Scheduling - Practical Assessment
Objective. Test mastery of Scheduling for the certification gate: can the learner show a shortage, separate plan from optimise from publish, validate a schedule before trusting it, and avoid presenting a plan as a result or clean-against-config as compliant.
This is the certification assessment for Scheduling. The learner works one or two realistic scenarios and is scored on a checklist that mirrors the app's core guardrails. The pass bar is the Academy gate: practical score at or above 80, no critical error, and manager review. The critical errors that fail the assessment are the ones that would mislead a customer - presenting a plan as a result, trusting a stale-forecast draft, or calling a draft a commitment. The self-check items below are the assessment questions with model answers.
See it in the demo
How this assessment is scored
The learner must demonstrate, not recite. Scoring follows the Academy certification gate: practical score at or above 80, no critical error, and a manager review. A critical error fails the assessment regardless of the rest of the score, because each critical error is a statement that would mislead a customer.
The assessor watches for the same spine the whole app teaches: a schedule is a plan against a forecast, coverage shown is planned not actual, Generate/Optimise produces a draft bound by the rules you set, and a published schedule is a commitment still subject to real-world change. Run the scenarios live on the demo where possible; keep any capacity/cost framing separate from a cash claim.
Scenario A (CSM lean) - the fully-covered draft
Setup: A customer's scheduler shows you a W42 draft with a green coverage view and zero rule violations, and wants to publish it this afternoon. The draft was built earlier in the week.
The learner must:
- Not accept the green view at face value. Confirm which forecast version the draft is bound to - and recognise that a draft built earlier in the week may be on a stale forecast.
- Confirm the configured rules and skills are the customer's real policy, so 'zero violations' is meaningful.
- Confirm the time zone for the team/site.
- Confirm violations are cleared (not just low) and read coverage by interval, not the daily total, to check the peak.
- Separate plan from delivery: state that green coverage is the plan, not proof the day will run - adherence is RTA.
- Name who should own publication and the notice it should go out with.
A correct answer publishes only after the four validations pass, or withholds publication and escalates if the forecast integration or a required rule cannot be confirmed.
Scenario B (Sales lean) - the live demo and objection
Setup: In a live call, a prospect says: 'Just run your optimiser - it'll fix our coverage, right?' You have the demo open on the seeded W42 week.
The learner must:
- Reframe honestly: optimization reshapes the draft against the rules and cost you set; it does not change the forecast underneath and cannot invent a missing rule - a step, not magic.
- Show the five-beat sequence: a shortage window on the planner/coverage, the optimization before/after (shortage and overtime down, rules respected), then publication as the point a draft becomes a commitment.
- Keep the three steps distinct and say why (a draft is not live; optimization does not touch the forecast).
- State the limitation out loud: coverage is planned against the forecast, not actual delivery (RTA/Intraday), on seeded data with customer specifics confirmed with product.
- Close on a customer-owned pilot with success measured as coverage by interval, cleared violations, acknowledgment and publication notice.
A correct answer never promises the optimiser fixes everything and never presents the optimised draft as an achieved result.
The scored checklist
Award the practical score against these (weight the forecast-version and plan-versus-actual items most heavily):
- [ ] Confirmed the live forecast version before trusting coverage.
- [ ] Validated rules, skills and time zone against the customer's policy.
- [ ] Read coverage by interval, not the daily total; found or acknowledged the peak.
- [ ] Kept planning, optimization and publication as distinct steps and could say why.
- [ ] Framed optimization as reshaping a draft against set rules, not a fix for bad inputs.
- [ ] Treated the optimised draft as a draft, not a result, until published.
- [ ] Stated that coverage is planned, not actual delivery (RTA/Intraday for the live day).
- [ ] Named publication ownership and notice, and that a published schedule is still subject to real-world change.
- [ ] Separated any capacity/cost framing from a cash claim, and used confirm-with-product for unverified specifics.
- [ ] Proposed a customer-owned pilot with the right success measures.
