DKM&X is a large industrial company manufacturing tools and equipment and offering consultancy services, with 450 staff at its United Kingdom headquarters and strategic business units and offices across the country. The author leads a team of full-time and part-time staff, some headquarters-based and some mobile, and liaises with customers, suppliers and regulatory organisations. Performance and benchmarking information is presented monthly to the senior management team and compliance data is provided regularly to regulatory bodies. Organisational detail beyond the scenario is illustrative.
Table of Contents
Task 1a: Account on the use of data and information in decision making
AC 1.1 Analyse the use of data and information in decision making
The distinction between data and information. Data are raw, unprocessed facts. Information is data organised, contextualised and given meaning. Analysing why this matters practically, the number 4,212 is data; 4,212 units produced against a target of 4,500 in a week when two machines were down is information, and only the second supports a decision. DKM&X generates data continuously and produces information intermittently, and the gap between the two is where most managerial frustration sits.
Use at different organisational levels. Analysing how use varies, strategic decisions at board level draw on aggregated, external and forward-looking information over long horizons, such as market data and benchmarking. Tactical decisions at business unit level use monthly performance information to allocate resources within a year. Operational decisions use daily and real-time data on production, quality, customers and suppliers. Analysing the implication, the same underlying data serves all three and must be presented entirely differently for each, which is a common point of failure.
Reducing uncertainty rather than removing it. Analysing what data actually contributes, decisions are made under uncertainty and data narrows the range of plausible outcomes without eliminating it. A manager who believes data delivers certainty will over-rely on it; one who dismisses it will decide on instinct alone. The defensible position treats data as evidence to be weighed alongside experience and judgement.
Specific uses in this organisation. Monitoring performance against target. Identifying trends before they become problems, since a rising quality rejection rate is visible in data weeks before it appears in customer complaints. Benchmarking against comparable operations. Supporting the business case for investment. Meeting compliance obligations to regulators. Allocating resources between business units. Evaluating whether a previous decision worked, which is the use most often omitted.
Analysing the limits and risks. Data can mislead as readily as inform. Correlation is routinely mistaken for causation. Measures drive behaviour, so what is measured becomes what is optimised regardless of whether it matters. Mullins (2022) observes that control information shapes the conduct it monitors. Selective use to justify a decision already taken is common and difficult to detect. And an over-supply of data produces paralysis, since a monthly pack of forty pages guarantees that nothing in it receives attention.
Analysing the balance. Data-informed decision making outperforms intuition alone in stable, well-measured domains and offers less advantage in novel situations where no relevant history exists. Buchanan and Huczynski (2023) note that managerial decisions are frequently made under conditions of bounded rationality, where information is incomplete and time is short, which describes most real decisions at DKM&X rather than the idealised version.
AC 1.2 Examine types of data and information used in decision making
Quantitative and qualitative. Quantitative data are numerical and support measurement and comparison, such as production volumes and rejection rates. Qualitative data are descriptive, such as customer feedback and consultancy client comments. Examining their relationship, quantitative data establish what is happening and qualitative data explain why, and a decision informed by only one will be either unexplained or unevidenced.
Primary and secondary. Primary data are collected first-hand for a specific purpose, such as a customer survey DKM&X commissions. Secondary data already exist, having been gathered for another purpose, such as industry benchmarking reports. Examining the trade-off, primary data are relevant and expensive; secondary data are cheap and may not fit the question.
Internal and external. Internal sources include production systems, quality records, sales figures, finance and human resources data. External sources include market reports, competitor information, regulatory publications and benchmarking data. Examining the scenario specifically, DKM&X uses both for benchmarking, performance management and reporting, and the two must be reconciled to a common basis before comparison means anything.
Structured and unstructured. Structured data sit in defined fields and are readily analysed. Unstructured data include emails, meeting notes, engineer reports and free-text customer comments. Examining the practical position, the majority of an organisation’s data is unstructured and the minority that is structured receives nearly all the analytical attention.
Real-time, historical and predictive. Examining the temporal dimension, real-time data support immediate operational intervention, historical data support trend analysis and accountability, and predictive information supports planning. DKM&X’s daily production and quality data are real-time; the monthly senior management pack is historical; the annual demand forecast is predictive.
Compliance and regulatory information. Examining a category specific to the scenario, information supplied to regulatory bodies is a distinct type with its own accuracy, format and timeliness requirements, and it is not optional.
Big data and its characteristics. Examining a category increasingly present in industrial manufacturing, high-volume machine-generated data from production equipment arrives at speed and in varied formats, and is conventionally described by its volume, velocity, variety and veracity. Examining its relevance to DKM&X, sensor data from production lines could support predictive maintenance, and the constraint is not the data’s availability but the absence of anyone able to analyse it. Examining a general point this illustrates, data volume without analytical capability produces storage cost rather than insight.
