a secret preference for one action label.

How do prerequisites change a ranking?

In the original example, prerequisites put A ahead of B and D, and a confirmed B ahead of C; this is a reasoning constraint in the fictional brief, not a McKinsey ranking rule. If you put B first before checking capacity, you are assuming the existing hall can take the exposed group. The brief deliberately withholds that confirmation. If you put C before B is settled, the notice may name a room or route that cannot be used.

A prerequisite is different from a vague preference. Ask whether doing the later action first is impossible, unsafe or merely less efficient. A cannot be skipped if it supplies the missing capacity and route facts. D can be prepared after A, but its benefit is smaller than B's for indoor exposure. You might rank a quick warning ahead of B if the brief says families are already travelling toward an unsafe site; that would be new information, and your reason should say so. The method is to trace the chain of consequences instead of applying a fixed “data before people” slogan.

The original SFL method guide uses a different project to show how a prerequisite works. For a wider habit of tracking dependencies across Solve games, see the Solve tips guide; the same discipline of checking every rule before you commit is what the Sea Wolf mistakes guide drills. Here the ranking itself is the unit of practice: explain every neighboring pair, especially B versus C and C versus D.

How should the ranking guide the next two decisions?

For the candidate-reported chain of SFL decisions, carry your opening objective and prerequisites into each later choice, but treat consistency as our coaching check because McKinsey has not published an SFL scoring criterion. The fictional heat-alert story now continues. These are original follow-up prompts, not recalled assessment content.

Follow-up decision: the hall loses power

After A verifies capacity, a power fault closes the cooled hall before B can open. A partner hall may be available, but its capacity and accessible route are unconfirmed. Choose between announcing the original hall anyway, moving the group there despite the fault, or checking the partner hall and route before issuing a revised notice. The third choice preserves the actual objective: reduce exposure at a usable place. It changes the operational plan because B's premise failed; it does not abandon the priority of direct protection.

The first choice looks “consistent” with C only if you ignore why C was third. C was a confirmed notice, not any announcement. The second choice copies the original ranking letter while discarding its safety condition. If the partner hall cannot be confirmed in time, the next defensible step is to find a feasible safe option or seek the authorized decision maker, not invent certainty. What matters in this practice exercise is the condition attached to the choice.

Follow-up decision: families ask for an immediate update

While the partner hall is being checked, families ask where to go. Choose between saying the move is confirmed, staying silent until every detail is known, or sending a short status update that says what is being checked and when the next update will arrive. The third choice respects C's purpose without claiming the new route is ready. Once staff confirm the place and route, send the actionable notice. The first option creates a false destination; the second leaves people without a useful time for the next answer.

These two choices test the same reasoning twice: use accurate evidence to protect people and communicate the plan at the right level of certainty. They are not a prediction of how branching or grading works in the real assessment. For the broader question of how SFL fits into an invitation, use the invitation guide rather than inferring the next screen from this exercise.

When should new evidence change your first order?

In this fictional ranking, change the order when a new fact removes a prerequisite, makes an action infeasible or changes the exposure risk; candidate reports describe linked SFL decisions, but McKinsey does not say that keeping the same order earns points. A power failure removes B as an immediate protective action. A verified partner hall could restore a version of B. A transport problem could make route verification more urgent than shade. Each revision needs a one-sentence explanation of which original assumption broke.

The revision costs time, so check it against the window again. The partner check takes 4 hours with 1 coordinator.

Do not treat “consistent” as “never change your mind.” That would lock you into the failed hall. Also do not treat every emotional or vivid statement as a new operating fact. If someone is upset but the hall remains safe and accessible, acknowledge the concern and keep the protective sequence. If someone provides evidence that the route excludes a group, the sequence changes because access to the protection has changed. You need not know a hidden weight to distinguish a changed fact from changed pressure.

What mistakes make a plausible ranking fail?