Critical errors that fail the assessment
Any one of these fails the assessment, regardless of the rest of the score, because each would mislead a customer:
- Presenting planned coverage as actual delivery - e.g. 'coverage is green so the day went well.'
- Trusting a stale-forecast draft - publishing or endorsing a clean-looking draft without confirming it is on the live forecast version.
- Treating a draft as a commitment - calling an unpublished (however optimised) draft the live schedule.
- Claiming optimization fixes everything - presenting the optimiser as magic that repairs a wrong forecast or missing rules.
- Reading 'zero violations' as compliant without confirming the configured rules are the customer's real policy.
- Asserting a compliance, security, integration or contractual specific that should be confirm-with-product.
- Conflating capacity value with cash recovered.
Interpretation check
The assessment passes the learner who demonstrates the spine under pressure: confirm the forecast version, keep plan/optimise/publish distinct, read coverage by interval, and never present a plan as a result or clean-against-config as compliant. The gate is score at or above 80, no critical error, manager review.
In the field
The Scenario A pattern is the real KnackGlobal-scale risk: at thousands of seats, a single stale-forecast draft published on a Friday is a multi-site SLA miss - the assessment fails anyone who would have published it without confirming the live forecast version.
Practical task
Complete both scenarios live where possible. For Scenario A, produce the validation record (forecast version, rules/skills, time zone, cleared violations, interval coverage read) and your publish/withhold decision with owner and notice. For Scenario B, deliver the five-beat response with the limitation stated out loud and the pilot next step. Submit for manager review against the scored checklist; any critical error is an automatic fail.
Self-check
ASSESSMENT Q1: A scheduler shows a green, zero-violation W42 draft built earlier in the week and wants to publish today. Walk through what you confirm before agreeing.
Confirm the forecast version the draft is bound to (a draft built earlier may be on a stale forecast - the top trap); confirm the configured rules and skills match the customer's real policy so 'zero violations' means something; confirm the time zone; confirm violations are cleared and read coverage by interval to check the peak. State that green coverage is the plan, not proof of delivery (RTA). Agree to publish only after those pass, with a named owner and adequate notice - otherwise withhold and escalate.
ASSESSMENT Q2: A prospect says 'just run your optimiser, it'll fix our coverage.' Respond as you would on the call.
Reframe: optimization reshapes the draft against the rules and cost you set, does not change the forecast underneath, and leaves a draft - a step, not magic. Then show a shortage window, the optimization before/after with rules respected, and publication as the point a draft becomes a commitment. State the limitation: coverage is planned, not actual delivery. Close on a customer-owned pilot measured by coverage-by-interval, cleared violations, acknowledgment and notice. Never promise the optimiser fixes everything or present the optimised draft as a result.
ASSESSMENT Q3: Name three critical errors that fail this assessment and why each misleads a customer.
Presenting planned coverage as actual delivery (it is a plan, not a result); trusting a stale-forecast draft (a clean view on an old forecast fails current demand); treating an unpublished draft as the live commitment (a draft carries no commitment until published). Each would have the customer act on something that is not true.
ASSESSMENT Q4: A draft shows zero rule violations. Is the schedule compliant? Justify.
Not necessarily. Zero violations only means zero against the rules configured. If those rules are not the customer's full real policy, it is clean against the wrong policy. Validate the configured labour rules and time zone before making any compliance statement - and treat specific compliance claims as confirm-with-product.
ASSESSMENT Q5: Where does Scheduling's responsibility end, and what owns the rest?
Scheduling produces a plan: a feasible, rule-respecting, published and acknowledged schedule. It does not guarantee the day runs to plan. Actual activity and output against the schedule are Real-Time Adherence; re-planning the live day when reality diverges is Intraday Management. A published schedule is a commitment still subject to real-world change.
Verified vs demo. Scenarios run on seeded demo data; exact controls, labels and figures are SME-validation-pending. Any compliance, integration, security or contractual specific raised in the scenarios is confirm-with-product and asserting it is itself a critical error.