Understanding change and making its case Review questions Integrated summary: Understanding change and making its case Change management begins with a testable problem, not a tool selected in advance. Calculate a return rate with a defined numerator, denominator and period, then inspect its distribution across teams and tasks. Map changes in behavior, decisions and data, including dependencies and exclusions: an interface edit may be limited while moving approval authority has broader consequences. Compare expected benefits, effort, uncertainty and alternatives; distinguish temporary containment from a permanent remedy. A future state describes proficient use and measurable outcomes, not merely a system launch. Track adoption and results alongside a service-time or quality guardrail, and set a review point with expansion and stop conditions. If the denominator changes or evidence contradicts the assumed cause, revisit the diagnosis before scaling the intervention.
Review questions Test your understanding with these questions. Marking and saved results are available with a free account.
Question 1 Eighteen of 90 cases closed for the first time in June were returned. A May report says 12 of 80 cases received were returned. What is the soundest conclusion before claiming that rework rose? It rose from 15% to 20%, so the increase can be reported immediately. June's rate is 18% because 90 counts closures rather than receipts. June's rate is 20%, but the denominator definition must match May's before comparing the rates. Neither rate can be calculated until the causes of the returns are known. Question 2 After a new intake tool launched, returns increased in one team. Seven of ten returned cases lack a request description. Which step can distinguish unclear handoff instructions from a data-transfer defect? Buy a replacement tool because the timing proves the current tool is defective. Observe real handoffs and compare the field before and after transfer with the current instructions. Train every team on the new tool without checking whether non-returned cases also lack the description. Measure general satisfaction alone; a lower score identifies which hypothesis caused the missing fields. Question 3 A fictional service log labels 30 of 80 cases as returned, but 12 are duplicate scans without a new request. What should the diagnosis correct first? Treat 30/80 as the final return rate Separate duplicates and recount unique cases Change every team's request screen immediately Question 4 A team reports slow closure, but half its open cases await an external approval unchanged by the proposed project. Which problem boundary keeps the change decision accurate? Attribute all waiting time to the intake screen Exclude pending cases from all service monitoring Separate approval wait from internal work time while tracking both Question 5 After a new route launches, 14 of 20 late requests in a report entered before launch. Which analysis avoids blaming the new route for all delay? Separate prelaunch cases and compare new cases under one definition Use all twenty late cases as launch-failure evidence Remove old cases from service monitoring entirely Question 6 Complaints about “waiting” rise at a walk-in office, yet observation shows a fast queue and delays before attachments are complete. What initial diagnosis fits? Add reception counters immediately Inspect attachment guidance and the previsit completion check Rename the waiting indicator only Question 7 A central dashboard reports average rework, but three service types have very different review steps. What should precede a common remedy? Pool the types to hide variation Treat the type with the most returns without its case denominator Calculate type-specific rates and link returns to their review steps Question 8 An association proposes a new intake form, while approval authority, eligibility rules and disbursement stay unchanged. What best describes the first initiative's scope? A complete service transformation because intake becomes digital. An intake and completeness improvement with a clear supervisor dependency. There is no organizational change because approval authority stays put. Disbursement must be added so the work deserves the label transformation. Question 9 In a trial of a new description field, a downstream reviewer relies on that description but never uses the intake screen. How should the plan treat the reviewer? Exclude them because only screen users are affected. Give them sole design authority because they read it later. Record the dependency and test whether the resulting description supports their decision. Automatically rate their impact high because they work downstream. Question 10 The team proposes reordering intake fields without changing eligibility rules or decision authority. How should scope be classified? A bounded interface change with a handoff check A transformation of organizational authority No need to involve reviewers Question 11 An intake error message changes, but the team must ensure rejected cases follow the same route. Which dependency belongs on the boundary map? Only the message colour Link the message to correction and rejected-case handoff Only executive training dates Question 12 A project map includes “digitisation” but omits a backlog of paper requests. How should the boundary be repaired? Assume paper cases migrate automatically Exclude all mention of backlog because it is old Assign