The short answer. The Sustainable Futures Lab (SFL) is the judgment game reported as the third McKinsey Solve module, after Red Rock and Sea Wolf, with no calculator and no data exhibit. In 85-minute invitations it is the standard 20-minute version: you rank actions for a sustainability project, then make about 12 linked decisions. In 95-minute invitations candidates report a newer 30-minute version, first reported on September 11, 2026, in which you lead four researchers over three simulated days. McKinsey has published neither format nor a scoring rubric, and 65-minute invitations do not include the SFL.
| Fact | Detail | Status |
|---|---|---|
| Standard SFL | About 20 minutes in 85-minute invitations; one ranking, then about 12 decisions | Candidate-reported |
| New SFL | About 30 minutes in 95-minute invitations; four researchers, three simulated days | Candidate-reported since September 11, 2026; a pilot |
| 65-minute invitations | No SFL | Candidate-reported |
| Format and rubric | Not published | Official absent from McKinsey's Solve page and August 2025 FAQ |
Checked October 2 to 3, 2026. McKinsey has not published either SFL format. Everything below is labelled as official, reported or reconstructed. Some searches use “Sustainable Future Lab”; this guide uses Sustainable Futures Lab, or SFL.
What does the Sustainable Futures Lab test?
Prep sites say the Sustainable Futures Lab tests five traits: prioritization, decision-making under uncertainty, interpreting messy information, balancing trade-offs, and team and stakeholder effectiveness. McKinsey has published no SFL rubric, so that list is a reconstruction, not an official scoring model.
What McKinsey states: nothing SFL-specific. Its Solve page and August 2025 FAQ do not name the game, its traits or a scoring model. The guide to what McKinsey Solve measures separates what is official from what is reported, and the Solve FAQ guide collects the rest.
What the lists share: several write-ups list the same five traits in the same order. Identical lists suggest a shared source rather than independent confirmations, so read them as a working map. The right-hand column shows what each trait looks like in this guide's own water scenario.
Prep-site reconstructionNot published by McKinsey
- Prioritization
- Putting the action that unlocks others first. In the scenario below, inspection A goes before repair B because B cannot start without it.
- Decision-making under uncertainty
- Acting on what is known without waiting for perfect information. When the repair part slips to 72 hours, temporary storage D moves up instead of the plan stalling.
- Interpreting messy information
- Separating the fact that changes the decision from noise. A part delay outside the 48-hour window matters; a resident's tone does not change the dependency.
- Balancing trade-offs
- Seeing what each option buys and what it costs. Moving the residents' briefing earlier buys trust and costs two hours of the window.
- Team and stakeholder effectiveness
- Answering the actual objection without inventing certainty: agree an update time, do not promise a fix tonight.
Trait names as listed online, reviewed October 3, 2026. Examples are from this guide's original scenario, not McKinsey content.
For the 30-minute version, one write-up (updated after the redesign) lists a different, behaviour-level set: prioritizing questions when requests are scarce, delegating with a defensible reason, handling interpersonal pressure without swinging between extremes, tracking how each team member is doing across the days, and staying consistent as the situation develops. That is one site's reading of a format weeks old.
Consistency, and what is known about scoring. No public source has McKinsey's SFL scoring. Write-ups agree that the run is read as a whole, so a choice on a late question that contradicts your opening priorities without new evidence is the expensive error. That is the habit worth drilling: keep later decisions consistent with earlier ones.
One write-up also reports that the end-of-assessment survey asks how many hours you spent preparing. That is a single, unverified report; answer it honestly if you see it.
How is the SFL different from Red Rock and Sea Wolf?
Red Rock is data analysis with an in-game calculator and Sea Wolf is rule-based selection under numeric ranges, while the Sustainable Futures Lab is written judgment with no calculator, where several options can be defensible and later choices are checked against earlier ones (reports, October 2026).
