Performance measurement systems in the health and care sector: Are targets and monitoring additional demands or resources for employees?

1. 1Define the construct before you measure it. Supportive management is the employee's perception that managers value and care about them. It is not what managers intend or what policy says. Write down which employee groups you will ask and what 'valued and cared about' means in your setting. Then treat their answer as the data.
2. 2Map the causal chain on one page: management practice → employee perception of work conditions → wellbeing and attitude. The source's chain runs from targets and monitoring, through job control, supportive management and job demands, to job satisfaction, anxiety and organisational commitment. Place supportive management as the middle link, a perception you can read and a lever you can work on. Do not treat it as an outcome.
3. 3Inventory your performance practices and separate targets from monitoring. Targets are the objectives and milestones you set. Monitoring is tracking progress and collecting performance data. Count the targets per team and note whether the emphasis is productivity, financial performance or employee performance. The source finds these practices relate to different perceptions, so lumping them together hides the signal.
4. 4Cut target volume and productivity emphasis where they are heavy. In the study, setting many targets and focusing on productivity correlate with higher perceived job demands, and job demands correlate negatively with wellbeing. Fewer, clearer targets protect the ground on which supportive management can register.
5. 5Run monitoring of financial performance and of employee performance as an explicit part of how you manage. In the study, both relate positively to perceptions of job control and of supportive management, which in turn relate positively to job satisfaction and organisational commitment and negatively to anxiety. Make the data visible, feed it back, and use it to help people work towards agreed goals. Do not use it to assign blame.
6. 6Close the feedback loop so monitoring reads as care. The source describes systems as designed to provide clarity, objectivity and feedback, and as a language for collaboration and consensus. Employees also expect recognition and support in return for their effort (the Social Exchange and reciprocity argument). After each monitoring cycle, tell people what the data showed, what it means for their work, and what support you will offer.
7. 7Measure perception directly, together with its neighbours. Ask about supportive management, job control and job demands, then about job satisfaction, anxiety and commitment. Read the pattern across them. If supportive management is high but job demands are also high, you have not removed the demand problem.
8. 8Treat clarity as a trade-off, not a rule. A case study of English child and adolescent mental health services found unclear targets cause confusion but also enable creative, proactive solutions. Where you leave targets loose, say so and say why, so ambiguity reads as trust and not as management not understanding the work.

The source is a study of the British health and care sector. It uses matched employee-workplace data from the 2011 Workplace Employment Relations Study (WERS) and structural equation models. Its core finding is that performance measurement systems are neither a resource nor an extra demand in themselves. The effect depends on which practice, and on what employees perceive. Path one: setting many targets and focusing on productivity correlates with perceived job demands, which correlate negatively with wellbeing. Path two: monitoring financial performance and monitoring employee performance correlate positively with perceived job control and supportive management. Those perceptions correlate positively with job satisfaction and organisational commitment and negatively with anxiety. Read as an operating lesson, a manager facing a crowded target list should not conclude that measurement damages morale. The better reading is to trim the targets that load demand and keep transparent monitoring that employees experience as support. The data are correlational, so treat the pattern as a strong hypothesis to test with your own staff, not a guarantee.

• If employees report high job demands and you run many targets or a productivity focus, then cut or consolidate targets before adding any support programme, because the study ties those practices to demands and lower wellbeing.
• If supportive-management scores are low and monitoring is mainly punitive or absent, then add visible, feedback-driven monitoring of financial and employee performance, since these relate to higher perceived support and job control.
• If anxiety is high alongside low supportive management, then fix the management perception first, since supportive management correlates negatively with anxiety and positively with satisfaction and commitment.
• If targets are deliberately ambiguous, then state the intent and offer support for interpreting them. Ambiguity can enable autonomy and agility but also causes confusion.
• If you plan to rely on commitment as a management tool, then check that employees feel recognised and supported in return for effort, because they expect reciprocity.

• Assuming the system itself is the problem or the cure. The study finds it is neither a resource nor a demand by default. The tell is a debate about 'targets: good or bad?' with no data on employee perceptions.
• Equating monitoring with surveillance. Data collected but never fed back, or used only to assign blame, risks the blame culture Deming warned about. The tell is staff who describe monitoring as checking up and cannot name a decision it helped.
• Piling on targets and a productivity focus while expecting supportive-management scores to hold. These practices correlate with job demands, which correlate with lower wellbeing. The tell is rising workload complaints alongside lower satisfaction.
• Treating manager intention as employee perception. The construct is what employees perceive. The tell is managers rating themselves supportive while staff survey results show weak job control and weak support.
• Reading correlational evidence as proof of causation. The tell is promising specific wellbeing gains from a monitoring change without testing it against your own staff's perceptions before and after.

• ▸Team / employee group: name the group and the number of people you will ask.
• ▸Targets in force: list every target and tag each as productivity, financial, employee-performance or other.
• ▸Monitoring in force: list what is tracked, how often, and whether it is financial or employee-performance monitoring.
• ▸Feedback given: record what employees were told about the data last cycle and by whom.
• ▸Perceived supportive management: ask whether managers value and care about them and note the score or comments.
• ▸Perceived job control: ask how much discretion and decision latitude they feel over their work.
• ▸Perceived job demands: ask how much overload and time pressure they feel.
• ▸Outcomes: record job satisfaction, anxiety and commitment readings for the same group.
• ▸Action and review date: name one target to cut or clarify, one monitoring or feedback change, and when you re-ask.

1. 1Fix the definition before you measure anything: job demands are an employee's perception of the intensity of requirements at work. The two core parts are work overload and time pressure, meaning the effort needed and the difficulty of getting work done in the time available. Measure what staff report, not what the rota or workload data imply.
2. 2Measure demands separately from the two resources the Job Demands-Control model pairs with them: job control (decision latitude over how work is organised and delivered) and supportive management (support, acknowledgement and care). Record all three for the same team, because the model predicts demands lower wellbeing while control and support raise it.
3. 3Inventory the targets set for the team. Count every type, including the human-resources ones: training, workforce job satisfaction, absenteeism and labour turnover. The source builds its 'intensity of targets' measure as a standardised count against all target types available in its dataset.
4. 4Tag each target by type: financial, productivity, quality or HR. In the source, setting many targets and a focus on productivity are the two features that correlate positively with perceived job demands. Mark those first as the likely demand carriers.
5. 5Benchmark the count against the sector. The source compares a workplace's number of target types with the sector average (about 4.7) and finds the sector mean does not differ significantly from comparison groups. Use a sector or peer figure so you can tell whether your team sits above or near the norm.
6. 6Inventory the monitoring separately: financial performance monitoring and employee performance monitoring. In the source these are not linked to higher demands. They are positively associated with job control and supportive management. Do not merge monitoring and targets into one 'performance system' score.
7. 7Link demands to outcomes. Read perceived demands against job satisfaction and anxiety (wellbeing) and organisational commitment. In the source, higher demands go with lower wellbeing, so a demand rise with falling satisfaction or rising anxiety is the signature you track.
8. 8Act on the target portfolio, not on measurement itself. The source finds performance measurement systems as a whole are neither perceived as resources nor as extra demands. Prune or consolidate target types, especially productivity ones, and keep monitoring that supports control and support. Re-survey after the change.

