Alternative Performance Indicators

CIMA-P2 · Chapter 8

Alternative Performance Indicators

1 Introduction

Chapter 7 measured performance in money. That is indispensable and it is not enough, for one reason above all others: financial measures report the past, and almost every decision a manager takes is about the future.

It is easy to raise this year's profit in ways that damage next year's — buy cheaper materials, cut the training budget, defer maintenance, pay staff less. The costs fall immediately and the consequences arrive later, in lost customers and lost staff. A measurement system built only on financial results will report the first effect as success and the second as bad luck.

This chapter covers non-financial performance indicators and how to construct them; the difference between a lead and a lag indicator; Kaplan and Norton's balanced scorecard and the cause-and-effect logic that holds it together; and how a small set of measures is reported to a responsibility-centre manager on a dashboard.

2 Non-financial performance indicators

This recording covers sections 2 and 4 and works Example 1. Its central point is the one the chapter opens with: measure quality and satisfaction because they drive the financial results that follow.

Notice one. The tape does not cover the cause-and-effect linkage between the four perspectives (section 4.3), how a scorecard is built (4.4), or the problems of using one in practice (4.6). It lists the four perspectives and their measures.

Notice two. Sections 3 and 5 — critical success factors, lead and lag indicators, and dashboards — are notes-only.

One point from the tape is worth carrying: a customer survey is not itself a performance measure. Running the survey is how the data is collected; the measure is the average grade, or the percentage rating the service good or excellent. Section 2.3.

YouTube video

2.1 What they are

A non-financial performance indicator measures an aspect of performance that is not expressed in money and contains no monetary value in its calculation. That is the whole of the definition; everything else about them follows from the fact that they can be chosen freely, because there are no rules about what to measure.

They let a company measure anything it believes is critical to its success — and the corollary is that the choice of measures is itself a management decision, and a revealing one.

2.2 What is commonly measured

Area	Example indicators
Quality	Defect rate · units reworked as a % of output · warranty claims per 1,000 units sold · first-time-right %
Customer satisfaction and loyalty	Complaints per week · % of goods returned within 14 days · % of customers rating the service good or excellent · repeat-purchase rate · net promoter score
Delivery and operational effectiveness	% of orders delivered on time and in full · order-to-delivery lead time · cycle time · capacity utilisation · % of scheduled services completed
Staff	Staff turnover % · absenteeism days per employee · training days or courses completed per employee · % of staff reporting fair treatment in the annual survey
Innovation	Number of new products launched · % of revenue from products launched in the last three years · time to market
Environment and society	Tonnes of waste to landfill per period · % of packaging recycled · energy used per unit of output · % of suppliers audited

2.3 Turning a quality into a measure

The subjective nature of many non-financial qualities makes them harder to measure than money, and the discipline that solves it is simple: decide in advance how the quality will be measured, and put a number to it. It is no use one manager saying quality feels better this year and another saying it feels worse.

A customer survey is not a performance measure. Running the survey is how the data is collected; the measure is what comes out of it — the average grade, or the percentage of respondents rating the service good or excellent. Reporting "we sent out 4,000 surveys" measures the measuring, and it is a common way of losing marks.

A useful test of any proposed indicator is whether it would move if performance changed, and whether the manager being measured could move it. Both come straight from chapter 5: an indicator outside a manager's control is as unfair non-financially as an apportioned head-office cost is financially.

2.4 Advantages

They are lead indicators of future performance. Quality, satisfaction and staff morale change before revenue and profit do, so a fall in them is an early warning that a financial report cannot give.
They can be tied to long-term objectives in a way that a period's profit cannot.
Improving them tends to drive the financial results — better quality and service lead to more sales and lower costs of failure.
They are understood at every level. A production supervisor can act on a defect rate; a return on capital employed means much less on the factory floor.
They are harder to manipulate than profit, which depends on accounting estimates.

2.5 Drawbacks

Time and cost. Somebody has to collect and record the data, and the cost of measuring can exceed the benefit of knowing.
Difficulty of obtaining reliable information, particularly for anything subjective.
No agreed formulae. Unlike a ratio, there is no standard definition, so results are hard to compare between periods, divisions or companies, and they come in different units — times, percentages, days, counts.
The causal link may be weak. Improving an indicator does not always improve the business; factors outside the company's control intervene.
Too many measures. A manager given thirty indicators pays attention to none of them. This is the single commonest failure in practice.
Measure fixation. What is measured gets managed, including at the expense of what is not — the same dysfunctional behaviour as chapter 5, section 5.5, in a different currency.