Examining type selection. The type required follows the decision. Whittington et al. (2023) note that strategic questions demand different evidence from operational ones, and the common error at DKM&X is applying operational precision to strategic questions, where a defensible estimate would serve better than a precise number that arrives too late.
Task 1b: Proposal on the impact of stakeholders, organisational and legal frameworks on the use of data and information
AC 1.3 Analyse the impact of organisational and legal frameworks on the use of data and information in decision making
Data protection legislation. The UK General Data Protection Regulation and the Data Protection Act 2018 govern personal data. Analysing the impact on DKM&X, the framework requires a lawful basis for processing, limits collection to what is necessary, restricts retention, requires accuracy, and confers rights on individuals including access and erasure. Practically this constrains what customer and employee data the team may hold, for how long, and what analysis is permitted. Analysing a specific consequence, mobile staff carrying customer information on devices create a security and compliance exposure that headquarters-based working does not.
Accountability and demonstrability. Analysing a requirement organisations underestimate, the framework requires not merely compliance but the ability to evidence it, through records of processing, impact assessments for higher-risk activities and documented decisions. A team that complies without documentation cannot demonstrate compliance.
Freedom of information and transparency obligations. Analysing their relevance, DKM&X is a private company and not directly subject to these, and information it supplies to regulatory bodies may enter the public domain through them, which affects what is committed to writing.
Sector and regulatory reporting requirements. Analysing the impact of the compliance reporting described in the scenario, regulatory bodies specify format, content, frequency and accuracy, and these requirements determine how underlying data must be captured. A reporting obligation imposed externally becomes a systems requirement internally.
Intellectual property and confidentiality. Analysing this dimension, consultancy work generates client information held under confidentiality obligations, and the boundary between insight DKM&X may reuse and client information it may not is a live constraint on internal knowledge sharing.
Organisational frameworks. Analysing the internal dimension, data governance policy determines ownership and access rights; information security policy determines how data is stored and transmitted; retention schedules determine deletion; and the scheme of delegation determines who may authorise disclosure. Analysing which of these bites hardest day to day, access rights do, since a manager who cannot obtain data cannot use it, and DKM&X’s access controls were designed for security rather than for analysis.
Systems architecture as a de facto framework. Analysing a constraint rarely recognised as one, the business unit systems at DKM&X record data on different bases, so cross-unit comparison requires manual reconciliation. The systems architecture therefore determines what analysis is practically possible, irrespective of what policy permits.
Employment and monitoring frameworks. Analysing a constraint relevant to a team with mobile members, monitoring staff activity through vehicle tracking or system logging engages both data protection law and the employment relationship, and monitoring introduced without consultation and a stated purpose damages trust disproportionately to whatever it detects.
Analysing the combined effect. Legal frameworks set the outer boundary and organisational frameworks determine practice within it. Mullins (2022) observes that internal controls shape behaviour most directly at the point where permission ends, and at DKM&X that point is the access request rather than the statute.
AC 1.4 Discuss the impact of stakeholder needs on the collection, analysis and interpretation of data and information for decision making
Identifying whose needs are in play. The scenario names several stakeholder groups with genuinely different requirements: the senior management team receiving monthly performance and benchmarking information; regulatory bodies receiving compliance data; customers and professional services clients; suppliers; and the author’s own team, both office-based and mobile.
Impact on collection. Discussing how stakeholder needs to shape what is gathered, the requirement drives the capture. Regulatory reporting requires specified fields recorded at specified points, which obliges the team to collect data it would not otherwise need. Client reporting on consultancy engagements requires activity and outcome data captured at engagement level rather than aggregated. Discussing the cumulative effect, a team serving five stakeholder groups collects for all of them, and the collection burden falls on staff who see no use for most of it, which is a principal cause of poor data quality at source.
Impact on analysis. Discussing how needs shape treatment, senior management wants variance against target and trend; regulators want absolute compliance figures with no interpretation; customers want performance against their own service expectations. Discussing the practical consequence, the same production data must be cut three ways, and an analysis produced for one audience is rarely usable by another.
Impact on interpretation and framing. Discussing the most sensitive dimension, stakeholders bring different questions and different tolerances. A regulator interprets a variance as a potential breach; the senior management team interprets the same figure as a performance issue; the operational team interprets it as a resourcing problem. Discussing the risk this creates, the temptation to frame interpretation to suit the audience shades quickly into presenting the same facts to imply different conclusions, which is a professional integrity question rather than a communication one.
Conflicting stakeholder needs. Discussing a real tension at DKM&X, customers request performance data that would reveal capacity constraints the commercial team would prefer not to disclose. The resolution is a policy decision about disclosure rather than a data decision, and it must be taken above the level of the person compiling the report.