The source study uses matched employee-workplace data from a nationally representative British survey of the health and care sector and estimates structural equation models. It does not treat 'the performance system' as one thing. Financial targets and financial monitoring are each reduced to a single component, which explains 72% of the variance in the four items behind it. Productivity targets, quality targets and employee-performance monitoring stay as binary indicators. A further measure counts how many target types a workplace uses, standardised against every target type in the dataset, including training, workforce job satisfaction, absenteeism and labour turnover. The result: many targets and a productivity focus correlate positively with perceived job demands, and demands correlate negatively with wellbeing. Monitoring financial and employee performance instead associates with more job control and supportive management, which go with higher job satisfaction and commitment and lower anxiety. For a manager the lesson is to audit target count and productivity emphasis as demand sources. Monitoring is not the demand source, and the data do not support treating the whole system as one.

• If a team carries many target types or a heavy productivity emphasis, then expect higher perceived demands and check satisfaction and anxiety first.
• If the question is whether to keep monitoring, then judge it separately from targets. Financial and employee-performance monitoring link to control and support, not to higher demands.
• If perceived demands are high, then do not treat control and support as a cure by assumption. Measure all three, because the model treats them as distinct perceptions with distinct effects on wellbeing.
• If your target count is at or below the sector average, then do not assume demands are low. The effect is a perception, so confirm with employee reports.
• If targets are unclear, then weigh both effects before clarifying. The case study of English child and adolescent mental health services found unclear targets cause confusion but also enable autonomy and creative solutions.

• Equating job demands with objective workload figures. The tell: the rota looks manageable but staff report overload and time pressure.
• Treating performance measurement as one lump that raises or lowers demands. The tell: you cut monitoring and demands do not move, because the source ties demands to target volume and productivity focus.
• Counting only financial or quality targets and missing HR targets such as training, absenteeism and turnover. The tell: your count falls below the sector norm yet staff still feel overloaded.
• Reading high demands as a motivation problem and answering with commitment drives. The tell: demands rise alongside falling job satisfaction and rising anxiety, the wellbeing pattern the model links to demands.
• Judging unclear targets as simply bad and tightening them at once. The tell: autonomy and creative problem-solving drop after clarification, because ambiguity was also supplying discretion.

• ▸Team / unit and survey date: name the group and when perceptions were collected.
• ▸Perceived overload and time pressure: summarise staff-reported effort needed and difficulty of finishing work in the time available.
• ▸Target types in use: list each type (financial, productivity, quality, training, workforce job satisfaction, absenteeism, labour turnover) and the total count.
• ▸Productivity emphasis: yes or no, with the target or message that signals it.
• ▸Sector or peer benchmark for target count: record the comparison figure and your team's position against it.
• ▸Monitoring in place: financial performance and/or employee performance; note separately from targets.
• ▸Perceived job control and supportive management: record both so demands are read against resources.
• ▸Outcome signals: job satisfaction, anxiety and organisational commitment, plus the direction of change since the last survey.
• ▸Action and re-check date: which targets to prune or consolidate, which monitoring to keep, and when to re-survey.

1. 1Fix the definition before you measure. Job satisfaction is the pleasure people derive from evaluating their work when it meets or exceeds their expectations (Warr, 1990). It is one of two wellbeing dimensions the source studies. The other is anxiety, a state of heightened unease and restlessness. Record both, because a practice can move them in different directions.
2. 2Separate satisfaction from its neighbours. Organisational commitment (attachment to the organisation, a sense of belonging, acceptance of its values) is a different outcome. Job control, supportive management and job demands are perceptions of work conditions, not satisfaction. Keep them as separate variables in any survey or model.
3. 3Treat satisfaction as the end of a causal chain, not a thing you manage directly. The chain runs from management practice, to employee perception of work conditions, to feelings and attitudes. Map your practice to three perceptions: job control (discretion over how work is organised and delivered), supportive management (support, acknowledgement and care), and job demands (effort, time pressure, task difficulty).
4. 4Break the practice into parts before you attribute effects. The source splits performance measurement into targets (identifying and setting objectives and milestones) and monitoring (tracking progress and recording data). It further separates financial, productivity and quality targets, financial monitoring, and monitoring of employee performance. List which of these your workplace actually uses.
5. 5Predict the sign for each part. In the source, many targets and a productivity focus correlate positively with perceived demands, and demands correlate negatively with wellbeing. Monitoring financial performance and monitoring employee performance correlate positively with control and supportive management, and those correlate positively with satisfaction and commitment and negatively with anxiety.
6. 6Measure the perception, not just the practice. The source's central point is that performance measurement systems are neither resources nor additional demands in themselves. Their effect depends on how employees perceive the work context. Ask staff about control, support and demand, and read satisfaction through those scores.
7. 7Check target intensity. The source counts how many target types a workplace uses, including human-resource targets (training, workforce job satisfaction, absenteeism, turnover), and standardises the count against the dataset. Count your own target types, and compare to a benchmark if you have one. High counts are the warning sign for demands.
8. 8Aggregate to the workplace level and watch it over time. Bryson et al. (2017), cited in the source, conjecture that one employee's wellbeing affects others', so the workplace average matters for productivity. Track the average satisfaction alongside performance, since satisfaction and performance are positively correlated across meta-analyses, experiments, and cross-sectional and longitudinal studies.

The source study (de Menezes and Escrig-Tena, 2023) tests this chain in the British health and care sector, using matched employee-workplace data from the 2011 WERS survey and structural equation models. Targets and monitoring are coded as binary indicators. The four financial targets collapse into one component explaining 72% of their variance, and the four financial monitoring items do the same. Productivity targets, quality targets and employee-performance monitoring stay as single binary variables. The results split by practice. A broad set of targets and a productivity focus raise perceived job demands, which lower wellbeing. Monitoring financial performance and monitoring employee performance are associated with higher perceived job control and supportive management. These in turn go with higher job satisfaction and organisational commitment and lower anxiety. So the same performance measurement system can push satisfaction down through demands or up through control and support. A manager reading this would first count target types, then ask staff whether monitoring leaves them discretion and support, and only then judge likely satisfaction effects.

• If a workplace sets many target types or leans heavily on productivity targets, then expect higher perceived demands and lower satisfaction, and check demand perceptions before adding any more targets.
• If you add or change monitoring (of financial or employee performance), then measure whether staff perceive more control and supportive management. The satisfaction and commitment gains in the source run through those two perceptions.
• If satisfaction falls after a measurement change, then look first at job demands (overload, time pressure) and at the number of targets. Do not assume monitoring itself caused it.
• If targets are unclear, then do not treat that as automatically harmful. Ojiako et al. (2022) found ambiguity is both a source of confusion and an enabler of autonomy and agility, so ask staff which it is.
• If you want to claim an effect on satisfaction, then test the mediating perceptions (control, support, demands). The source's hypotheses H1 to H3 place the effect there, not directly in the practice.