Suggest some non-financial performance measurements that would be suitable for a large supermarket chain to assess its performance in the following areas.

(a) Environmental performance
(b) Customer service
(c) Employee satisfaction

There are no standard measures; what is being tested is whether the suggestions are genuinely measurable, genuinely relevant to a supermarket, and genuinely non-financial. Many answers are possible.

Area	Possible measures
(a) Environmental	Kilograms of food waste per store per week · tonnes of plastic packaging recycled per quarter as a % of packaging used · energy consumed per square metre of selling space · % of own-brand suppliers audited against the environmental standard
(b) Customer service	Average number of customers queuing before an additional till is opened · average queuing time at peak · % of customers rating service good or excellent (from the survey, or from the smiley-face terminals at the exit) · % of listed lines available on shelf · complaints per 1,000 transactions
(c) Employee satisfaction	Staff turnover %, against last year and against target · absenteeism days per employee · % of staff agreeing they are treated fairly in the annual survey · number of staff attending employee participation days against target · internal promotion rate

Two points of technique. First, each measure names a comparator — against target, against last year, per 1,000 transactions — because a bare count means nothing (chapter 7, section 2). Second, the survey and the exit terminal are data collection, not measures; the measure is the percentage that comes out of them.

3 From objectives to indicators

3.1 Critical success factors and key performance indicators

Measures should not be invented one at a time. The chain runs:

Objective — what the organisation is trying to achieve. Win commuter traffic from road.
Critical success factor — the few things that must go right if the objective is to be achieved. Punctuality; comfort; the price of a season ticket.
Key performance indicator — the measure that tells you whether the critical success factor is being delivered. Percentage of trains arriving within three minutes of schedule.

Working in that order keeps the number of indicators small and keeps every one of them attached to something the organisation actually cares about. Working in the opposite direction — measuring what is easy to measure and then looking for a reason — is how organisations end up with thirty indicators and no strategy.

3.2 Lead and lag indicators

	Lag indicator	Lead indicator
What it tells you	What has already happened.	What is likely to happen, because it measures a driver of the result.
Examples	Revenue · operating margin · return on capital employed · market share achieved	Customer satisfaction score · order book · staff turnover · defect rate · training days completed
Strength	Objective, and it is the outcome that actually matters.	It moves early, so there is still time to act.
Weakness	By the time it moves, the period is over.	The link to the outcome is a hypothesis and may be wrong.

The pairing

Every financial measure in chapter 7 is a lag indicator. A performance report needs both: lag indicators to confirm the result, and lead indicators to give a manager time to change it.

3.3 What makes a good indicator

Relevant to an objective or critical success factor, and not merely available.
Measurable on an agreed definition, so that this period and last period mean the same thing.
Controllable by the person it is reported to (chapter 5, section 4.1).
Timely enough to act on.
Few. A manageable set beats a complete one.
Balanced — financial and non-financial, lead and lag, short and long term. Which is exactly what the balanced scorecard sets out to guarantee.

4 Kaplan and Norton's balanced scorecard

4.1 What it is for

The balanced scorecard, developed by Kaplan and Norton in 1992, views a business from four perspectives and requires objectives, and measures for those objectives, under each. Its purpose is to give an overall health check that no single family of measures can give: internal and external, financial and non-financial, lead and lag, short term and long term.

The name is often misread. It is not a scorecard full of scores — it is a card on which the measures are balanced against one another so that no one of them can be improved quietly at the expense of the rest.

4.2 The four perspectives

Perspective	The question it asks	Typical objectives	Possible measures
Customer	What do existing and potential customers value from us?	Win new customers; retain existing ones; deliver reliably	% of sales from new customers · % of on-time deliveries · orders as a % of enquiries · customer survey score · returns and complaints
Internal business	What must we excel at to achieve our customer and financial objectives?	Produce efficiently; eliminate non-value-adding activity; shorten the cycle	Unit cost · process and cycle time · capacity utilisation · defect and rework rate · value analysis findings
Learning and growth	How can we continue to improve and create future value?	Innovate; develop and keep capable people	Number of new products introduced · time to market · % of revenue from recent products · training days per employee · staff turnover
Financial	How do we create value for our shareholders?	Grow profitably; use capital well	Profitability · sales growth · return on investment · cash flow and liquidity — the ratios of chapter 7

4.3 Cause and effect

The four perspectives are not a list. They are a chain, and this is the part of the scorecard most often left out:

Read it upwards and it is a hypothesis about how the business works: capable, motivated people improve the processes; better processes deliver what customers value; satisfied customers buy more and stay; that produces the financial result. Read it downwards and it is a plan: if this is the financial result we want, these are the customers we must satisfy, these are the processes we must excel at, and these are the capabilities we must build. A diagram setting the chain out explicitly is called a strategy map.

