| 1 | Attendance, absence, and overtime control | Analyze where overtime and unplanned absence actually occur. Review schedules, coverage rules, workload peaks, and how supervisors respond when someone is out. Separate absence that comes from workload or work design from absence that is an individual pattern. Reset who may authorize overtime, and which causes must be fixed instead of covered with premium hours. | 30–60 days | Stabilizes daily coverage and makes true capacity visible. Teams stop using overtime as the automatic response to uneven workload, rework, absence, or late decisions. Errors and burnout that follow chronic overtime become easier to see and interrupt. | Moves payroll expense that is already being paid at a premium: overtime wages, paid absence, absence-coverage cost, last-minute contractor fill-in, and lower output per paid labor hour. The usual cash effect is immediate cost reduction, not new revenue. It does not, by itself, create gross profit. | In 30 days, take one unit with a persistent overtime pattern. Classify the last four weeks of overtime by cause: volume, absence, rework, late inputs, or approval delays. Change one rule — who can authorize overtime — and require a cause code so premium hours cannot be approved without a reason. | Overtime hours and overtime premium paid; unplanned absence rate; cost of coverage, including contractors used to fill gaps; output per paid labor hour; share of overtime caused by rework or late decisions rather than true volume. |
|---|
| 2 | Selection quality and hiring decisions | Analyze the work before writing the requisition. Define the outcomes the role must produce. Replace unstructured interviews with job-related questions, work samples, or job simulations scored on a shared rubric. Use a realistic preview of the work. Review the process for consistency and adverse impact before it becomes the standard. | 60–90 days to install; early job results at 90 days and 6 months | Fewer early failures in roles that drive output, service, or risk. Managers spend less time correcting a hire the process could have informed. The organization learns which evidence actually predicts performance in its roles, instead of relying on confidence in the interview. | Affects recruiting expense, replacement cost, wasted onboarding and training expense, the productivity gap while a poor fit remains in the seat, and the operational risk of an unqualified decision. This is primarily avoided future expense and risk reduction. It is not an immediate payroll cut, and it does not guarantee a better hire. | In 60 days, install a structured interview and a one-page scorecard for one high-volume role or one high-cost role. Every interviewer scores the same job-related factors before the debrief. Stop adding unscored 'culture' impressions after the fact. | 90-day and 6-month retention of new hires in the target role; time to reach the agreed proficiency standard; hiring-manager quality ratings collected on a rubric, not a vibe; cost per hire; early turnover within the first 90 days; offer-to-accept where the preview of the work changed acceptance. |
|---|
| 3 | Retention of critical employees | Identify the roles where a resignation creates outsized cost or risk. Review regretted losses, exit themes, and stay conversations with people still in those roles. Separate controllable causes — manager, workload, pay fairness relative to the work, growth in the role — from causes the organization will not change. Direct action at the controllable causes. Do not launch a blanket perk and call it retention. | 30–90 days for leading indicators; turnover usually follows later | Protects institutional knowledge and continuity in roles the operation cannot casually re-cover. Reduces the disruption that follows a regretted loss: open work, rushed hiring, and uneven service. Makes retention a managed portfolio of critical roles rather than a company-wide slogan. | Affects replacement cost, recruiting expense, onboarding and training expense, overtime paid to cover the vacant work, and productivity lost while the new person ramps. The cash effect is avoided future expense and preserved productivity. It is not immediate cash unless a departure was already about to create those costs. Retention improvement does not guarantee that any specific person will stay. | In 30 days, list the critical roles where a single resignation would hurt delivery, cash, or customers. Hold a structured stay conversation with each incumbent. Record the top three controllable reasons they would consider leaving. Assign one action to one owner before the month ends. | Voluntary turnover in named critical roles; regretted loss versus total loss; time-to-fill when a loss occurs; overtime and contractor cost during vacancies; recruiting and onboarding cost per replacement; incidents of knowledge loss that delayed work. Calculate avoided cost from your own baseline, not from a generic multiple of salary. |
|---|