On small screens, scroll the table horizontally to compare all columns.
| Red Rock | Sea Wolf | Sustainable Futures Lab | |
|---|---|---|---|
| Reported time | About 35 minutes | About 30 minutes | About 20 (standard) or 30 minutes (new) |
| In every invitation? | Yes, in 65, 85 and 95 minutes | Yes, in 65, 85 and 95 minutes | Only in 85 and 95 minutes |
| Calculator | Yes, in-game | No | No |
| What you work with | Charts, tables and a research brief | Microbe attributes against site ranges and trait rules | Written scenarios about a project and its people |
| What a good answer shows | Correct numbers from the right data | A valid combination that meets every rule | Defensible, consistent priorities |
| Right answer? | Yes, mostly checkable | Yes, constraints are definite | Several options can be defensible; consistency separates them |
The practical difference: Red Rock and Sea Wolf reward checking, the SFL rewards judgment you can explain. Reports differ on Sea Wolf's length, with a few saying 35 minutes and most accounts giving 30. The Red Rock and Sea Wolf guides cover those games in depth.
Which Sustainable Futures Lab version will you get?
Your McKinsey Solve invitation length tells you the SFL version: about 85 minutes means the standard 20-minute Sustainable Futures Lab, about 95 minutes means the newer 30-minute SFL, and about 65 minutes means no SFL, as reported. McKinsey's August 2025 FAQ says only that your assessment length is stated in the invitation and that each task starts with an untimed tutorial.
| Invitation | Games reported | SFL version | Evidence |
|---|---|---|---|
| About 65 minutes | Red Rock (35) and Sea Wolf (30) | None | Consistent reports through 2026 |
| About 85 minutes | Red Rock, Sea Wolf, SFL (20) | Standard: a ranking, then about 12 decisions | Consistent reports; still sent after late August 2026 |
| About 95 minutes | Red Rock, Sea Wolf, SFL (about 30) | New: lead four researchers over three days | 95-minute invitations first reported August 24, 2026; new SFL first described September 11; a pilot |
- Red Rock
- Sea Wolf
- SFL
- New SFL
The extra 10 minutes in a 95-minute invitation are the longer SFL, not setup buffer. Reported game times; tutorials are untimed and sit outside them.
- Both versions are live. In public accounts, a candidate in the Middle East asked about a 95-minute invitation on August 24, 2026, another sat an 85-minute one on August 25, and on September 11 a candidate described a new 30-minute SFL with a “much more complex interface”.
- Treat the 95-minute version as a pilot. It is weeks old, no region pattern has been established and McKinsey has announced nothing. It may become standard, change or disappear.
- Some advice is already wrong. Several write-ups first said the extra 10 minutes were buffer time, and one guide described the 13-question SFL as replaced by the team version; both claims were contradicted by later accounts of 85-minute invitations.
With a 65-minute invitation you will not sit SFL: use the Red Rock and Sea Wolf guides instead. With 85 minutes, practise the standard version. With 95, read the new-version section, then practise the standard one too, because the judgment it trains is shared.
What is the difference between the standard and the new 30-minute SFL?
The standard Sustainable Futures Lab is a 20-minute ranking task followed by about 12 linked decisions on one project; the new version is a 30-minute, three-day team-management game in which you lead four researchers across four workstations. Both are reported, and the new one, first described on September 11, 2026, is best treated as a pilot.
| Standard SFL | New SFL (reported pilot) | |
|---|---|---|
| Invitation | About 85 minutes | About 95 minutes |
| Game time | About 20 minutes | About 30 minutes |
| Structure | One ranking task, then about 12 linked scenario decisions | Onboarding, then Day 1, Day 2 and Day 3 |
| Your role | Team member on one sustainability project | Leader of four researchers across four workstations |
| What you do | Rank actions, then pick one course of action per decision | Request information, assign people, resolve dilemmas, rate the team |
| Evidence | Many candidate accounts; start dates in write-ups begin in mid-2025 | One first-hand account plus reconstructions |
Neither version has a calculator or data exhibits. The screens differ more than the skills: both reward setting the objective first, spending scarce attention on facts that change the decision, updating when new evidence lands, and handling people without dropping the plan.