backlog migration or closure and a cutover reconciliation test Question 13 The proposal moves approval of defined exceptions from a branch supervisor to a central officer while entry stays the same. What matters most on the map? Decision right, escalation route and response time Number of new interface colours Training the intake worker alone Question 14 The project pilots in two branches, but duplicate blocking uses a database shared with an excluded third branch. What is a safe boundary? Treat the third branch as unaffected because staff are not trained Include shared-rule impact and test third-branch cases before cutover Stop all branches permanently without testing Question 15 Forty requests a month are returned for a missing attachment; each repair takes 18 minutes. A new check adds one minute to each of 400 requests and prevents half the returns. What is the monthly operating-time effect before setup cost? Forty extra minutes; it saves 360 and uses 400. A 320-minute saving because all 40 returns disappear. A 360-minute saving because an automated check uses no human time. A 720-minute increase because that is current repair effort. Question 16 A pilot case assumes 70% of returns come from missing descriptions, but only three of ten reviewed cases show that omission. What is the best response before promising a 70% reduction? Keep the promise because a small sample does not disprove the assumption. Conclude that descriptions cause no returns. Change the return definition until the rate matches the forecast. Revise the benefit to an uncertain range and inspect causes further before expansion. Question 17 A proposal predicts fewer repeat calls, but no baseline calls-per-request count exists. What first condition makes the business case testable? Call the benefit “large” without records Collect baseline repeat calls per unique request before the trial Launch everywhere and establish baseline afterward Question 18 A manual check may avoid 40 hours of monthly rework but consumes 30 new hours, its effect on customer waiting is unknown. What is an honest case? A potential ten-hour net saving subject to waiting-time testing A certain forty-hour saving because checking is free Reject all trials because the estimate is uncertain Question 19 Management compares one project reducing case errors with another shortening email response, but the teams report benefits over different periods. What makes comparison defensible? Add the reported numbers and pick the larger Pick the proposal with the longer narrative Align time horizon and units and show nonconvertible benefits separately Question 20 A funding request assumes a mandatory field prevents every current error, although some occur after entry. What sensitivity test fits? Model lower prevention rates and inspect where remaining errors arise Treat all errors as preventable because the field is mandatory Remove later errors from organizational records Question 21 A small pilot improves median time but worsens the slow tail for complex cases. What belongs in the investment decision? Only the median because it represents everyone Median benefit plus a complex-case guardrail and support cost Stop measurement because the tail is small Question 22 Of 110 eligible cases, 90 pass through the new form and 14 of those need correction before handoff. What is correct-use rate among all eligible cases? 90÷110≈81.8%, because passage establishes proficiency. 76÷110≈69.1%, if correct means passage without correction. 14÷90≈15.6%, because that is the correct-case share. 76÷90≈84.4%, the requested rate among all eligible cases. Question 23 Returns fall after a new form, but median closure time rises by one day. Which dashboard reading supports a sound expansion decision? Final success because returns are the sole target. Final failure because any time increase cancels quality improvement. Possible improvement requiring an investigation of delay against the service guardrail before expansion. Remove delayed cases from the denominator so the rate reflects good adoption. Question 24 A success plan targets training completion only, though the goal is a changed request process. Which measure comes closer to the outcome? Share of eligible requests completed correctly without later repair Number of training slides used Share of staff who opened the invitation Question 25 Digital entry reaches 90%, but one fifth of cases revert to paper for completion. Which pair of measures is needed? Route entry and account count only Route use and correct end-to-end completion with fallback tracking Launch-message count and timing Question 26 Returns fall in a pilot service while eligible request volume halves. What dashboard check is essential? Compare return counts alone Delete the low-volume period Show return rate with numerator, denominator and case mix Question 27 A dashboard shows a good overall service average, but assisted cases exceed their local limit. How should expansion be gated? Gate expansion on overall performance and the harmed group's limit Ignore the group because it is small Stop measuring the overall average Question 28 A plan targets 80% correct use but does not define who counts as “eligible.” How should measurement be repaired? Use app openers as the automatic denominator Define eligible cases before sampling and log exclusions Change the target weekly to match available counts Create a free account to mark your answers and save your progress.