| 4 | Manager effectiveness | Define the few managerial routines that actually move the work: priority setting, feedback on the work itself, removal of blockers, and fair handling of performance. Assess current managers against those routines, not against a personality model. Coach to a weekly cadence. Where the span, the tools, or the role definition makes the routine impossible, change the design of the manager job rather than sending the manager to another course. | 60–90 days for observable routines; team results after that | A manager's routines scale across a whole team: clarity of priorities, speed of decisions, quality of feedback, and whether people stay. Improving the routine improves the operating system of the team. It does not require every manager to become charismatic. | Shows up indirectly, which is why it must be tied to a team metric: productivity per paid labor hour, overtime in that team, regretted turnover, rework, and the cycle time of decisions the manager controls. The cash effect may be productivity, avoided expense, or gross profit, depending on what the team produces. There is no standard dollar return to claim in advance. | In 60 days, install one routine on one team: a weekly 30-minute priority huddle with a written list, one feedback standard for the work that matters, and one escalation rule for decisions that are stuck. Review whether the team used the routine, not whether they liked the workshop. | Share of weeks the routine actually happened; team output or cycle time; overtime hours on that team; regretted turnover on that team; employee-reported clarity of priorities; age of decisions waiting on the manager. |
|---|
| 5 | Sales and customer-service performance | Study what higher-performing employees actually do in the sales or service interaction, using observation or call review rather than opinion. Turn those behaviors into a short standard. Coach to the standard. Check that selection, onboarding, and incentives are not rewarding a conflicting behavior. Remove scripts or policies that force the wrong conversation. | 30–90 days where the behavior is observable | Changes conversion, repeat business, complaint handling, and the consistency of service across employees. Gives supervisors something coachable. Reduces the gap between the best and the typical interaction without assuming every employee will perform alike. | The relevant figure is additional gross profit, not additional sales alone, plus any reduction in refunds and service-recovery cost. Better employee behavior can lead to higher conversion or retention, then additional gross profit, then improved cash generation. That chain is a hypothesis. Revenue that is unprofitable, or cash that is collected later, is not a win. No conversion lift is promised. | In 30 days, review a sample of real interactions in one team. Write the three behaviors that distinguish better outcomes. Coach those behaviors twice a week for a month. Track conversion or complaint resolution for that team against its own prior month. | Conversion rate or resolution rate; gross profit per transaction, not revenue alone; repeat purchase or customer retention where the interaction should affect it; complaint and refund rate; revenue per employee; handle time only alongside quality, so speed is not mistaken for improvement. |
|---|
| 6 | Onboarding and time-to-productivity | Define what 'productive' means for the role at day 30, 60, and 90. Strip the first weeks of administrative delay and unstructured shadowing. Give the manager a checkpoint schedule and the new hire a short list of work to perform, not only content to read. Identify the errors that are predictable during ramp and put a job aid there. | 30–90 days for the next cohort | Shortens the period when the company is paying for a seat that cannot yet carry the work. Reduces early errors and early quits. Gives managers a standard so onboarding quality does not depend on who has time that week. | Affects onboarding and training expense, paid time before independent output, supervisor hours consumed by the new hire, early error and rework cost, and early replacement cost if the person leaves during ramp. The cash effect is productivity and avoided expense. A faster ramp is not additional revenue unless the role's output is sold and the extra output is real. | In 30 days, rewrite the first 30 days for one role: a day-by-day work plan, a manager checkpoint at the end of week 1 and week 4, and a written proficiency standard for day 30. Use it with the next person hired. Do not wait for a learning-system implementation. | Days to independent output against the written standard; 30-, 60-, and 90-day retention of the cohort; early error rate; supervisor hours per new hire; cost of the onboarding time itself. |
|---|
| 7 | Rework, errors, and service recovery | Build a short taxonomy of the errors and rework that actually consume time or trigger refunds. Separate skill gaps from unclear standards, bad inputs, and conflicting goals. Install a job aid or a feedback loop at the point of the error. Do not start with a general quality slogan. | 30–60 days when the error has a stable definition | Raises first-time-right work, shortens the path to a billable or usable output, and reduces the customer harm that becomes a complaint or a credit. Gives supervisors a concrete coaching target. | Affects rework labor, wasted materials, refunds and service-recovery cost, delayed billing, and customer loss where the error is visible to the buyer. Some of this is immediate cost reduction. Some is avoided future expense. Some protects gross profit. An error-rate change is not cash until the labor, the credit, or the delayed invoice is counted. | In 30 days, pick the one error that creates the most rework hours or credits. Define it so two reviewers agree. Sample a week of occurrences, name the most common cause, and put one job aid or checkpoint in place. Recount the same error for the next four weeks. | Rework hours; first-time-right rate for the named error; defect or error count; refund and credit dollars tied to that error; complaint volume tied to that error; days of delayed billing if the error holds up an invoice. |
|---|