How does the standard 20-minute Sustainable Futures Lab work?
Published write-ups agree on the shape of the standard Sustainable Futures Lab: a short project brief, one ranking question, then about a dozen scenario questions that follow the same project from kickoff to recommendation. The SFL ranking task guide works that opening ranking step by step. Every option is a full sentence describing a course of action, and later questions carry earlier choices forward, so the run reads as one story.
How does the new 30-minute Sustainable Futures Lab work?
In the new 30-minute Sustainable Futures Lab, reported in 95-minute invitations, you lead four researchers across four workstations over three simulated days, and each day repeats four stages: explore, assign, support and reflect. The detail comes from write-ups and candidate accounts, and McKinsey has confirmed none of it. The new 30-minute Sustainable Futures Lab guide covers each day, each stage and how to prepare for it.
Practice coverage: Road to Offer offers both SFL Classic and SFL Island, with one complete run of each free. Classic includes all 13 linked decisions, one ranking and twelve situational decisions, without a countdown. Island includes onboarding and all three project days, with Explore, Assign, Support and Reflect each day, four researchers and four workstations, on a 30-minute Simulation clock. Start without an account, then create a free account to finish and keep your score and worked debrief. No payment is needed for either first run or its worked review.
The free review shows where points went and worked feedback on each submitted decision, with decision and dependency diagrams in Classic, and evidence chains, placement maps and team feedback in Island. Learning mode provides a step-by-step path with coaching and feedback as you go in both SFL formats, with untimed Island practice. It is paid access included with the SFL pack, the Solve pack or qualifying Annual access.
What are the most common Sustainable Futures Lab mistakes?
The six most common Sustainable Futures Lab mistakes are skimming the brief, answering each question in isolation, stalling for certainty, escalating what you could decide, ignoring the stakeholder and burning the clock on the opening ranking. They come from mistake lists that circulate online, merged where they overlap. They describe the standard version only. Each example uses the water scenario later in this guide, so you can test the fix yourself.
Skimming the brief
- Looks like
- Ranking repair B first because it sounds decisive, missing that B cannot start until inspection A finishes.
- Do instead
- Before ranking anything, write the hard limits in one line: budget, hours, crew, dependencies.
Answering each question in isolation
- Looks like
- Promising residents a fix tonight to win one question, then planning around a 72-hour part delay in the next.
- Do instead
- Keep a one-line rule, such as "verify first, never promise what is unverified", and check each choice against it.
Stalling for certainty
- Looks like
- Pausing all work until the supplier confirms a new delivery date.
- Do instead
- Act on what is still feasible: inspection A and temporary storage D fit the 48-hour window regardless of the part.
Escalating what you could decide
- Looks like
- Sending the whole ranking up for approval when every near-term action is inside your mandate and budget.
- Do instead
- Decide the in-mandate actions yourself and escalate only what genuinely needs approval, here the permanent plant E.
Ignoring the stakeholder
- Looks like
- Keeping the order without discussion because technical work matters more than the residents' complaint.
- Do instead
- Acknowledge the gap, commit to an update time, and consider moving briefing C earlier without breaking a dependency.
Burning the clock on the opening ranking
- Looks like
- Spending five or six minutes perfecting the first ranking, then rushing the dozen decisions that follow.
- Do instead
- Budget about 2 minutes for the ranking and 90 seconds for each of the 12 decisions that follow (2 + 18 = 20 minutes). A defensible ranking now beats a perfect one late.
How should you approach any Sustainable Futures Lab decision?
Run every Sustainable Futures Lab decision, in either version, through four checks before you click: objective, constraints, people and new evidence. This is our editorial frame, not a McKinsey rubric.
- Objective: what outcome is the team responsible for now?
