Advanced Performance Management · Technology and information systems

Using Data to Drive Performance Insights in APM

Updated 11 October 2026 · Fact-checked

Data are raw facts. Information is data that has been processed, put in context and made useful for a decision. Analytics and visualisation tools, such as dashboards, do this processing. In APM, you show how they link to strategy, support decisions and what limits their value.

Understand Using Data to Drive Performance Insights

Data are raw facts and figures, such as individual sales transactions, machine sensor readings or website clicks. On their own they tell a manager very little. Information is data that has been processed, organised and put in context, so it helps someone make a decision. A list of 10,000 sales lines is data. A chart showing that sales in one region fell for three months while costs rose is information.

Analytics is how you move from data to information. Descriptive analytics shows what happened. Diagnostic analytics asks why it happened. Predictive analytics estimates what is likely to happen next. Prescriptive analytics suggests what to do about it. Good performance reporting uses more than the first type.

Data visualisation presents data in charts, graphs and maps so patterns, trends and outliers are easy to see. A dashboard brings the key measures onto one screen, often updated in near real time. It usually shows KPIs against targets, using traffic lights, gauges, trend lines and drill-down to detail. It lets managers spot exceptions quickly and act.

The link to strategy matters most. A dashboard is only useful if it shows the measures that matter for the organisation's critical success factors and objectives. Too many measures hide the signal. The wrong measures steer behaviour the wrong way. The right design gives each user, such as board, divisional manager or supervisor, the level of detail and time frame they need.

Tools also have limits. Poor data quality gives misleading insight. Attractive visuals can hide weak assumptions or mislead through scale choices. Analytics shows correlation, not always cause. Real-time data can encourage short-term reactions. Data privacy, security and bias in models are also concerns. In the exam, give both benefits and limits, tied to the scenario.

How to solve Using Data to Drive Performance Insights questions

Use this method for any written requirement on data, dashboards or analytics in performance reporting.

1Read the requirement and note the verb: explain, evaluate, recommend or advise. Note who the audience is.
2Identify the decision or objective the data must support from the scenario, such as cost control, customer retention or sustainability targets.
3Link to the strategy. Name the critical success factors and the KPIs that measure them.
4Say how data would be turned into information: which analytics type (descriptive, diagnostic, predictive, prescriptive) and which visual format suit the user.
5Apply to the scenario facts. Use the organisation's data sources, users and problems, not a generic list.
6Evaluate limits: data quality, cost, overload, bias, privacy and security, and over-reliance on numbers.
7Give a clear recommendation with priorities, and write it in the format asked, such as a report or briefing note, to earn professional skills marks.

Quickest way: Purpose, user, measure, limit

When to use it: Use when time is short and the requirement asks you to discuss or evaluate a dashboard or analytics tool.

Purpose: state the decision or strategic objective in one line.
User: say who uses it and what level of detail and update frequency they need.
Measure: name two or three KPIs tied to critical success factors, with targets.
Format: pick the visual that fits, such as trend line, traffic light or drill-down.
Limit: give at least two risks, such as poor data quality and information overload, and one safeguard each.

Common mistakes in Using Data to Drive Performance Insights

Treating data and information as the same thing.
The words are used loosely in everyday work.
Fix: Define both early. Show that information is processed, in context and useful for a decision.
Listing features of dashboards without linking them to the scenario.
Students write memorised textbook points.
Fix: Tie each point to the named company, its users and its strategy. Use scenario facts in every paragraph.
Suggesting too many measures for a dashboard.
More data feels more helpful.
Fix: Choose a small set linked to critical success factors. Explain that overload hides the important signals.
Ignoring data quality, bias and security.
Students focus on benefits and forget the evaluation.
Fix: Always include a limits paragraph. Cover accuracy, completeness, timeliness, privacy and governance.
Assuming a pattern in the data proves a cause.
Charts make links look obvious.
Fix: Say that correlation needs further testing before action. Suggest diagnostic analysis or a controlled trial.
Giving no recommendation or professional judgement.
Students stop after describing the tool.
Fix: End with a clear, prioritised recommendation and state any assumptions. This earns professional skills marks.