Timing and decision cycles. Discussing this dimension, stakeholder needs operate on different cycles: daily for the operational team, monthly for senior management, quarterly or annually for regulators and clients. Discussing the design implication, data must be captured at the finest granularity any stakeholder requires, since aggregation is possible afterwards and disaggregation is not.
Accessibility and capability of the audience. Discussing an often-neglected factor, stakeholders differ in their ability to interpret data. Senior managers are not analysts and regulators are specialists. Northouse (2025) notes that communication which the recipient cannot act upon has not achieved its purpose, which argues for tailoring presentation to capability rather than to the analyst’s preference.
Trust and willingness to supply data. Discussing a factor that determines whether collection succeeds at all, stakeholders who supply data need confidence that it will be used for the stated purpose. Suppliers asked for cost breakdowns will provide them where the purpose is joint improvement and will not where they suspect the intention is to squeeze price. Evidence links the willingness to disclose to whether disclosure is perceived as safe (Capezio et al., 2023), and the same principle governs internal reporting of quality problems.
Discussing stakeholder needs overall. Stakeholder requirements are the starting point for the whole data cycle rather than a consideration at the reporting stage. A collection system designed without reference to who will use the output produces data that satisfies nobody and is resented by those who capture it.
Task 2: Good practice guide on interpreting data and information to support decision making
AC 2.1 Discuss criteria used for selection of data and information
Relevance to the decision. Discussing the first criterion, data must bear on the question actually being asked. The commonest failure at DKM&X is assembling available data rather than required data, producing reports full of what the system generates easily.
Accuracy. Discussing this criterion, data must be correct within a tolerance appropriate to the decision. Absolute accuracy is expensive and frequently unnecessary; a strategic decision tolerates approximation that a compliance return does not.
Currency. Discussing timeliness, data must reflect the period the decision concerns. Production data three weeks old supports explanation rather than intervention.
Completeness. Discussing the criterion most often overlooked, a dataset with material gaps will mislead in ways that are invisible. Missing records are rarely missing at random, and the pattern of absence usually carries information.
Reliability and source credibility. Discussing external data in particular, benchmarking reports differ in methodology, sample and independence, and a figure from an industry body with a commercial interest requires more scepticism than one from a statistical authority.
Consistency and comparability. Discussing a constraint specific to DKM&X, business units record on different bases, so comparison requires reconciliation to a common definition first. Comparing unlike figures produces confident and wrong conclusions.
Sufficiency and proportionality. Discussing this criterion, enough data to support the decision without so much that it obscures. This also engages the data protection principle of collecting only what is necessary.
Cost and accessibility. Discussing the practical criterion, data that would be ideal but requires three weeks to assemble is not available for a decision due on Friday, and the realistic question is what is the best evidence obtainable in the time.
Bias and provenance. Discussing a criterion requiring judgement, who collected the data and why affects what it shows. Customer satisfaction data collected by the account manager whose performance it reflects should be treated accordingly.
Ethical acceptability. Discussing a criterion that sits alongside the technical ones, data may be relevant, accurate and lawfully obtainable while its use remains inappropriate. Analysing an individual’s performance from system access logs collected for security purposes is an example, since the collection was lawful and the repurposing was not disclosed.
Discussing the criteria collectively. They frequently conflict, and selection is a judgement about which to prioritise. Relevance and accuracy are non-negotiable; currency, completeness and cost are traded against one another according to the decision.
AC 2.2 Evaluate the use of tools and techniques for analysing and interpreting data and information to support decision making
Descriptive statistics. Measures of central tendency and dispersion. Evaluating their use, they summarise a dataset quickly and are widely misused: an average conceals distribution, and reporting mean production without variance hides the fact that one shift performs consistently while another oscillates. Reporting a measure of spread alongside a mean is the single cheapest improvement available to DKM&X’s reporting.
Trend and time series analysis. Plotting a measure over time to identify direction and seasonality. Evaluating this technique, it distinguishes a genuine movement from a normal fluctuation, which a month-on-month comparison cannot. Its limitation is that it requires enough history and assumes the underlying conditions have not changed.
Variance analysis. Comparing actual against plan and decomposing the difference by cause. Evaluating its value, separating volume effects from price and usage effects converts a single unexplained number into three actionable ones.
Benchmarking. Comparing performance against internal units or external comparators. Evaluating this technique in the scenario’s context, internal benchmarking across DKM&X’s business units is more reliable than external because the definitions can be controlled, and external benchmarking is more challenging because comparators differ in ways raw figures conceal.
Correlation and regression. Examining relationships between variables. Evaluating their use, they identify association and quantify its strength, and they do not establish causation. Evaluating the risk, a correlation between overtime and quality defects may reflect fatigue, or may reflect that overtime is worked when volumes are high and volumes drive defects.
Data visualisation and dashboards. Evaluating this category, visualisation reveals pattern faster than tabulation and is the technique most likely to be misapplied, since chart design choices including axis scaling can materially change the impression a figure creates.