- Constraints: what limits action, such as time, budget, capacity, dependencies or approval?
- People: whose information or cooperation changes delivery?
- New evidence: which assumption changed, and which decisions should move because of it?
The exercise below puts the frame to work: rank five actions, answer a stakeholder objection, then revise when a fact changes. It is original fictional practice, not a leaked question or an official answer key.
The scenario: restore a town's reliable water supply
You coordinate a town response team. Its objective is to reduce water-service disruption over the next 48 hours without committing to an unverified permanent solution. An inspection is needed to confirm a suspected leak and a connection point for temporary storage. The repair part is initially believed to be available.
- Your immediate budget is 60 funding units.
- You have two crew members. For this exercise, actions are performed sequentially, one at a time; do not assume parallel crews.
- Actions B and D cannot begin until inspection A finishes.
- A replacement part is required for B. Permission for temporary storage D is already granted.
- The permanent plant cannot be completed or funded within this response window.
Rank all five actions by priority. This is a priority list, not a promise that every item will be executed now. Put infeasible long-term work after the immediate response.
Scroll horizontally on a small screen to compare action costs, duration and conditions.
| ID and action | Cost (units) | Hours and crew | Condition |
|---|---|---|---|
| A: Verify the baseline with a site inspection | 4 | 4 h, 1 crew | No prerequisite. Confirms the leak location and tank connection. |
| B: Repair the confirmed leak | 24 | 16 h, 2 crew | Needs A and an available replacement part. |
| C: Brief residents and publish the next update time | 2 | 2 h, 1 crew | No prerequisite. State uncertainty rather than promising a fix. |
| D: Install a temporary storage pilot | 30 | 24 h, 2 crew | Needs A. Permission is already granted in this fictional scenario. |
| E: Build a permanent treatment plant | 120 | 480 h, 2 crew | Separate funding and approval. Not feasible in this response window. |
Write your order before revealing ours. For your top two actions, name the dependency or risk that determines their position. Which action would you defer even if it sounds most ambitious?
Reveal one defensible initial ranking
A → B → C → D → E is a defensible order, not the only possible answer.
- A first: verify the condition before choosing a repair or connecting a storage pilot. A unlocks B and D.
- B next: the part is initially available, and repairing the confirmed leak addresses the suspected source of disruption.
- C next: explain what is known, what remains uncertain and when residents will hear more. It should not promise a result that the team has not verified.
- D next: temporary storage adds resilience while the team monitors the response. It is feasible but does not fix the leak itself.
- E last: the permanent plant costs 120 units and takes 480 hours. It requires separate funding and approval, so it does not fit this immediate mandate.
The first four actions cost 4 + 24 + 2 + 30 = 60 units and take 4 + 16 + 2 + 24 = 46 sequential hours. No action needs more than two crew members. E is deferred, not squeezed into the remaining two hours.
Stakeholder challenge: improve communication without inventing certainty
A residents' representative says: “We have heard promises before. Why are we only hearing from you after the repair work?” Choose a response before revealing the discussion:
- Keep the order without discussion because technical work is more important.
- Promise that the water problem will be solved tonight to reassure the group.
- Acknowledge the information gap, agree a concrete update time, and consider moving C earlier while preserving the inspection dependency.
Reveal the response and an alternative defensible order
Option 3 addresses the concern without inventing certainty. For example: “You need a dependable update, not another unsupported promise. We will explain what the inspection will check and publish the next update time. We will report what it finds before confirming the repair plan.”
C → A → B → D → E can also be defensible when the immediate communication gap is causing harm. C has no prerequisite, costs two hours, and the first four actions still fit the 48-hour window. A still precedes B and D. Another reasonable order is A → C → B → D → E if the initial inspection result is important to the first briefing.
Option 1 ignores a relevant operational risk. Option 2 offers a deadline unsupported by the evidence. Moving C is not automatically right in every situation; explain why its benefit justifies delaying another action here.