| 8 | Goals, feedback, and incentives | Review whether goals describe a result the person can influence. Check line of sight from the goal to the operational metric. Look for incentives that pay for a behavior the operation is trying to stop — volume without quality, speed without accuracy, individual credit for shared work. Reset the goal and the feedback cadence before redesigning the whole pay plan. | 60–90 days for a goal reset; payout effects follow the pay cycle | Aligns daily effort with the result the business actually needs. Reduces gaming, neglected quality, and the cynicism that follows goals nobody believes. Gives managers a feedback target that is not a personality comment. | Affects productivity per paid labor hour, the mix of profitable versus unprofitable sales, the cost of incentives that do not buy the needed behavior, and rework created by a conflicting target. Misaligned pay is both a payroll expense and a gross-profit problem. Changing a plan has implementation cost — design time, communication, and possible payout change — and is not presumed to pay for itself. | In 60 days, take one role. Rewrite the goals so each one names a metric, a baseline, and a target the person can see. Remove or suspend one incentive rule that pays for a conflicting behavior. Review progress at day 30 and day 60 with the metric, not with a rating label. | Movement in the named operational metric; quality or error rate if the old goal ignored it; incentive payout versus the gross profit or cost the payout was supposed to influence; incidents of the unwanted behavior the old rule rewarded; productivity per paid labor hour for the role. |
|---|
| 9 | Decision speed and approval design | Map one process where work waits. Record who approves, how many steps exist, and which decisions are reversible. Set a threshold beneath which a named role decides without a meeting. Remove approvals that do not change the risk. Give stuck items an escalation age. | 30–60 days | Cuts waiting time in the operation. Clarifies authority so work is not re-explained at every layer. Reduces meetings whose only product is a delayed yes. Protects the approvals that actually manage risk by separating them from habit. | Affects delayed revenue, labor hours spent waiting or re-approving, overtime created by late decisions, and missed opportunities that are real only if you can point to a lost order, a held invoice, or idle capacity. Faster approval improves cash conversion when the delay was holding cash. It is not automatic cash. Some approvals prevent losses; removing those would be a false saving. | In 30 days, take one approval chain that employees name without being asked. Count the steps and the median age of items. Eliminate or delegate every step that has not changed an outcome in the last quarter, and set a maximum age before escalation. | Cycle time from request to decision; number of approval steps; median age of open items; staff hours spent in the approval loop; revenue or invoices delayed by the chain; overtime caused by decisions that arrived after the work window. |
|---|
| 10 | Workforce planning and critical-role coverage | Name the roles where a vacancy stops revenue, cash, safety, or delivery. Compare demand with the capacity you actually have, including absence and time lost to non-core work. Build coverage for those roles — cross-training, a successor, or a hiring pipeline with a real cycle time. Set a rule for when a contractor is a bridge and when a contractor is an unexamined habit. | 90 days for a usable plan; value appears as vacancies are avoided | Reduces surprise gaps in the roles that carry the business. Makes contractor use and overtime a choice with an end date. Connects hiring plans to work demand instead of to last year's headcount. | Affects contractor expense, overtime from uncovered roles, payroll committed earlier or later than demand, vacancy cost, and revenue per employee where an empty critical role blocks output. This is mostly avoided future expense and reduced operational risk. A plan on paper has no return until coverage, contractor spend, or vacancy days change. | In 90 days, name the critical roles, the current coverage if the incumbent is out for two weeks, and the contractor or overtime spend of the last quarter tied to those roles. For the single most exposed role, assign a backup or an active pipeline with a date. | Critical-role vacancy days; coverage ratio for named roles; contractor spend and overtime tied to those vacancies; time-to-fill versus the plan; revenue or output delayed by an empty critical role. |
|---|
| 11 | Work design: remove, redesign, reassign, or automate | Inventory the tasks in one process. Mark which tasks change the customer or financial result and which are handoffs, re-entry, or waiting. Decide what to remove, what to redesign, what to reassign to a different role, and what is actually ready to automate. Automation is last, after the work is simplified, so the organization does not automate a bad process. Specify the new role expectations and the manager routines that keep the new design in place. | 60–90 days for one process | Frees capacity inside the current payroll. Reduces handoffs that create delay and error. Makes a later technology investment more likely to be used, because the work it supports has been specified. Changes what a role is, not only how hard people try. | Affects cost per transaction, productivity per paid labor hour, contractor expense for work employees should not be doing, and payroll only if hours are actually reduced or redeployed onto revenue-producing work. Freed hours are not savings until the hours disappear from paid time or produce output that was previously foregone. This is productivity first. Cash follows only after that redeployment is real. | In 60 days, take one high-volume process. Have the people who do it list every handoff and re-key. Remove one handoff and one re-key. Measure hours per transaction before and after on the same volume definition. | Hours per transaction; share of time in non-value tasks; output per full-time employee on that process; handoff count; error rate after the redesign; contractor hours no longer used for the removed work. If automation is involved, adoption and the hours actually released — not the vendor's projected savings. |