In this original exercise, a ranking fails its own brief when it mistakes a long-term project for an alert response, announces a plan before confirmation, or treats a dependency as optional; these are our diagnoses, not measured McKinsey failure patterns. Each of the three orders below sounds reasonable until you test it against the deadline and the dependencies.

Before settling the order, test every action with three short questions: what does it change before the deadline, what must already be true, and whom does it leave out? If two options both pass, compare their effect on the stated outcome. If the brief gives no way to choose, acknowledge uncertainty and identify the deciding fact. It is better to have a conditional reason than to invent a universal priority formula. The Solve difficulty guide covers pressure across games, while the prep plan helps allocate practice without turning this one ranking into your whole preparation. For the arithmetic-heavy game, the seven Red Rock mistakes follow the same pattern of a plausible answer that fails one check.

A common temptation is to reverse-engineer hidden scoring from practice screens. McKinsey has published no SFL pass mark, ranking weight or trait formula. For what McKinsey does say about the skills behind the games, read what Solve measures in each game; for what can actually be said about outcomes, use the scores and results guide and the note on how the Solve result is used. Your job here is to defend an order against the facts you have, then update it when the facts change.

How can you practise the ranking without memorising a script?

For the candidate-reported standard SFL ranking, practise explaining why each action comes before the next, then change one premise and rerank; this is independent preparation, and McKinsey says no preparation is required. Try the fictional heat brief without looking at the worked schedule. Say the objective aloud, name the constraint, choose an order, and explain the weakest adjacent pair. Then suppose the hall is closed and decide which part of the plan survives. That exercise trains a reason you can transfer to a different story.

The practice page offers a complete free SFL Classic run with all 13 linked decisions: one ranking and twelve situational decisions, with no countdown. SFL Island has its own complete free run: onboarding and all three project days, with Explore, Assign, Support and Reflect each day on a 30-minute Simulation clock. These are two of our four practice formats, alongside Red Rock and Sea Wolf, each with a complete free run. Start without an account; create a free account to finish and keep your score, points breakdown and worked feedback on each submitted decision, with decision and dependency diagrams in Classic and evidence chains, placement maps and team feedback in Island. No payment is required for these first runs or their worked reviews. This independent practice cannot predict your official score. For what candidates report about that version, read the 30-minute Sustainable Futures Lab guide; for invitation differences, follow the 85 versus 95-minute guide. If your invitation has no SFL at all, which game to practise first, Red Rock or Sea Wolf, is the better next read. For practical setup and rules, read the test-day checklist and the system requirements and tech issues guide before your sitting.

Do this preparation before the assessment. McKinsey's official rules say to take Solve alone without websites, applications, AI tools or prepared notes, except an approved written accommodation. During the live test, let its untimed tutorial and actual brief govern the task. If format reports change, the dated Solve update log separates reported changes from what McKinsey confirms, and the short Solve FAQ answers the common rule questions in one place.

Sources and evidence boundaries

• Official, checked October 11, 2026: McKinsey's Solve page says Solve is a gamified recruiting assessment, preparation is not needed, and the result is considered with the rest of the application. The August 2025 Solve FAQ says your invitation states the assessment length, each task begins with an untimed tutorial, and outside aids and prepared notes are prohibited except an approved written accommodation.
• Candidate-reported: The Sustainable Futures Lab name, its appearance in some 85-minute invitations, an opening ranking and about 12 linked decisions. McKinsey does not describe this game in the official sources above. The reported 95-minute team-management version is a different question covered in our invitation-length guide.
• Not published: McKinsey's SFL tasks, exact ranking interface, answer key, scoring weights, pass mark, treatment of later consistency and whether any one order is uniquely correct. The criteria grid and its judgment marks, the heat-alert dataset, the schedule, the order and the follow-ups on this page are original fictional teaching material; the figures come from the accompanying dataset and are recomputed by its test. They do not reproduce assessment content or a scoring model.