New evidence: the repair part will arrive in 72 hours
Before work starts, the supplier corrects its stock report: the part required for B cannot arrive until 72 hours from now. The 48-hour response window and budget are unchanged. A and D remain feasible, and the storage connection still needs inspection.
Revise your order. What changed, what stayed the same, and which earlier assumption was invalidated? Do not invent a second crew or assume the supplier can deliver sooner.
Reveal the revised plan and explain the change
A → D → C → B → E is one defensible revision. A still provides the required evidence. D becomes the feasible near-term intervention because B cannot start within the window. C explains the temporary approach and the known part delay. B is planned after delivery; E remains a separate long-term proposal.
The executable first three actions cost 4 + 30 + 2 = 36 units and take 30 sequential hours. Reserve the remaining 24 units rather than pretending B has been completed. The 60-unit budget has not increased.
Moving C before D can again be reasonable if prompt communication is critical. What is not defensible is keeping B ahead as an executable immediate action while ignoring its unavailable part, or skipping A when D still depends on it.
Before and after: what moved, and why?
The new supplier fact changes one action's feasibility. It does not erase the inspection dependency, increase the budget or make the permanent plant fit the response window.
What one new fact changes in the plan
Initial information
Part believed available
- 1AVerify the baseline with a site inspection
- 2BRepair the confirmed leak
- 3CBrief residents and publish the next update time
- 4DInstall a temporary storage pilot
- 5EBuild a permanent treatment plant
Revised information
Near-term plan changes
- 1AVerify the baseline with a site inspection
- 2DInstall a temporary storage pilot
- 3CBrief residents and publish the next update time
- 4BRepair the confirmed leak · wait for delivery
- 5EBuild a permanent treatment plant
How do you check a Sustainable Futures Lab answer without an official score?
Check a Sustainable Futures Lab practice answer against four questions: objective, feasibility, stakeholders and updating. McKinsey publishes no SFL percentile or pass mark, so this is our editorial checklist:
- Objective: does the action reduce near-term disruption, or did a grander but infeasible project distract you?
- Feasibility: did you check budget, hours, crew capacity, prerequisites and approvals?
- Stakeholders: did you answer the actual objection without promising facts you cannot establish?
- Updating: can you point to the new fact that changed the plan, rather than switching randomly?
A different priority list can be strong when its assumptions are explicit and consistent with the brief. A polished explanation cannot repair a broken hard constraint. If you propose parallel work, an extra budget or an accelerated delivery, you have changed this exercise rather than solved it.
Connect the practice to your next step
Try writing a two-sentence rationale before comparing it with the explanations: “I prioritize X because Y. I would change that order if Z.” Review whether Z actually appears in the new evidence. This makes the exercise useful even without an automatic score.
Road to Offer covers four practice formats: Red Rock, Sea Wolf, SFL Classic and SFL Island, with one complete run of each free. Red Rock includes the full Study and six Cases on a 35-minute practice clock; Sea Wolf includes all three sites on a 30-minute Simulation clock. Learning mode is available with the matching game pack, the Solve pack or qualifying Annual access across all four formats, untimed in the clocked games. The separate Solve plan is free with a free account: enter your invitation length and test date for dated practice steps selected for that invitation, up to 28 days. These independent practice formats do not establish which modules or scenarios will appear in your real assessment.
For arithmetic rather than prioritization, try the Red Rock study. For rules with definite validity checks, try Sea Wolf combinations. Keep preparation separate from the actual assessment using the test-day checklist.
Sources and evidence boundaries
McKinsey's official Solve page and its August 2025 FAQ state that your length is in the invitation email and that tutorials are untimed. They do not describe SFL by name, either version, or a scoring rubric.
First-hand accounts of the 95-minute invitation date from August 24 to September 11, 2026. Reconstructions were reviewed October 2 and 3, 2026 for the trait list, mistake lists and game comparison. We treat these as reports, not authority, and none of them publishes a verified McKinsey scoring model.