• Treating performance measurement as inherently a demand or inherently a resource. The tell: a blanket policy to cut or expand targets with no staff perception data. The source finds it is neither on its own.
• Merging all targets into one 'target pressure' measure. The tell: you cannot say whether financial, productivity or quality targets drive the change, when the source shows productivity focus and monitoring have opposite routes.
• Equating satisfaction with commitment or with absence of anxiety. The tell: a single 'engagement' score. Satisfaction, anxiety and commitment are separate outcomes with distinct definitions, and the source reports them separately.
• Ignoring target count. The tell: satisfaction complaints cluster in areas with the most metrics while each metric looks reasonable alone. The source measures intensity as the standardised number of target types.
• Reading satisfaction only at individual level. The tell: individual scores are reviewed but the workplace average is not. Bryson et al. (2017) link average wellbeing to overall productivity.

• ▸Workplace / team: name the unit being assessed.
• ▸Target types in use: tick financial, productivity, quality, and any HR targets (training, workforce job satisfaction, absenteeism, turnover); record the total count.
• ▸Monitoring in use: tick financial performance monitoring and employee performance monitoring.
• ▸Perceived job demands (staff rating): record overload, time pressure and task difficulty perception.
• ▸Perceived job control (staff rating): record how much decision latitude staff report over organising and delivering work.
• ▸Perceived supportive management (staff rating): record how much staff feel supported, acknowledged and cared for.
• ▸Job satisfaction (staff rating): record how far the job meets or exceeds expectations; keep separate from commitment.
• ▸Anxiety and commitment: record each separately so effects on satisfaction are not conflated.
• ▸Reading and action: note which pathway is dominant (demands down, or control and support up) and the one change you will make.

1. 1Define the target state precisely before measuring. Anxiety here is a heightened sense of unease and restlessness: high activation with negative affect. It is the opposite of contentment (low activation, positive affect). It is not the same thing as job dissatisfaction. Warr (1990) treats job satisfaction, the pleasure derived from work, as a separate dimension of wellbeing, so measure the two separately.
2. 2Choose the unit of analysis deliberately. Collect anxiety at employee level, then also report the workplace or team average. Bryson et al. (2017) conjecture that one employee's wellbeing affects others', so the average level is what bears on productivity.
3. 3Map the causal chain you are testing: management practice, then perceived work conditions, then wellbeing. The study follows the 'causal chain' (Peccei; Purcell and Kinnie; Boxall; Wood et al.), in which links from practices to outcomes are mediated by employee perceptions. Measure the three perceptions: job control (decision latitude over how work is organised and delivered), supportive management (support, acknowledgement and care), and job demands (effort needed, time pressure, task difficulty).
4. 4Inventory the performance measurement practices as the study does. Record binary indicators for each target type (financial, productivity, quality) and for each monitoring type (financial performance, employee performance). Then add a target-intensity count: the number of target types in use, including HR targets (training, workforce job satisfaction, absenteeism, labour turnover). Compare it with a sector benchmark. In the study the sector mean was about 4.7 types against about 5 across all workplaces, with no significant difference at the 5% level.
5. 5Trace each practice to anxiety through the perception that carries it. Expect two routes. Many targets and a productivity focus correlate positively with perceived job demands, which correlate negatively with wellbeing. Monitoring financial performance and monitoring employee performance correlate positively with job control and supportive management, which correlate negatively with anxiety.
6. 6Segment before acting. Compare anxiety across teams by (a) perceived control, (b) perceived support, (c) perceived demands, and (d) target count. Limited job control is linked to higher anxiety, dissatisfaction and disengagement (Fila et al., 2017), so low-control pockets are the first place to look.
7. 7Pick the lever that matches the diagnosed perception: restore discretion over how work is organised, strengthen visible managerial support, or reduce the number and intensity of targets that feed demands. Do not drop monitoring wholesale. The evidence treats performance measurement as neither a resource nor an extra demand on average.
8. 8Pair anxiety with its downstream indicators: absences and labour turnover, which anxiety has been linked to. Track both so the cost of unease shows up in operational terms.

The study draws on the 2011 Workplace Employment Relations Study (WERS), matched employee-workplace data for the UK health and care sector, and estimates structural equation models. Wellbeing is measured as job satisfaction and anxiety. Targets and monitoring are coded as binary practices. Financial targets and financial monitoring each reduce to a first principal component explaining 72% of the variance. A separate intensity measure counts how many target types a workplace uses. Reading the results as an anxiety manager, you find two opposite paths. A workplace with many targets and a productivity focus shows higher perceived job demands, and demands correlate with lower wellbeing. A workplace that monitors financial performance and employee performance shows higher perceived job control and supportive management, and these correlate with lower anxiety and higher satisfaction and commitment. The net conclusion is that performance measurement is neither a resource nor an added demand overall. The lever is therefore which practices you run and how many, not whether you measure. Treat the findings as associations from one cross-sectional survey, not proof that cutting targets lowers anxiety.

• If anxiety is high where perceived job control is low, then fix discretion over how work is organised and delivered before changing the targets themselves.
• If a team's target count is well above the sector norm (about 4.7 types in the study) and its perceived demands are high, then cut or consolidate target types, especially productivity targets, before adding any more.
• If monitoring of employee or financial performance coexists with strong perceived control and supportive management, then keep it. In the study these monitoring practices were associated with lower anxiety through those two perceptions.
• If satisfaction scores look healthy but anxiety is elevated, then treat them as separate problems and do not infer contentment from satisfaction. Contentment is low activation and positive affect, so check it directly.
• If one team's anxiety is rising, then check neighbouring teams too, because one employee's wellbeing may affect others'. Act at workplace level, not only on individuals.

• Equating anxiety with dissatisfaction. The tell: a single 'wellbeing' score, or a satisfaction survey used as the anxiety measure.
• Blaming targets and monitoring wholesale. The tell: removing all monitoring, even though monitoring was associated with more perceived control and support, and with less anxiety.
• Counting targets without measuring perceptions. The tell: the practice inventory exists but no job demands, control or support data sit beside it, so the mediating path is invisible.
• Omitting HR targets (training, satisfaction, absenteeism, turnover) from the intensity count. The tell: the target tally looks low while staff still report time pressure.
• Reading associations as proven causation. The tell: a claim that a specific target 'causes' anxiety, drawn from one cross-sectional dataset with no change over time tracked.