Worked examples

Example 1

A retail chain has 120 stores. Head office receives long monthly spreadsheets of sales and costs for each store. Regional managers say they cannot see problems until too late. Advise how a dashboard could improve performance reporting. (8 marks)

Show the solution

Point out the problem: the spreadsheets are data, not information. They are detailed, late and not linked to strategy.
Purpose: a dashboard should help managers spot and fix problems early, such as falling sales per store or rising stock losses.
Measures: link to critical success factors. Examples are sales growth against target, gross margin, stock availability, customer satisfaction and staff turnover. Keep it to a small number.
Users: the board needs a summary by region. Regional managers need drill-down to store level. Store managers need daily data on their own store.
Format: use trend lines, traffic lights against target and exception alerts. Update weekly or daily where the data allow, rather than monthly.
Analytics: add diagnostic analysis, such as why a store is underperforming, and predictive analysis, such as expected stock-outs.
Limits: data from different store systems may be inconsistent. Too many alerts will be ignored. Daily figures may prompt short-term reactions. Set data standards and review the measures regularly.
Recommendation: pilot the dashboard in one region, check data quality, then roll it out.

Answer: A dashboard turns the spreadsheet data into timely, role-specific information. It should show a few KPIs tied to critical success factors, with targets, trends and drill-down. It needs good data quality, careful design to avoid overload, and a phased rollout starting with a pilot.

Example 2

A logistics company uses telematics data from its vehicles. The finance director says: 'We collect millions of data points but still make decisions on gut feel.' Explain how analytics could turn this data into performance insight, and give two risks. (8 marks)

Show the solution

Explain data versus information. Raw telematics readings, such as speed, location and fuel use, are data. They need processing to be useful.
Descriptive: report fuel use per kilometre, delivery times and idle time by route and driver against targets.
Diagnostic: investigate why some routes cost more, for example traffic patterns, driver behaviour or vehicle age.
Predictive: forecast vehicle breakdowns from engine data so maintenance can be planned. Predict delivery delays.
Prescriptive: suggest the best routes and schedules to cut cost and improve on-time delivery.
Link to strategy: show the results on a dashboard with KPIs such as cost per delivery, on-time rate and vehicle utilisation.
Risk one: poor or incomplete data, such as faulty sensors, would lead to wrong decisions. Check and validate the data.
Risk two: privacy and morale. Tracking drivers raises data protection issues and may harm trust. Be open about use and follow data protection law.

Answer: Analytics moves the company from data to decisions. It uses descriptive reports, diagnosis of causes, prediction of breakdowns and delays, and prescriptive route optimisation, shown on a KPI dashboard. Two key risks are poor data quality and privacy or staff morale issues, each managed through validation and clear governance.

Exam tips

Start answers by separating data from information. It shows understanding and sets up the rest of the answer.
Always tie dashboards and analytics to the scenario's strategy and users. Generic lists score poorly.
Cover both benefits and limits. Evaluate questions need balance and a conclusion.
Name the type of analytics (descriptive, diagnostic, predictive, prescriptive) when you propose a use. It makes answers precise.
Use the format asked, such as a briefing note, and finish with a clear recommendation to earn professional skills marks.

Practice questions from Technology and information systems

Using Data to Drive Performance Insights: frequently asked questions

What is the difference between data and information in management reporting?

Data are raw facts with no context, such as a list of transactions. Information is data that has been processed and put in context so it helps a decision. A good report turns the first into the second.

What should an APM dashboard include?

It should show a small set of KPIs linked to critical success factors, with targets and trends. It should suit the user, with drill-down where needed. Visual cues such as traffic lights help spot exceptions quickly.

Are the four types of analytics examinable in APM?

They are useful tools for structuring answers. Descriptive, diagnostic, predictive and prescriptive analytics help you explain how data supports decisions. Use them with scenario facts rather than as a definition list.

What are the main risks of using data analytics for performance measurement?

Poor data quality, information overload, bias in models, wrongly assuming cause from correlation, and privacy and security issues. Over-reliance on numbers can also ignore context. Give safeguards such as governance, validation and regular review.