Two consequences follow. First, the learning-and-growth measures are the earliest lead indicators in the whole system — the ones that move first and the ones that are cut first when a manager is under short-term pressure. Second, if a scorecard's four sets of measures cannot be joined into a chain like this one, it is not a scorecard; it is four unrelated reports on one page.

4.4 Building one

For each perspective, four things are needed, and a scorecard missing any of them will not change behaviour:

Objectives — what we are trying to achieve in that perspective.
Measures — how we will know.
Targets — the level of the measure we are aiming at, and by when.
Initiatives — what will actually be done to get there, and who owns it.

Keep it to three or four measures per perspective. Kaplan and Norton's own guidance is that a scorecard with more than about twenty measures has stopped being a focusing device and become a report.

Ravenna Rail operates a regional passenger rail franchise. Its strategy is to win commuter traffic away from road travel by being the most punctual and most comfortable operator in its region.

(a) For each of the four perspectives of the balanced scorecard, suggest one objective and one measure that would be appropriate for Ravenna Rail.

(b) Explain the cause-and-effect chain that links the four objectives you have suggested.

(a) Objectives and measures

Perspective	Objective	Measure
Learning and growth	Recruit, train and retain drivers and on-board staff	% of operational staff completing the annual safety and customer-service programme (with staff turnover % as a second measure)
Internal business	Run the timetable reliably	% of trains arriving within three minutes of the scheduled time (with services cancelled as a % of services scheduled)
Customer	Increase commuter satisfaction and retention	% of passengers rating the journey good or excellent, and the season-ticket renewal rate
Financial	Grow revenue and margin from commuter traffic	Growth in season-ticket revenue, and revenue per train-kilometre

Any measures that are genuinely measurable, genuinely within Ravenna Rail's influence and genuinely aligned to its stated strategy would earn the marks. Note that the strategy is specific — punctuality and comfort for commuters — so a generic measure such as "passenger numbers" is weaker than one that tracks the thing the strategy turns on.

(b) The cause-and-effect chain

Trained drivers and on-board staff who stay with the company operate the service better and maintain the rolling stock better → the timetable is kept and cancellations fall → commuters arrive on time, are more satisfied, renew their season tickets and recommend the service → revenue from commuter traffic grows and the margin improves.

Read upwards, that is the company's hypothesis about how its strategy will work. Read downwards, it is the plan. And it shows why the learning-and-growth measures matter most for the long run: they are at the start of the chain, they move first, and they are the first thing cut when a manager is under pressure to hit this period's financial number.

4.5 Advantages

It forces attention onto non-financial and long-term measures alongside the financial ones, so short-termism becomes visible.
It links measures to strategy rather than to what is easy to count.
The cause-and-effect chain makes the organisation's assumptions explicit, and therefore testable.
It communicates the strategy: a manager can see where their own objective sits in the chain.

4.6 Problems in practice

Too many measures. The four perspectives are an invitation to add, and a scorecard with forty measures focuses on nothing.
Conflict between perspectives. Improving customer service usually costs money in the short run. The scorecard exposes the trade-off; it does not resolve it, and somebody still has to decide.
No weighting. If three perspectives improve and one worsens, the scorecard does not say whether that is a good year, and rewarding managers against it therefore needs a judgement the scorecard cannot supply.
The causal links are assumed, not proved, and they may be wrong, or right with a much longer lag than the reporting period.
Data cost. Several of the measures do not exist in the accounting system and have to be collected specially.
It can become a reporting exercise. Without targets and initiatives (section 4.4) it changes nothing; it merely describes.

5 Reporting the measures: dashboards

A scorecard is a design. What a responsibility-centre manager actually sees each week is a dashboard: a small set of measures, presented visually, with the ones outside tolerance highlighted. Chapter 5, section 6 dealt with the analytics behind it; this section deals with what belongs on it.