• ▸Unit and period: team or workplace, and survey date. Name the unit and the period being assessed.
• ▸Anxiety level: record the unit's anxiety score using your validated instrument (the passages do not specify items) and also the unit average.
• ▸Job satisfaction level: record separately so the two dimensions are not conflated.
• ▸Perceived job control: score employee perceptions of discretion over how work is organised and delivered.
• ▸Perceived supportive management: score perceptions of support, acknowledgement and care from managers.
• ▸Perceived job demands: score perceptions of effort, time pressure and task difficulty.
• ▸Target inventory: tick financial, productivity, quality and HR targets in use (training, satisfaction, absenteeism, turnover) and record the total count against the sector norm.
• ▸Monitoring inventory: tick monitoring of financial performance and of employee performance.
• ▸Downstream indicators: record absence and labour turnover figures for the unit.
• ▸Diagnosed path and action: state whether the main route is demands (target load) or low control/support, and name one lever with an owner and a re-measure date.

1. 1Fix the definition before you measure anything. Organisational commitment is affective attachment: a sense of belonging, acceptance or internalisation of organisational values, and willingness to put in greater effort for the organisation's good. Satisfaction, anxiety and mere compliance with targets are different constructs. Write the three components (belonging, internalised values, extra effort) as separate survey or interview probes.
2. 2Map the causal chain before you intervene. Management practice (targets, monitoring) shapes perceived work conditions (job control, supportive management, job demands), which shape outcomes (commitment, job satisfaction, anxiety). Treat commitment as the end of a chain, not something a practice produces directly. Your levers are the perceptions upstream.
3. 3Inventory your performance measurement practices in two groups. Group 1 is targets: how many you set and whether the focus is productivity. Group 2 is monitoring: of financial performance and of employee performance. The source treats targets and monitoring as the core of any performance measurement system, so audit these first.
4. 4Measure the three work-condition perceptions as employees define them. Job control is decision latitude over how work is organised and delivered. Job demands are the intensity of requirements: effort, time pressure, task difficulty. Supportive management is perceived support, acknowledgement and care from management. Ask about each separately; do not merge them into a general 'morale' score.
5. 5Read your own data against the source's pattern. Many targets plus a productivity focus correlate positively with perceived job demands, and demands correlate negatively with wellbeing. Monitoring of financial and employee performance correlates positively with job control and supportive management, and these correlate positively with commitment and satisfaction and negatively with anxiety. Check which pattern your workplace shows before changing anything.
6. 6Where demands are high, reduce target count and productivity pressure before adding engagement activity. Where monitoring exists, make it visible to staff as feedback and a shared language for collaboration, not only as oversight, and pair it with discretion over how work is done and with visible managerial support.
7. 7Close the reciprocity loop. Committed employees expect recognition and support in return for their effort. Pair any request for extra effort or proactive problem solving (as Quality Management wants) with explicit acknowledgement and support, and check commitment again afterwards.
8. 8Interpret results as associations. The source is a cross-sectional matched employee-workplace survey analysed with structural equation models, so it shows correlated pathways, not proven causes. Treat your changes as tests and compare before and after.

The source study (de Menezes and Escrig-Tena, 2023) used matched employee-workplace data from a nationally representative British survey of the health and care sector and estimated structural equation models. It started from a common criticism: that targets and monitoring burden staff and erode attachment. The data did not support that blanket claim. Performance measurement systems were perceived as neither resources nor additional demands. The effect depended on the practice. Setting many targets and focusing on productivity correlated positively with perceived job demands, which correlated negatively with wellbeing. Monitoring financial performance and employee performance correlated positively with job control and supportive management, and those perceptions correlated positively with job satisfaction and organisational commitment and negatively with anxiety. For a manager the implication is concrete: if commitment is weak and the unit runs many productivity targets, the study's pattern points to demands as the first thing to examine. Trimming targets and keeping monitoring that staff experience as support and discretion is the route the data favour. The paper does not claim causation and gives no thresholds for how many targets is 'many', so you establish your own baseline and test it.

• If perceived job demands are high and the unit runs many targets with a productivity focus, then reduce or consolidate targets before launching any commitment programme, because that pattern is linked to demands and lower wellbeing.
• If monitoring of financial or employee performance is in place and staff report low job control or low supportive management, then change how the monitoring is used (feedback, discretion, support), not whether it exists, because those perceptions are what link monitoring to commitment.
• If commitment is low but satisfaction is acceptable, then probe belonging, internalised values and willingness to give extra effort separately, because they are different constructs.
• If you ask staff for proactive problem solving or extra effort, then pair the request with recognition and support, because committed employees expect these in return.
• If you cannot show before-and-after perception data, then do not claim a practice raised commitment, because the evidence is correlational.

• Treating performance measurement as inherently corrosive to commitment and scrapping monitoring. The tell: control and support scores fall while demands stay unchanged, since monitoring was the practice associated with higher control and support.
• Adding more targets to 'drive effort'. The tell: target count and productivity focus rise, perceived demands rise, and wellbeing and attachment fall.
• Conflating commitment with satisfaction or compliance. The tell: your survey has one 'happy at work' item and no items on belonging, values or willingness to give extra effort.
• Asking for discretionary effort without offering recognition or support. The tell: staff comply with targets but do not volunteer problem solving, and supportive-management perceptions are low.
• Reading correlations as proof of cause, or skipping the mediating perceptions. The tell: you report 'monitoring raised commitment' with no data on job control, supportive management or demands.

• ▸Unit / workplace: name the team or site being audited.
• ▸Targets in force: list the number of targets and whether the emphasis is productivity.
• ▸Monitoring in force: note whether financial performance and employee performance are monitored, and how results reach staff.
• ▸Perceived job demands: summarise staff-reported effort, time pressure and task difficulty.
• ▸Perceived job control: summarise staff-reported discretion over how work is organised and delivered.
• ▸Perceived supportive management: summarise staff-reported support, acknowledgement and care.
• ▸Commitment probes: record separate readings for belonging, internalised values and willingness to give extra effort.
• ▸Wellbeing side-check: record job satisfaction and anxiety readings to compare with commitment.
• ▸Change to test and re-measure date: state the single change (e.g. fewer targets, monitoring used as feedback) and when you will re-read all fields.

