9 Key Performance Indicators and Business Metrics Framework

9.1 Why KPIs and Metrics Matter

What gets measured gets managed is half the truth. The other half is that what gets measured badly gets managed badly.

Every organisation, by the time it reaches any meaningful scale, runs on a set of numbers. Sales targets, customer-satisfaction scores, on-time delivery rates, production yields, cycle times, churn percentages — these are the numbers that determine which projects are funded, which managers are promoted, which suppliers are renewed, and which products are continued.

A coherent system of Key Performance Indicators (KPIs) and business metrics is therefore one of the highest-leverage decisions a firm makes. Done well, it aligns thousands of daily choices with the firm’s strategy. Done badly — too many metrics, the wrong metrics, metrics that can be gamed — it produces motion without progress, and energy spent on the wrong things.

This chapter introduces the vocabulary, the leading frameworks, and the practical disciplines for designing a system of KPIs that actually drives the behaviour the strategy requires.

9.2 Defining KPIs, Metrics, and Measures

The three terms are often used interchangeably, but they are usefully distinct.

A Measure is any quantitative observation about the business — a count, sum, ratio, or duration. Number of orders shipped yesterday is a measure.
A Metric is a measure that is tracked over time and used to monitor or communicate performance. Daily orders shipped, week on week, is a metric.
A Key Performance Indicator (KPI) is a metric that has been deliberately selected because it materially affects the achievement of a strategic or operational objective. KPIs are the small, focused subset of metrics that the leadership team uses to steer the organisation.

A useful test is the one-question test: if the answer to “would I act differently if this number changed by ten per cent?” is no, the number is a metric, not a KPI.

Measures, Metrics, and KPIs at a Glance

Term	Scope	Decision Relevance	Example
Measure	Any single observation	Background information	Number of orders shipped yesterday
Metric	A measure tracked over time	Monitoring and communication	Weekly trend in orders shipped
KPI	A metric tied to a strategic or operational objective	Drives action and accountability	On-time, in-full delivery percentage against target

9.3 Characteristics of a Good KPI

A good KPI is more than just a number. The widely used SMART test, extended for analytics, is a useful checklist:

Specific: It refers unambiguously to one defined behaviour or outcome.
Measurable: It can be quantified, and there is an agreed definition and source.
Achievable: The target can plausibly be reached through actions the team controls.
Relevant: It is genuinely linked to the firm’s strategy or to a defined operational objective.
Time-bound: It carries a frequency and a time horizon.
Actionable: A change in the KPI implies a change in someone’s behaviour.
Aligned: It is consistent with the KPIs above and below it in the cascade — local optimisation does not damage the whole.
Auditable: The source data, calculation, and interpretation can be inspected.

Parmenter argues in his practitioner classic (David Parmenter, 2015) that most organisations track far too many KPIs and that a tightly chosen set of ten or so genuine KPIs at the corporate level is more effective than a sprawling list of dashboards no one acts on. The discipline of selection is at least as important as the discipline of measurement.

9.4 Types of KPIs

KPIs differ on several dimensions, and a balanced system uses several types in combination.

Classification of KPIs

Dimension	Type	Description	Example
Time orientation	Lagging	Measures the result after the period has closed	Quarterly revenue, annual customer churn
	Leading	Measures an early signal that predicts a future result	Sales-pipeline coverage, training hours per agent
Strategic level	Strategic	Reviewed at board and executive level over months and quarters	Return on capital employed, market share
	Tactical	Reviewed by senior managers over weeks and months	Campaign ROI, regional gross margin
	Operational	Reviewed by line managers daily or weekly	Call-handle time, on-time delivery percentage
Nature	Financial	Currency-denominated outcomes	Revenue, cost, EBITDA, ROCE
	Non-financial	Customer, process, people, or quality outcomes	NPS, defects per million, employee engagement
Causation	Outcome	Measures what happened	Customer churn rate
	Driver	Measures what causes the outcome	First-call-resolution rate
Form	Quantitative	Numerical measure	Average handling time in seconds
	Qualitative	Structured rating or category	Customer-effort score, audit grade

A common failure mode is to manage exclusively with lagging financial outcome KPIs. By the time these move, it is usually too late to influence the period in question. A balanced system pairs each strategic outcome KPI with the leading driver KPIs that are believed to cause it.

9.5 Major KPI Frameworks

9.5.1 Balanced Scorecard

flowchart TD
    V["Vision and<br>Strategy"]
    V --> F["Financial<br>How do we look<br>to shareholders?"]
    V --> C["Customer<br>How do customers<br>see us?"]
    V --> P["Internal Process<br>What must we<br>excel at?"]
    V --> L["Learning and Growth<br>Can we continue<br>to improve and learn?"]
    style V fill:#e3f2fd,stroke:#1976D2
    style F fill:#e8f5e9,stroke:#388E3C
    style C fill:#fff8e1,stroke:#F9A825
    style P fill:#fff3e0,stroke:#EF6C00
    style L fill:#fce4ec,stroke:#AD1457

The Balanced Scorecard, introduced by Robert S. Kaplan & David P. Norton (1992) in Harvard Business Review, was the first widely adopted framework to insist that financial KPIs alone do not measure organisational performance. The Scorecard frames performance across four perspectives:

Financial: How do we look to shareholders?
Customer: How do customers see us?
Internal Process: What must we excel at?
Learning and Growth: Can we continue to improve and learn?

