Most candidates hunt for an "exact" Sustainable Futures Lab (SFL) clone and then treat the percentage it prints as a McKinsey forecast. That is the wrong purchase. SFL sits inside Solve, McKinsey's gamified assessment. The firm does not publish a module walkthrough, a scoring matrix, or a pass mark. What candidate reports describe is a judgment task: rank initiatives against stated criteria, respect constraints, then defend the order when someone objects.
The thesis: a useful SFL simulator must force criteria-first ranking under time and give you a way to score the process, not a fake pass/fail. Format lore belongs in the sibling SFL guide. This piece is about how to practice with a simulator, how to grade your own runs, what transfers to the live window, and how to use CoachNed's Port Casilda sampler at /mckinsey-sfl without pretending it is McKinsey's item bank.
After reading, you should audit any SFL-style tool in two minutes, score a ranking with a four-line rubric, and run a five-day plan that builds the habit the live tutorial will reward.
What McKinsey publishes about Solve practice
McKinsey's public line on Solve is thin on purpose. The official Solve page frames a gamified assessment of problem-solving ability, says there is no need to prepare, and states that results are considered with the rest of the application. The interviewing overview places Solve as a step for most consulting roles and repeats the curiosity-and-creativity framing.
The FAQ PDF (as of August 2025) adds the operational rules that matter for practice design: length is in your invitation email, each task starts with an untimed tutorial, the assessment runs on a PC or Mac, and you should sit an uninterrupted slot. McKinsey's game-based innovation lab has said Solve reaches on the order of 300,000 early-career candidates a year. That scale explains the thin public documentation. It does not give you an SFL answer key.
Integrity is not folklore. On the Solve page, McKinsey prohibits applications, websites, AI tools, pre-prepared notes, recordings, and screenshots, and expects you to participate alone. Practice before the window. During the window, you work alone with what the tutorial allows.
What the firm does not publish: whether your invite includes SFL, per-module timers, scoring weights, or a pass threshold. Candidate reports that longer invitations add a judgment module next to Redrock and Sea Wolf are orientation, not primary fact. If a simulator claims "exact McKinsey SFL scoring," treat that as marketing.
What a good SFL simulator must reproduce
When I watch candidates flop on prioritization tasks, the break is almost never "they lacked sustainability knowledge." It is that they ranked by narrative urgency instead of the rule on the screen. A simulator earns its keep only if it reproduces pressure on the moves below.
| Move | What you do | What flimsy practice skips |
|---|---|---|
| Criteria capture | Write the primary metric and tie-breaker before touching the order | Skimming the story and ranking on gut |
| Ratio work | Compute the stated ratio for every initiative | Ranking by largest numerator alone |
| Constraint check | Enforce dependencies and budget windows as hard rules | Treating prerequisites as soft preferences |
| Defense | Acknowledge an objection, restate evidence, propose a next step | Capitulating or stalling until "more data" |
If a tool only quizzes "which project sounds most important," it is a vibes quiz. If it only drills ratios with no clock and no objection, it is half a case math sheet. Use /drills/math or /drills/quick-math for arithmetic. They are not a substitute for the ranking loop.
The defense step is the one most free tools omit. Candidate reports of SFL-style modules often include a stakeholder pushback after the order is set. Even a short coach prompt that forces acknowledge-evidence-action trains the disposition a partner listens for later: stay with the decision rule under social pressure.
How CoachNed's sampler maps to those moves
CoachNed's SFL practice at /mckinsey-sfl is an original, independent ranking cut. It is not McKinsey's game, not a screenshot of Sustainable Futures Lab, and not a promise of any score. It is a short stand-in for the criteria-and-constraint feel: five flood initiatives for Port Casilda, an explicit year-one lives-per-dollar rule with a reversibility tie-break, an eight-minute clock, and a model order when you finish.
Year-one lives per dollar first, then reversibility. Cheap and reversible beats a wetland that takes a decade. The model order is early warning, pump upgrades, seawall patch, zoning buyouts, wetlands restore. No account is required to start. Everything on CoachNed is open for seven days with no card; then $120 for a recruiting season or $49 a month.
Honest gaps matter. The sampler is one ranking round. It does not run a multi-scene stakeholder tree or claim McKinsey's unpublished process model. Score is proximity to the model order (up to 100), not a firm cut score. Use it to rehearse criteria capture, ratio instinct, and timed commitment. Pair /mckinsey-red-rock and /mckinsey-sea-wolf for the other Solve skill families.
Worked example: score Port Casilda like an interviewer
Reproduce this on paper in under five minutes, then compare to the sampler's model order.
Five initiatives, rough year-one lives saved and cost (invented figures for practice, not McKinsey data):
| Initiative | Year-one lives | Cost ($m) | Lives per $m | Reversible? |
|---|---|---|---|---|
| Early warning | 120 | 0.4 | 300 | Yes |
| Pump upgrades | 180 | 1.2 | 150 | Mostly |
| Seawall patch | 90 | 0.9 | 100 | Partly |
| Zoning buyouts | 200 | 4.0 | 50 | Hard |
| Wetlands restore | 400 | 5.0 | 80 | No (decade) |
Primary rank by lives per $m: early warning (300), pumps (150), seawall (100), wetlands (80), buyouts (50). Then apply the tie-break and the budget window: wetlands save the most lives eventually but fail year-one and reversibility, so they drop below seawall and buyouts in the model used by the sampler. The committed model order becomes early warning, pumps, seawall, buyouts, wetlands.