1. 1Fix the construct as the final link in the chain, not a standalone KPI. The chain runs performance measurement practices (targets, monitoring), then employee perceptions (job control, supportive management, job demands), then employee reactions (job satisfaction, anxiety, organisational commitment), then workplace performance. Write this chain down before you collect anything.
2. 2Measure performance with the managerial respondent's assessments of two items: labour productivity and quality of service, each relative to the sector average. Keep them as two separate outcomes. The source runs each as its own final outcome in the structural models.
3. 3Use a 5-point scale (below average to well above average) against the sector average. Do not ask for an absolute rating. The construct only means something as a comparison with the sector.
4. 4Check each item's distribution before modelling. In the source, the two lowest labour-productivity categories are merged into a 4-point scale. Quality of service is reduced to a 3-point scale. Merge sparse categories on the evidence of the distribution, and record the merge.
5. 5Match the managerial performance rating to employee-level data at the same workplace. The source uses matched employer-employee data from a nationally representative survey (WERS 2011). The manager supplies the outcome and employees supply the perceptions and reactions.
6. 6Add each performance item as the final outcome in a separate model, with job control and supportive management feeding employee reactions, which then feed performance. This is the double mediation. Estimate by maximum likelihood, weight the employee-level paths with the employee sampling weights, and include controls: workplace size, public or private status, and individual controls (manager, gender, low earnings, full-time, qualification, ethnicity).
7. 7Apply the mediation test the source uses (MacKinnon et al., 2007). Infer mediation only if the independent variable correlates significantly with the mediator and the mediator correlates significantly with the dependent variable. Test each link in the chain on its own.
8. 8Interpret the output cautiously. Correlations in the source are weak, and the data are cross-sectional and rely on manager perception. Report which links in the chain hold, not a single total effect.

de Menezes and Escrig-Tena first test how targets and monitoring relate to employee perceptions and to job satisfaction, anxiety and organisational commitment in the British health and care sector. They then extend the chain, following the management-performance literature, so that the final outcome is workplace-level performance. WERS 2011 offers two relevant measures: the managerial respondent's assessment of labour productivity and of quality of service, each against the sector average on a 5-point scale. After checking the distributions, they merge the two lowest labour-productivity categories into a 4-point scale and reduce quality of service to a 3-point scale. Each variable is added as the final outcome in the structural equation models, and a double mediation is tested. The results confirm positive associations between some employee perceptions and reactions and these performance ratings. The supplied text does not specify which paths hold. The rationale for including the ratings is that average wellbeing in a workplace may affect overall productivity, and that commitment supports the proactive problem-solving behind quality of care.

• If a performance item has sparse lowest categories, then merge them (as the source does for productivity and service quality) and report the recoding.
• If you want to claim that targets or monitoring improve performance, then show both links of each mediation path (practice to perception or reaction, then reaction to performance). Do not claim it from the practice-to-performance correlation alone.
• If the manager's rating is the only performance source, then call it a managerial assessment relative to the sector, not an objective output measure.
• If you analyse productivity and quality of service together, then keep them as separate outcomes with separate models, because the source scales and models them separately.
• If correlations between chain variables are weak, then state that and avoid causal language beyond the associations the models support.

• Treating performance as independent of employee outcomes. The tell: the model has no mediators between measurement practices and performance, so the double mediation is missing.
• Using the raw 5-point scale without checking the distribution. The tell: near-empty bottom categories that destabilise estimates.
• Reading managerial ratings as objective productivity. The tell: conclusions that cite output levels when the data are relative-to-sector perceptions.
• Claiming mediation from one significant link. The tell: a path is called mediated although the mediator-to-outcome (or practice-to-mediator) association is not significant.
• Collapsing productivity and quality of service into one index. The tell: a single 'performance' score hides that the source scales and models them differently, with 4 and 3 points.

• ▸Productivity item: write the exact wording of the managerial question, comparing labour productivity with the sector average.
• ▸Quality item: write the exact wording of the managerial question, comparing quality of service with the sector average.
• ▸Original scale and recoding: record the original points, which categories you merged and why (cite the distribution).
• ▸Respondent: record who rates performance (manager) and how it links to employee-level data.
• ▸Chain map: list practices (targets, monitoring), perceptions (job control, supportive management, job demands), reactions (satisfaction, anxiety, commitment), then the performance outcome.
• ▸Mediation check: for each path, record whether practice-to-mediator and mediator-to-outcome are both significant.
• ▸Controls and weights: list workplace controls (size, public/private), individual controls, and the sampling weights used.
• ▸Claim level: state the strongest claim the evidence supports (association, not proof of causation), noting weak correlations and perception-based ratings.

1. 1List every distinct target type in use at the workplace, as binary yes/no flags. The study records use or non-use per type from a manager interview and does not count how many individual targets sit under each type.
2. 2Cover all target families, not only the headline ones. The study's pool includes financial targets (four items), productivity, quality, and the HR targets: training, workforce job satisfaction, level of absenteeism and labour turnover. HR targets are the ones most often left off the list.
3. 3Count the flags to get a raw intensity score, then standardise it against the full pool of possible targets. Intensity is a relative measure, so a raw count alone does not tell you whether the load is high.
4. 4Benchmark against the sector. In the source, the sector mean is 4.7 target types and the whole-data mean is 5, which is not a significant difference at the 5% level. Read your score against a mean like this, not against zero.
5. 5Keep intensity separate from the type of target. The study found no significant direct or indirect effect for any single type on work-condition perceptions. The likely impact came from intensity, so do not infer harm from one target in isolation.
6. 6Test the mechanism through job demands. Measure demands as the mean of two items, time pressure and workload (composite reliability 0.68). Rising intensity should show up first as higher perceived demands, which then relate to lower wellbeing.
7. 7Measure the offsets in the same survey: job control (five items on influence over work) and supportive management. Both relate positively to job satisfaction and organisational commitment and negatively to anxiety, so they show whether intensity is being absorbed.
8. 8Analyse by workplace. Employees in one workplace share the same target set, so cluster errors by workplace and weight employee responses. Treat intensity as a workplace-level variable, not an individual one.

de Menezes and Escrig-Tena (2023) analyse 319 workplaces and 3,600 employees in the British health and care sector (WERS2011). They build an intensity measure as the standardised number of target types used, drawing on all target possibilities in the data, including training, workforce job satisfaction, absenteeism and labour turnover. Individual target types are less common in this sector than elsewhere, yet the sector's average number of target types (4.7) is not significantly different from the whole-data average (5). Intensity is therefore not a matter of the sector having fewer or more targets than other sectors. Run through separate models for job satisfaction, anxiety and organisational commitment, no single target type shows a significant association. Many targets and a productivity focus, however, correlate positively with perceived job demands, and job demands correlate negatively with wellbeing. The lesson is to count the target families, compare the count with a sector benchmark, and watch demands, not any one target. As a hypothetical illustration, a care provider running financial, productivity, quality and all four HR targets would score well above the 4.7 sector mean. You would expect its staff to report higher time pressure and workload, and you would check whether job control and managerial support are rising to compensate.

• If your target count is above the sector mean (4.7 in the source), then survey time pressure and workload before adding any new target.
• If a manager defends a single target as the problem or the fix, then look at the whole set. The evidence ties effects to the number of target types, not to any one type.
• If intensity and a productivity focus are both high, then treat it as the higher-risk combination for demands and wellbeing, and cut or consolidate before monitoring more.
• If job control and supportive management are high, then expect demands to be offset. Intensity still needs watching, but it is less urgent than where those two are low.
• If you only have binary use-of-target data, then report intensity as a relative, standardised score and say that targets cascading beneath each type are uncounted.