5.1 What belongs on a divisional dashboard

A few measures — six to ten, not thirty.
A mix of lead and lag, so the manager sees both the result and its drivers.
Only controllable items above the line the manager is judged on (chapter 5, section 4.3).
Each against a comparator — a flexed budget, a target, or the same period last year — never as a bare number.
A tolerance fixed in advance, so that attention goes to genuine exceptions and not to noise.
Trend, not point: twelve periods of a measure say far more than one.

5.2 Two design failures to recognise

Both are examinable and both come from measuring the wrong thing rather than from measuring badly:

Measuring the activity instead of the outcome — surveys sent rather than satisfaction scored, training courses run rather than competence achieved, calls answered rather than problems solved.
Measuring one half of a trade-off — speed without quality, cost without service, volume without margin. Any measure that can be improved by damaging something else must be reported paired with the something else.

The manager of a customer-service call centre, which is a cost centre, receives a monthly dashboard showing four measures:

Calls answered per agent per day (higher is better)
Average call handling time in minutes (lower is better)
Total divisional cost, including apportioned head-office costs, compared with the same month last year
Number of customer satisfaction surveys sent out

Identify four weaknesses in this set of measures and suggest an improvement for each.

Weakness	Why it matters	Improvement
Measures 1 and 2 reward speed at the expense of resolution	An agent hits both targets by ending calls quickly — and the customer rings back, so the same problem is counted twice as productivity. This is measuring one half of a trade-off.	Report first-contact resolution % alongside them, and read the three together. Add a call-quality score sampled by a supervisor.
Measure 3 includes apportioned head-office costs	Those costs are attributable to the division but are not controllable by its manager, so they do not belong in the figure the manager is appraised on (chapter 5, section 4.3).	Show controllable cost as the headline, with the apportioned costs below the line.
Measure 3 compares with last year rather than with a flexed budget	Call volumes vary. A month with 20% more calls will cost more, and the comparison says nothing about how well the centre was run (chapter 5, section 5.2).	Compare with a budget flexed to the number of calls actually handled, and show cost per call as well as total cost.
Measure 4 counts the data collection, not the performance	Sending surveys is an activity. The measure is what comes back.	Replace it with the average satisfaction score, or the % of respondents rating the service good or excellent, and report the response rate separately as a check on reliability.

A fifth observation would also earn credit: there is no lead indicator in the set. Every measure reports what has already happened. Agent turnover, absence, training completed and the queue-abandonment rate would all move before the service level does, and would give the manager time to act.

And note what is not a weakness: the call centre is a cost centre, so the absence of a revenue or profit measure is correct. Measuring it on revenue would be a worse error than any of the four above.

6 Bringing the measures together

The three chapters of this section fit together as one system, and an examination question will usually want them treated that way:

	Supplies
Chapter 5	Whose performance is being measured — the four responsibility centres, what each manager controls, the budget as the yardstick, and the behavioural consequences of measuring anybody against anything.
Chapter 7	The financial half of the measurement — the ratios, benchmarking, and the limits of what money can tell you.
Chapter 8	The non-financial half, and the framework that keeps the two in balance.

The single sentence that connects them: a performance measure is an instruction, so choose measures whose optimum for the manager is also the optimum for the organisation, report only what the manager controls, and never let one measure be improvable at the expense of something you are not measuring.

Non-financial indicators measure what is not expressed in money, and they matter because they lead the financial results rather than following them. Build them in order — objective, then critical success factor, then indicator — and pair every lag measure with a lead one. A survey is not a measure; the score is. Kaplan and Norton's balanced scorecard sets objectives and measures under four perspectives — customer, internal business, learning and growth, financial — and its point is the cause-and-effect chain that runs up from capable people, through excellent processes and satisfied customers, to the financial result; a scorecard whose four sets of measures cannot be joined into that chain is four reports on one page. Each perspective needs objectives, measures, targets and initiatives, and about twenty measures in total is the practical limit. On a dashboard, report few measures, only controllable ones, each against a comparator, with a tolerance set in advance — and never measure one half of a trade-off.

7 Test your knowledge

Two quick checks before you move on: work through the flashcards to fix this chapter’s key terms and definitions, then sit the objective questions for exam-style practice. Both mark themselves and explain the answers as you go.

Answer the questions one at a time. Your progress is saved so you can leave and come back.

Open chapter practice