Root cause techniques. Structured investigation from a symptom to its underlying cause. Evaluating their contribution, statistical techniques identify where a problem sits and root cause techniques establish why, and analysis stopping at the first is incomplete.
Forecasting and scenario modelling. Evaluating these, they support planning under uncertainty, and their output is only as sound as the assumptions, which should be stated rather than embedded.
Artificial intelligence and automated analysis. Evaluating a category now genuinely available to an organisation of DKM&X’s size, machine learning techniques can identify patterns in large datasets that conventional analysis would not surface, and generative tools can accelerate the drafting of analysis and commentary. Evaluating the constraints honestly, three apply. The output requires verification, since a plausible-sounding summary may misstate what the data shows. Personal data processed through external tools engages the legal framework discussed earlier. And most managers reach their role without preparation for evaluating such output (Chartered Management Institute, 2023), which means the risk is not the technology but uncritical acceptance of what it produces.
Evaluative conclusion. Technique selection follows the question. Descriptive statistics summarise, trend analysis detects change, variance analysis attributes cause, correlation tests relationship, and visualisation communicates. As Maylor and Turner (2022) put it, most analysis goes wrong not because the method was weak but because the wrong question was asked well. Framing the question is therefore the step that precedes any choice of tool.
Task 3: Account on presenting data and information to meet stakeholder needs
AC 3.1 Evaluate TWO methods of presenting data and information used for decision making
Method one: the visual dashboard. A screen or single page presenting selected measures graphically, typically with current position, comparison against target and trend, updated automatically from source systems.
Evaluating its strengths for DKM&X, a dashboard suits the monthly senior management presentation because it conveys position quickly to an audience that is not analytical and has limited time. It permits comparison across business units at a glance, updates without manual effort once built, and the trend element distinguishes a genuine movement from a normal fluctuation. For the mobile members of the team it is accessible remotely, which a printed pack is not.
Evaluating its weaknesses honestly, a dashboard shows what is happening and rarely why, so it generates questions it cannot answer. Design choices carry more influence than users appreciate: a truncated axis exaggerates a movement and a chosen comparison period can flatter or condemn. Dashboards proliferate, and one carrying thirty measures directs attention to none. And automation creates a false impression of reliability, since a dashboard drawing on the inconsistent business unit definitions described earlier presents reconciled-looking figures that are not reconciled.
Adapting the dashboard to different decision cycles. Evaluating a refinement the scenario invites, the same dashboard cannot serve the team’s daily operational review and the monthly senior management presentation. The daily view needs current production, quality and exception alerts; the monthly view needs trend, benchmark comparison and variance against plan. Building one view and presenting it to both audiences means one of them receives detail it cannot use.
Method two: the written report with supporting tables. A structured narrative document setting out the position, the analysis, the interpretation and the recommendation, with tabulated data supporting the argument.
Evaluating its strengths, a written report carries reasoning that a dashboard cannot. It states assumptions, acknowledges limitations, explains why a figure moved and recommends a course of action, which is what a decision actually requires. It provides a record, which matters for the compliance reporting the scenario describes, since a regulator requires a documented and defensible position rather than a visual summary. It also allows the author to control sequence, ensuring a caveat is read before a conclusion.
Evaluating its weaknesses, reports are slow to produce and slower to read, and senior audiences frequently read only the summary, which means the reasoning the format exists to carry goes unread. Length invites padding. And a narrative can be shaped to lead a reader towards a preferred conclusion more subtly than a chart can.
Evaluating the two against each other. They serve different purposes and the choice should follow the decision rather than the preference of the producer. Where the audience needs to know the current position quickly and will ask questions, a dashboard is appropriate. Where a decision requires reasoning, assumptions and a recommendation, or where a defensible record is needed, a written report is appropriate. Buchanan and Huczynski (2023) note that information is only useful where the recipient can act on it, and at DKM&X the practical combination is a dashboard for the standing monthly review, supported by a short written commentary explaining the two or three movements that matter and what is proposed about them.
References
Buchanan, D.A. and Huczynski, A.A. (2023) Organizational behaviour. 11th edn. Harlow: Pearson.
Data Protection Act 2018, c. 12. Available at: https://www.legislation.gov.uk/ukpga/2018/12 (Accessed: 15 August 2026).
Maylor, H. and Turner, N. (2022) Project management. 5th edn. Harlow: Pearson.
Mullins, L.J. (2022) Management and organisational behaviour. 12th edn. Harlow: Pearson.
Northouse, P.G. (2025) Leadership: theory and practice. 10th edn. Thousand Oaks, CA: SAGE.
Whittington, R., Regnér, P., Angwin, D., Johnson, G. and Scholes, K. (2023) Exploring strategy: text and cases. 13th edn. Harlow: Pearson.