• Counting only financial and productivity targets and leaving out HR targets such as absenteeism and turnover. The tell is an intensity score below what staff report feeling.
• Judging load by the type of target instead of the count. The tell is a debate over which target to drop while the total stays the same.
• Assuming a count equals a workload. The data are binary indicators of use, and one type can hold dozens of cascaded targets (the NHS Outcomes Framework example), so a low score can hide a heavy load. The tell is staff reporting heavy pressure at a workplace with a modest score.
• Reading a sector-average score as safe. The sector mean (4.7) matches the whole-data mean (5), but intensity still tracks perceived demands. The tell is 'we are at average' offered as the whole argument.
• Treating intensity as an individual-level variable and ignoring clustering. The tell is narrow confidence intervals from pooling employees who share one workplace's targets.

• ▸Workplace / unit: name the workplace or team being audited.
• ▸Financial targets in use (Y/N each): list each financial target and mark it yes or no.
• ▸Productivity targets in use (Y/N): mark whether any productivity target is set.
• ▸Quality targets in use (Y/N): mark whether any quality target is set.
• ▸HR targets in use (Y/N each): training, workforce job satisfaction, absenteeism, labour turnover.
• ▸Total target types and standardised score: count the Y flags and divide by the total possible types.
• ▸Sector benchmark and gap: record the benchmark mean (source: 4.7 for the sector) and your difference from it.
• ▸Perceived demands (staff mean): average the time-pressure and workload items.
• ▸Job control and supportive management (staff mean): record both scores as the offsets.
• ▸Action and owner: state the one target to consolidate or drop, who owns it, and the review date.

1. 1Write down the financial objectives first, because monitoring tracks progress against objectives and cannot exist without them. The source defines targets as identifying and setting objectives and milestones on different dimensions of performance. Monitoring is then the tracking of progress against those objectives, with performance and related data collected and recorded. For each financial dimension you track (sales, costs, profit, labour cost), state the objective and at least one milestone.
2. 2Translate the high-level financial objective into the language of the team. The source says targets exist to translate strategy into team terms and signal to each employee what to do. A unit head who sees only the organisation-wide profit figure has nothing to act on. Give each unit a line it can read and influence.
3. 3Set up a recurring record for each dimension: the figure, the period, the objective, and the gap. The source's definition stresses collecting and recording data. A figure that is not written against its objective is a number, not monitoring.
4. 4Treat monitoring as a bundle of practices. The study measured monitoring of financial performance as one component built from four binary practice questions, and the first component explained 72% of their variance. Monitor sales, costs, profit and labour cost as one routine rather than as four separate projects.
5. 5Review the records on a fixed cadence and use them to decide something: where resources go, which inefficiency to fix, which operation to adjust. The source's case for performance measurement is timely, accurate data for allocating resources and spotting inefficiencies. A review that changes no decision is reporting only.
6. 6Feed the results back to the people who generate them, with the objective alongside the figure. The source links clear targets and data-driven feedback to learning that improves processes. Share the gap and what it means for their work, not a bare number.
7. 7Keep the target count low while you monitor. In the study, setting many targets and a productivity focus correlated with higher perceived job demands, and job demands correlated with lower wellbeing. Financial monitoring was not the source of that pressure. Adding financial monitoring does not require adding more targets.
8. 8Pair the numbers with visible management support. In the study, financial monitoring was associated with employees perceiving more job control and more supportive management, which in turn correlated with higher job satisfaction and commitment and lower anxiety. These are associations, not proven causes. Use them as a reason to run monitoring openly and consistently.

The best real case in the source is the study itself. De Menezes and Escrig-Tena used the 2011 Workplace Employment Relations Study (WERS) in British health and care. They scored 'monitoring financial performance' from the first four monitoring questions in their appendix, with the first principal component explaining 72% of the variance. Financial performance targets were scored the same way from the first four target questions. The two were measured separately. In the correlation table, financial monitoring at workplace level correlates 0.30 with financial targets, and with job satisfaction (0.16) and organisational commitment (0.13). Their structural models link financial monitoring to perceived job control and supportive management, which link to higher satisfaction and commitment and lower anxiety. By contrast, many targets and a productivity focus link to perceived job demands, which link to lower wellbeing. The authors conclude that monitoring financial performance 'may be helpful to managers', and that performance measurement systems are neither clearly demands nor resources. The operational lesson is to run financial monitoring as a distinct, well-recorded routine, and not to pair it with a long list of competing targets. The evidence is correlational and cannot show that monitoring causes better employee outcomes.

• If you cannot name the financial objective and milestone for a figure, don't track it yet; define the objective first.
• If a unit's target list is growing (the sector average in the study was about 4.7 target types), then cut targets before adding any monitoring, because the number of targets is what correlated with job demands.
• If a financial figure has no owner who can influence it, then re-express it at a level the team controls, or drop it. Targets must be in the team's language.
• If a review produces no resource or operational decision, then change the review's content or cadence, not just its attendance.
• If staff read monitoring as surveillance, then raise the management support around it (explain purpose, share results, act consistently) before tightening the data.

• Tracking figures with no stated objective. The tell: reports show numbers but no gap to target, and nobody can say what 'good' is.
• Stacking financial monitoring on top of many targets and a productivity push. The tell: staff describe the workload as rising and the measures as competing, which is the job-demands pattern the study found.
• Monitoring that is collected but never fed back. The tell: staff learn their results only when something goes wrong, so the learning benefit the source describes never appears.
• Using unrealistic or ambiguous targets, or measures meaningless to the employee. The tell: staff cannot explain how their daily work moves the figure. The source lists these as the common criticisms of performance measurement systems.
• Treating the study's associations as proof of causation. The tell: someone claims monitoring will raise satisfaction without managing the other conditions. The source says only that monitoring 'may be helpful'.

• ▸Financial dimension: sales, costs, profit or labour cost. Pick one per row.
• ▸Objective and milestone: state the target value and the interim checkpoint, with dates.
• ▸Team-level expression: restate the objective in terms the team can read and influence.
• ▸Data source and recorder: say where the figure comes from and who records it each period.
• ▸Period figure vs objective: enter the actual, the objective and the gap.
• ▸Review date and decision made: record what resource or operational choice the review produced, or 'none' as a warning sign.
• ▸Feedback given: record to whom, in what form, and when the results were shared.
• ▸Total active targets in this unit: count them to check against overload and job demands.
• ▸Management support check: note how the monitoring was explained and what help was offered to close the gap.