|---|
| 12 | Capability and skill-gap closure | Start from a performance gap, not from a course catalog. Specify the skill the work requires and gather evidence of the gap — errors, observation, or a work sample. Choose practice on the job with manager reinforcement. Assess whether the behavior transferred. Stop programs that cannot name the on-the-job behavior they are buying. | 60–90 days for on-the-job transfer; longer if the capability system itself must be built | Closes the specific skill gap that is producing errors, slow ramp, or dependence on a few experts. Builds a repeatable way to know whether training changed work. Protects the organization from equating attendance at a course with capability. | Affects training expense that does not transfer, rework from skill gaps, contractor expense paid because an internal skill is missing, and productivity per paid labor hour once the skill is used. Training cost is an implementation cost. It is not an investment return until on-the-job behavior and the operational metric move. Many programs should be stopped rather than scaled. | In 60 days, pick one skill tied to a measured error or delay. Replace a general course with a short practice cycle: demonstration, coached attempts, and a manager check on the live work. Assess transfer at 30 days after the practice, on the job, not in the classroom. | On-the-job proficiency against a written standard; the error or cycle-time metric that justified the training; training cost per person who reached proficiency; contractor use for that skill; time to proficiency versus the prior method. |
|---|
| 13 | Change adoption and transformation execution | Before go-live, identify who must change which behaviors, what will make the old behavior easier, and whether managers can explain the purpose in operational language. During the change, measure adoption and proficiency, not announcement reach. After go-live, reinforce the new behavior and remove policies that still reward the old one. Use the same discipline for restructuring, mergers, expansion, and technology changes. | Adoption indicators by 90 days; business-case value often 6–12 months | Raises the odds that a change the business has already funded is actually used. Surfaces resistance as a design problem — skills, incentives, workload, or a real flaw in the change — rather than as a communication failure alone. Protects continuity of service while the new way is learned. | The financial stake is usually the business case already approved: productivity, gross profit, or cost reduction that will not arrive if people do not adopt the change. Failed adoption shows up as a productivity dip, delayed benefit, contractor overrun, customer disruption, regretted turnover during the change, and a write-off of the implementation. This is risk reduction and protection of a future return. It is not, by itself, new cash. Early months can cost more before they return anything. | In 90 days, for one change already underway, write the five behaviors that constitute adoption. Measure how many people in the target group perform them. Ask managers to remove one old incentive, report, or approval that keeps the old behavior alive. Review adoption fortnightly. | Adoption rate of the named behaviors; proficiency, not login counts; productivity versus the pre-change baseline; benefit realized versus the original business case; regretted turnover during the change; customer complaints or billing errors attributable to the transition; implementation cost against the approved budget. |
|---|
| 14 | Organization design and aligned culture systems | Examine whether structure, spans, and decision rights match the work the strategy requires. Identify duplicated effort across units. Treat culture as the behaviors the systems reinforce — who is selected, promoted, paid, and protected — not as a slogan or a survey score to be managed. Align at least one system with the behavior the business needs. Use survey results as a diagnosis, never as the result itself. | 6–18 months | Reduces structural drag: too many layers, unclear ownership, and units optimizing against each other. Makes long-cycle execution more consistent. Builds the conditions under which faster levers — selection, management, goals — keep working after the project team leaves. | This is long-term strategic value and risk reduction, not immediate cash. The financial categories it can eventually touch include cost per transaction, productivity per paid labor hour, and revenue per employee, through less duplication and faster cross-unit work. A reorganization has real implementation cost and a productivity dip. Do not book a culture-score increase as a financial return. Do not promise that a redesign will pay back inside a quarter. | There is no honest 30-day transformation. In 90 days, produce a decision-rights and duplication brief for one value stream: where two units do the same work, where a decision has no owner, and which promotion or pay rule contradicts the behavior leaders ask for. Choose one structural or system change. Schedule the 6- and 12-month reviews before the announcement. | Number of layers and unclear decision rights in the scoped stream; indicators of duplicated work; cross-unit cycle time; retention of critical talent through the redesign; productivity trend against the pre-change baseline; observed behaviors the system was supposed to reinforce, not a survey score alone. Review at 6 and 12 months. |
|---|