A candidate who submits that order has the arithmetic. Necessary, not sufficient. I still ask: did they write "lives / cost" before dragging anything? Did they kill one initiative out loud? Did they refuse to promote wetlands because the story felt green?
Common miss patterns:
| Answer pattern | Likely break | Fix on the next rep |
|---|---|---|
| Leads with wetlands | Ranked total lives, not the year-one ratio | Circle the denominator before sorting |
| Leads with seawall because "infrastructure" | Narrative urgency beat the metric | Write the primary rule in one line above the list |
| Puts buyouts above seawall | Ignored lives per $m after computing it | Sort by the ratio column, then apply tie-breaks |
| Reorders after a vocal mayor objects with no new data | Social pressure replaced the process | Acknowledge, show the table, offer a validation step |
That fourth row is why a defense rep belongs even when the UI only accepts a drag order. Say the so-what aloud after every submit.
How to score yourself without a published pass mark
McKinsey will not give you an SFL threshold. Do not invent one. After each timed run, fill this four-line card before you look at any model order.
| Line | Grade 0 or 1 | Pass look |
|---|---|---|
| Primary metric and tie-breaker written before the first move | 0 / 1 | You can recite the exact ranking rule |
| Ratio or score computed for every option | 0 / 1 | No orphan "this feels urgent" picks |
| Constraints checked (dependency, window, budget) before submit | 0 / 1 | No dependent item above its blocker |
| Defense spoken in one sentence if challenged | 0 / 1 | Acknowledge, evidence, next step |
A run that hits 4/4 on process and misses one adjacent swap beats a lucky model match with 0/4 on process. I would rather hire the first candidate: they catch themselves when a partner invents a new constraint mid-case.
Track process score (0–4) and order fidelity across five days. Process should rise first. Order fidelity often lags a day until ratio discipline is boring.
Ned's rule. If you cannot name the primary metric without looking back at the prompt, you ranked the story, not the decision. Write the rule above the list every time. Memory of the narrative is what fails when the tutorial swaps the denominator.
What transfers from a simulator, and what does not
Transfers: capturing criteria before ranking; computing the stated ratio for every option; a five-second dependency scan; acknowledge-evidence-action under pushback; calm under a visible clock.
Does not transfer: pixel layout, initiative names, McKinsey's unpublished weights, a CoachNed percentage as a pass predictor, or sustainability jargon as a substitute for arithmetic. McKinsey's careers material stresses that Solve does not require prior business knowledge. Memorizing flood-policy talking points is preparation theater. Practicing with AI copilots or second screens and then bringing those crutches into the live window is an integrity violation on McKinsey's own terms, not an edge.
A five-day practice plan that respects the invite
Do not burn a week on folklore about which offices see which module. Burn it on the ranking loop. If your invitation is short and never mentions a third game, still keep one SFL-style session: the same prioritization habit shows up in cases.
| Day | Assignment | Done when |
|---|---|---|
| 1 | One untimed Port Casilda pass; stop the clock | You can explain the full order from the ratio table alone |
| 2 | Twenty minutes of ratio and ranking drills; kill one option out loud each time | You never promote the largest numerator without the denominator |
| 3 | One timed Port Casilda run; fill the four-line process card before the model order | Process score is at least 3/4 |
| 4 | Same run, then a written stakeholder defense to a skeptical mayor with no new data | Your sentence names the metric, one number, and a next step |
| 5 | One timed run plus ten minutes of /drills/structure or /drills/brainstorming | You can say what transfers: criteria before options |
If process is still below 3/4 on day 5, do not add a sixth simulator. Repeat days 3 and 4. More volume without the card just rehearses urgency bias.
Practice this today
Open /mckinsey-sfl, start the eight-minute Port Casilda ranking, and refuse to move any initiative until the primary metric and tie-breaker are on paper. When the model order appears, fill the four-line process card before you react to the score. Then write one defense sentence as if the mayor objected with no new numbers. That loop beats three untimed quizzes that never force a commitment.
CoachNed is independent and is not affiliated with McKinsey & Company.
Frequently asked questions
Is there a free McKinsey SFL simulator?
CoachNed's Port Casilda sampler at /mckinsey-sfl is free to start with no account. It is an independent SFL-style ranking drill, not McKinsey's assessment and not a clone of the live UI.
Can an SFL simulator predict whether I will pass Solve?
No. McKinsey does not publish an SFL pass score. Use a simulator to diagnose whether you apply the stated rule, respect constraints, and defend the order. Treat any percentage as a practice diagnostic, not a hiring forecast.
How should I prepare for SFL in one week?
Five short sessions beat a weekend of forum lore: untimed criteria practice, ratio drills, one timed ranking with a process card, one stakeholder defense, then a final timed run. Read your invitation for whether a third module appears at all.
Does SFL test sustainability knowledge?
McKinsey states that Solve does not require prior business knowledge. Treat sustainability framing as atmosphere. The skill under test is applying stated criteria and constraints under time, not recalling climate policy.
What if my invitation is only 65 minutes?
Candidate reports often associate shorter invitations with Redrock and Sea Wolf only. Follow your email and on-screen tutorial. The ranking habit still helps case interviews even if SFL never appears.
Sources
- Solve, McKinsey's assessment game — official description, integrity rules, and how Solve sits with the rest of the application. Checked 2026-09-24.
- Interviewing at McKinsey — places Solve in the consulting interview process. Checked 2026-09-24.
- From apprenticeship in space to selecting microbes — McKinsey on the game-based innovation lab and Solve volume. Checked 2026-09-24.
- Innovating recruiting through online gaming — origin of Solve with Imbellus and online rollout. Checked 2026-09-24.