1. 1Write down what you track and why. Use the paper's definition: a performance measurement system involves setting goals, developing measures, collecting, analysing, reporting, interpreting, reviewing, and acting on performance data. If productivity records stop at collection, you are keeping records, not monitoring.
2. 2List every productivity-related measure and tie each one to a decision it informs, such as allocating resources, spotting inefficiency, or adjusting operations. Drop any measure that informs no decision.
3. 3Count your targets separately from your monitoring. The paper finds that setting many targets and a focus on productivity targets correlate with higher perceived job demands, which correlate with lower wellbeing. Monitoring financial and employee performance showed the opposite pattern, with more job control and supportive management. Audit each lever on its own.
4. 4Translate each productivity measure into team-level language so each employee knows what is expected and how they add value to output. Targets should signal to each employee what to do. A measure staff cannot interpret is one the criticism literature calls meaningless to the employee.
5. 5Close the loop with data-driven feedback. Clear targets plus feedback support the process learning that can benefit employees and employers. Schedule a regular review where staff see the data and discuss process changes, not only individual numbers.
6. 6Separate productivity monitoring from employee-performance evaluation. A recurring criticism is evaluations unrelated to performance objectives. Check that anything used to judge a person traces to a stated objective.
7. 7Pair productivity monitoring with quality monitoring. In the data they move together (r = 0.42). Throughput tracked alone invites unethical behaviour by staff striving to meet expectations, and weakens the 'efficiency and quality of outcomes' balance the paper says managers need.
8. 8Survey staff on job control, supportive management, job demands, satisfaction, anxiety and commitment. These are the paper's mediators and outcomes. Compare before and after changes to the monitoring regime, and treat a rise in perceived demands as the early warning.

Florence Nightingale is the paper's origin case. In 1860 she collected and analysed data on hospital units to argue for operational change and performance improvement. The record served a decision, and the data was used to change operations rather than to police individuals. Apply the same test to a modern health or care workplace using the paper's evidence. In the WERS-based data, monitoring productivity correlates with productivity targets (0.43), quality monitoring (0.42) and financial monitoring (0.37). A manager who switches on productivity monitoring is therefore usually adding a bundle of practices, not one. At employee level, job demands correlate positively with anxiety (0.38) and negatively with job satisfaction (-0.15), while supportive management correlates strongly with satisfaction (0.66). The sequence is: (1) inventory the bundle you are actually running; (2) count the productivity targets; (3) check whether staff report rising demands; (4) if they do, cut targets before cutting the monitoring, and keep the feedback sessions that build supportive management. This follows the paper's finding that targets and productivity focus raise perceived demands while some monitoring is linked to control and support.

• If you are adding productivity targets alongside monitoring, then cap the number of targets first. Many targets, not monitoring itself, are tied to higher perceived job demands.
• If a productivity metric informs no resource, process or staffing decision, then stop collecting it. It adds records without the learning that justifies monitoring.
• If you monitor productivity, then also monitor quality, and report the two together. They co-occur in practice and guard against output-at-any-cost behaviour.
• If staff survey scores show rising job demands or anxiety after a rollout, then reduce target intensity and increase feedback and manager support before adding measures. Demands relate to lower wellbeing, support to higher satisfaction and commitment.
• If you want to know whether monitoring helps or hurts, then do not assume either. The paper finds performance measurement systems are neither perceived as resources nor as additional demands, so test in your own workplace.

• Treating monitoring and targets as one thing. The tell: you can't say which practice staff blame for pressure. The paper links targets and productivity focus to demands, and monitoring financial and employee performance to control and support.
• Collecting without acting. The tell: dashboards exist but no process has changed after a review. This misses the analyse, interpret, review and act steps of a performance measurement system.
• Measures staff don't recognise. The tell: employees can't explain how a metric reflects their work. This is the criticism of meaningless measures and ambiguous or unrealistic targets.
• Productivity metrics without quality metrics. The tell: output figures improve while error or quality complaints rise. The criticism literature points to unethical behaviour by employees striving to meet expectations.
• Monitoring as blame. The tell: data is discussed only after problems, naming individuals. Deming argued that targets and monitoring can produce a culture of blaming others rather than the desired employee outcomes.

• ▸Metric name: write each productivity measure you currently record.
• ▸Decision informed: state the resource, process or staffing decision this metric feeds.
• ▸Linked objective: name the team-level objective or national target it traces to.
• ▸Paired quality measure: list the quality metric reported next to it.
• ▸Number of active productivity targets: count them for the team, and note any added in the last period.
• ▸Feedback loop: record how and how often staff see the data and discuss process changes.
• ▸Staff-reported job demands and control: record the latest survey result for each, with date.
• ▸Staff-reported supportive management, anxiety and commitment: record the latest results alongside.
• ▸Action this period: record keep, cut, or change, with the owner and review date.

1. 1Set the scope first. In the source study, 'monitoring employee performance' counts only when all non-managerial staff have their performance formally appraised (coded 1; anything less is 0). Partial or ad hoc appraisal does not meet the definition, so list which non-managerial roles are covered and which are not.
2. 2Trace each appraised role back to strategy. Targets exist to translate high-level strategy into the language of teams and signal to each employee what to do. Write one line per role showing how its expectations connect to a workplace objective. If you cannot, the appraisal has nothing to measure against.
3. 3Cap the number of targets and avoid a productivity-only focus. The study finds that setting many targets and focusing on productivity correlate with higher perceived job demands, which correlate with lower wellbeing. Appraise against a short list that includes quality, not volume alone.
4. 4Check that each appraisal criterion is tied to a stated performance objective. Documented failures include evaluations unrelated to performance objectives, ambiguous or unrealistic targets, and measures that are meaningless to the employee. For every criterion, ask whether the employee can say what it measures and how they influence it.
5. 5Pair the appraisal with data-driven feedback. Clear targets plus feedback support the learning that improves processes and can benefit both employees and employers. Return the data to the person being appraised, not only to management.
6. 6Protect job control and supportive management while you appraise. These are the two perceptions the study links to employee monitoring. Keep employee influence over tasks, pace, method, task order and start/finish times in view, and make the appraiser's role supportive rather than only evaluative.
7. 7Measure the employee's side. Ask employees how much influence they have over the five job-control items (1 = none, 4 = a lot) and how strongly they agree with job-demand statements (1–5). Track job satisfaction, anxiety and organisational commitment alongside them.
8. 8Read results as associations, not proof. The study estimates structural equation models on matched employee-workplace survey data and infers pathways. It also notes that outcomes may be contingent, so test the effect in your own setting.

The study draws on the UK's WERS2011 matched employee-workplace survey for the health and care sector. Managers were asked, 'Is the performance of employees assessed?' The practice was coded 1 only if all non-managerial staff have their performance formally appraised, and 0 otherwise. Employees at the same workplaces rated how much influence they had over their tasks, pace, method, task order and working hours (1 = none, 4 = a lot). They also rated job demands, job satisfaction, anxiety and commitment. Formal appraisal of employees was positively associated with perceived job control and supportive management. Those two perceptions in turn correlated positively with job satisfaction and organisational commitment, and negatively with anxiety. By contrast, many targets and a productivity focus correlated with higher perceived job demands and lower wellbeing. The authors conclude that performance measurement systems are neither clearly resources nor additional demands, and that employee monitoring may help managers. The lesson is to appraise everyone formally, keep the target set lean, and check job control and support in the same cycle.