Each perspective carries its own set of objectives, measures, and targets. The four perspectives are linked by causal hypotheses: improvements in learning and growth drive better internal processes, which improve customer outcomes, which produce financial results.

The Balanced Scorecard remains, three decades later, the most widely adopted strategic-performance-management framework in industry, particularly in larger organisations and the public sector.

9.5.2 Objectives and Key Results (OKR)

OKR (Objectives and Key Results) is the framework popularised by Andy Grove at Intel and adopted at scale by Google. Each cycle (typically a quarter) the team or company sets:

An Objective — a qualitative, ambitious, time-bounded statement of intent.
Three to five Key Results — quantitative outcomes that, if achieved, would mean the Objective has been delivered.

OKRs are deliberately stretched: scoring 0.6 to 0.7 on a 0-to-1 scale is considered successful. They are usually transparent across the organisation and refreshed every quarter. OKRs are best suited to fast-moving organisations and to ambition-setting, less so to stable operational performance management.

9.5.3 North Star Metric

A North Star Metric is a single high-level metric chosen to represent the long-term value the firm is creating for its customers. It is a useful concept for product-led companies that wish to anchor cross-functional decisions around one number.

Spotify often cites Time Spent Listening.
Airbnb has used Nights Booked.
LinkedIn has used Active Connections.

The North Star is not a substitute for a Balanced Scorecard or an OKR system; it is a unifying focus that sits above them. The risk is that any single metric, used alone, will be gamed. A North Star must be paired with guardrail metrics that flag damage caused by chasing it.

9.5.4 AARRR Pirate Metrics

AARRR, attributed to Dave McClure, is a customer-funnel framework widely used in product and growth analytics:

Acquisition — how do users find us?
Activation — do they have a good first experience?
Retention — do they come back?
Referral — do they tell others?
Revenue — do they monetise?

Each stage of the funnel carries its own KPIs. The framework is particularly useful in subscription, freemium, and digital-product businesses, where the customer journey can be observed digitally and modelled end to end.

9.5.5 KPI Tree (Driver Tree)

A KPI Tree decomposes a top-line outcome KPI into its mathematical or logical drivers, then decomposes each of those, and so on, until a leaf-level operational KPI is reached. Revenue breaks into price × volume; volume breaks into traffic × conversion × repeat rate; traffic breaks into source-by-source acquisition; and each leaf becomes the responsibility of a specific team. The tree is particularly useful for diagnosing the cause of a movement in the top-line metric and for assigning accountability across functions.

9.5.6 Comparison

Comparison of KPI Frameworks

Framework	Best For	Cadence	Risk
Balanced Scorecard	Strategic performance management at enterprise scale	Quarterly to annual	Becomes static if not actively refreshed
OKR	Ambitious goal-setting in fast-moving organisations	Quarterly	Confusion between OKRs and operational KPIs
North Star Metric	Product-led firms unifying cross-functional focus	Continuous, reviewed quarterly	Can be gamed without guardrails
AARRR	Customer-funnel analytics in digital products	Continuous	Optimises the funnel, not the strategy
KPI Tree	Diagnostic and accountability mapping of an outcome	Continuous	Can become unwieldy if too detailed

The frameworks are not mutually exclusive. Mature firms often use a Balanced Scorecard at the corporate level, OKRs for cross-functional initiatives, a North Star Metric inside each product line, and KPI trees to diagnose outcome-metric movements. The choice is about coverage and cadence, not allegiance.

9.6 Designing a KPI System

flowchart LR
    A["1. Anchor in<br>strategy"] --> B["2. Identify<br>strategic outcomes"]
    B --> C["3. Map drivers<br>and leading<br>indicators"]
    C --> D["4. Define each<br>KPI precisely"]
    D --> E["5. Cascade to<br>functions and<br>teams"]
    E --> F["6. Set targets<br>and thresholds"]
    F --> G["7. Build dashboards<br>and review<br>cadence"]
    G --> H["8. Audit, adapt,<br>and prune"]
    H -.-> A
    style A fill:#fce4ec,stroke:#AD1457
    style B fill:#fff3e0,stroke:#EF6C00
    style C fill:#fff8e1,stroke:#F9A825
    style D fill:#e3f2fd,stroke:#1976D2
    style E fill:#ede7f6,stroke:#4527A0
    style F fill:#e8f5e9,stroke:#388E3C
    style G fill:#f3e5f5,stroke:#6A1B9A
    style H fill:#eceff1,stroke:#455A64

A pragmatic eight-step process to build a KPI system that actually shapes behaviour:

Anchor in strategy: KPIs measure progress against objectives. Without explicit objectives, KPIs measure activity, not progress.
Identify strategic outcomes: Translate each strategic objective into the small set of outcome KPIs that would prove the objective is being achieved.
Map drivers and leading indicators: For each outcome KPI, identify the operational drivers believed to cause it. These become leading-indicator KPIs.
Define each KPI precisely: Name, definition, formula, source, owner, frequency, target, threshold for alert. Document them in a KPI catalogue.
Cascade to functions and teams: Each strategic KPI translates into related departmental and team KPIs that are consistent with it. Avoid contradictions in which a team can hit its KPI while damaging the level above.
Set targets and thresholds: Targets express ambition; thresholds express acceptable variation. Both should be evidence-based, not aspirational guesses.
Build dashboards and a review cadence: Different audiences need different views. The board sees a quarter; an operating manager sees a week. The cadence determines the dashboard.
Audit, adapt, and prune: Review the system at least annually. Drop KPIs that no longer drive action. Add KPIs only when a clear gap is identified.