• If appraisal covers only some non-managerial staff, then treat it as outside the construct as the study defines it and extend coverage, or accept that you are running a different practice.
• If employees report high job demands after you add targets, then cut the number of targets and reduce the productivity emphasis before changing the appraisal itself. Many targets and productivity focus are the factors linked to demands.
• If an appraisal criterion cannot be linked to a stated performance objective, then drop or rewrite it. Evaluations unrelated to objectives are a documented source of criticism.
• If job-control or supportive-management ratings fall after appraisal begins, then redesign the process before expecting gains in satisfaction or commitment. These perceptions are the route to those outcomes.
• If your setting differs from the study's (UK health and care, survey-based), then test locally before generalising, because outcomes of performance management may be contingent.

• Appraising only some staff or only informally, then claiming a formal system. The tell: you cannot name the share of non-managerial staff with a formal appraisal.
• Piling on targets and stressing productivity. The tell: staff describe work intensification, information overload or role conflict, and demand ratings rise.
• Appraisal criteria that do not match stated objectives or mean nothing to the employee. The tell: staff cannot explain what a measure is for, or the ratings do not track the objectives.
• Using appraisal as blame assignment. Deming warned that targets and monitoring can create a culture of blaming others. The tell: discussions focus on who failed rather than on process learning.
• Treating the study's associations as guaranteed effects. The tell: you skip measuring job control, support, satisfaction, anxiety and commitment in your own workforce.

• ▸Role groups covered: list every non-managerial role group and mark whether all staff in it are formally appraised (yes/no).
• ▸Coverage check: state whether ALL non-managerial staff are formally appraised; if not, name the gap.
• ▸Strategy link: for each role, write the workplace objective its targets translate.
• ▸Target count and productivity weight: count the targets per role and note whether productivity dominates over quality.
• ▸Criterion-objective match: for each appraisal criterion, name the performance objective it measures.
• ▸Employee clarity test: record whether staff can state what each measure means and how they influence it (yes/no).
• ▸Feedback loop: describe how appraisal data return to the employee and by when.
• ▸Job-control pulse: average employee rating (1–4) on influence over tasks, pace, method, task order and working hours.
• ▸Wellbeing and commitment readout: record job satisfaction, anxiety, organisational commitment and job-demand ratings, and note changes since the last cycle.

1. 1Define job control as perception, not policy. Ask employees how much discretion they feel they have, not what the manual allows. The construct is perceived decision latitude over how work is organised and delivered, and the study treats it as an employee perception sitting between management practice and outcomes.
2. 2Split the construct into its five parts: tasks, pace, methods, order, and timing. Rate each separately, because a target regime can leave methods open while fixing timing. A single 'autonomy' score hides which part a practice is squeezing.
3. 3List every target and monitoring practice in use. The source treats these two as the core of a performance measurement system: targets identify and set objectives and milestones, and monitoring tracks progress and records performance data. Note for each practice what it measures (financial, employee, productivity) and how many targets each person carries.
4. 4Map each practice to its likely effect on job control, using the study's pattern. Monitoring financial performance and monitoring employee performance are positively associated with perceived job control and supportive management. Setting many targets and a focus on productivity are positively associated with perceived job demands, which correlate negatively with wellbeing. Treat these as associations, not proven causes.
5. 5Check the downstream outcomes tied to job control. Higher perceived control goes with higher job satisfaction and organisational commitment and lower anxiety. Collect those three measures next to the job-control rating so you can see whether a practice that raises or lowers control moves them in the expected direction.
6. 6Test the tension between standardisation and discretion. Much healthcare work is routine and benefits from standardised, measurable procedures. Health workers also face interdependent tasks and must prioritise, act quickly, and act independently. Mark which tasks are standardised and which need independent judgement, and protect discretion over method and order in the second group.
7. 7Review the clarity of targets as a control lever. In the English child and adolescent mental health case, unclear targets caused confusion but also enabled creative, proactive solutions and encouraged autonomy and agility. Decide deliberately where to leave room and where ambiguity only produces confusion.
8. 8Pair any change to targets or monitoring with supportive management. In the findings, monitoring practices are linked to both job control and supportive management, and both relate positively to satisfaction and commitment. Do not change monitoring without checking how managers present and use the data.

The source contrasts two readings of the same practices in healthcare. Critics such as Wankhade (2012) argued that emphasising targets and monitoring means control over the provision of care would be lost. McCann et al. (2015) equated performance measurement systems with information overload, work intensification and role-conflict. The case study of child and adolescent mental health services in England (Ojiako et al., 2022) found a different picture. Unclear targets were a source of confusion, but they also enabled creative and proactive solutions and encouraged autonomy and agility. The survey analysis of nationally representative matched employee data by de Menezes and Escrig-Tena then separates the practices. Monitoring financial performance and monitoring employee performance are positively associated with perceived job control and supportive management, which go with higher satisfaction and commitment and lower anxiety. Many targets and a productivity focus are instead associated with higher perceived demands and worse wellbeing. A manager applying this would not drop monitoring to protect control. They would audit target count and productivity emphasis first, and keep monitoring that staff experience as supportive.

• If staff report many targets or a strong productivity focus, then check perceived demands and anxiety before looking at control, because those practices are linked to demands rather than to control.
• If you are deciding whether to keep financial or employee-performance monitoring, then keep it where staff report higher control and supportive management, since the study links it to both.
• If tasks are interdependent and need quick, independent prioritising, then fix the objective but leave method, order and timing to the employee.
• If a target is unclear, then ask whether staff use the gap for creative solutions or are confused by it. Clarify it only in the confusion case.
• If job control is low and anxiety is high together, then treat the pattern as a warning that the practices are being experienced as demands, not resources, and review them before adding new ones.

• Equating monitoring with loss of control. The tell: you cut monitoring on the critics' argument, yet your own staff ratings show monitoring tracks with higher control and support.
• Measuring policy instead of perception. The tell: the rulebook shows wide discretion, but employees rate their latitude low on pace or timing.
• Using one blended autonomy score. The tell: the average looks fine while one dimension (for example timing or order of work) is tightly fixed.
• Adding targets because each is individually reasonable. The tell: target count rises, perceived demands rise, and satisfaction and anxiety worsen while control does not improve.
• Treating all ambiguity as a defect. The tell: you clarify every target and remove the room that staff were using for proactive solutions.

• ▸Practice and type: name the target or monitoring practice and whether it covers financial, employee or productivity measures.
• ▸Number of targets per employee: count the targets each role carries.
• ▸Task standardisation: mark whether the work is routine and standardised or interdependent and judgement-based.
• ▸Control rating by dimension: rate perceived discretion from low to high for tasks, pace, methods, order and timing.
• ▸Target clarity: mark clear or unclear. If unclear, note whether staff report confusion or creative solutions.
• ▸Perceived demands: rate whether staff experience the practice as extra demand, and note any overload or role-conflict.
• ▸Supportive management: rate how managers use the data, as support or as blame.
• ▸Outcome check: record job satisfaction, organisational commitment and anxiety for the same group.
• ▸Decision and owner: record the keep, adjust or remove decision, who owns it, and when you will re-rate.