9.6.1 KPI Cascading

KPI cascading translates a corporate-level KPI into the departmental, team, and individual KPIs that support it. The principle is line of sight: every employee should be able to trace how their daily work connects to a corporate objective.

A simple example: a bank’s corporate KPI of Cost-to-Income Ratio cascades to Operations Cost per Transaction, which cascades to a back-office team’s Errors per Thousand Transactions and Average Handling Time. Improvements at the leaf level produce improvements at the corporate level — provided the cascade is mathematically and behaviourally consistent.

9.7 Common Pitfalls

Vanity Metrics: Numbers that look impressive — total registered users, total social-media followers — but have no relationship to the firm’s outcomes. They make presentations longer and decisions worse.
Goodhart’s Law: When a measure becomes a target, it ceases to be a good measure. Any KPI tied to incentives will eventually be gamed unless guardrail metrics are in place.
Too Many KPIs: Long dashboards crowd out attention. Parmenter recommends roughly ten true KPIs at the corporate level; many firms operate with two to three times that, to little benefit.
Lagging-Only Dashboards: A dashboard composed only of lagging financial outcomes tells leadership what already happened. Without leading indicators, there is nothing to act on.
Misaligned Cascading: Local KPIs that contradict corporate KPIs. A call-centre that hits its Average Handling Time target by cutting calls short will worsen the corporate First-Call Resolution and NPS KPIs.
Definition Drift: The same KPI name carrying different definitions across functions. Active customer in marketing is not active customer in finance.
Unactionable KPIs: KPIs whose movement does not imply any action by anyone. They are noise.
Set-and-Forget: KPIs designed once and never revisited. The strategy evolves; the KPI set must evolve with it.
Dashboards Without Decisions: Dashboards that are reviewed but never produce a change of behaviour. The review meeting is theatre.
Neglecting Qualitative Measures: Over-quantifying the business until valuable qualitative signals — employee morale, customer effort, brand health — disappear from the leadership conversation.

9.8 Illustrative Cases

The following short cases illustrate how KPIs and metric frameworks play out in practice. They are based on the kinds of programmes commonly seen in industry; the framing is the author’s.

A Retail Bank’s Balanced Scorecard

A mid-size retail bank deploys a Balanced Scorecard at the corporate level. The Financial perspective tracks return on equity, cost-to-income ratio, and capital adequacy. The Customer perspective tracks net promoter score, complaint volume, and primary-bank-relationship penetration. The Internal Process perspective tracks loan-application turnaround, fraud-detection accuracy, and digital-channel availability. The Learning and Growth perspective tracks training hours per employee, employee engagement, and digital-skills coverage. Each KPI cascades into branch, regional, and product-line targets, and the four perspectives are reviewed quarterly by the executive committee.

A Software-as-a-Service Firm’s North Star and Pirate Metrics

A SaaS firm sets Weekly Active Customers Performing Core Action as its North Star Metric and runs every product team against the AARRR funnel. Acquisition tracks site visits, sign-up rate, and cost per sign-up. Activation tracks first-week feature usage. Retention tracks 30-, 60-, and 90-day cohort retention. Referral tracks invitations sent and accepted. Revenue tracks paid conversion, ARPU, and gross retention. Guardrails — uptime, support-ticket volume, and customer-effort score — protect against North Star gaming.

A Manufacturing Plant’s KPI Tree

A manufacturing plant decomposes its top-line Cost per Unit Produced into a KPI tree. Direct-materials cost decomposes into yield, scrap rate, and supplier price variance. Direct-labour cost decomposes into output per shift and overtime hours. Overhead decomposes into energy use per unit and maintenance cost per machine-hour. When the top-line metric drifts adversely, the plant traces the movement down the tree to the specific leaf KPI responsible, and assigns the action to the team that owns it.

An Indian E-Commerce Firm’s OKRs

An Indian e-commerce firm runs on a quarterly OKR cycle. A typical objective is Make first-time-customer experience the best in the category in tier-2 and tier-3 cities. Key results might be: increase first-purchase satisfaction in target cities from 78 to 85, reduce first-purchase return rate from 9 per cent to 6 per cent, and cut first-mile pick-up time in target cities from 36 hours to 18 hours. The OKRs are visible across the firm; teams self-score at quarter end, and a 0.7 average is treated as a success.

Summary

Concept	Description
Foundations
Why KPIs Matter	A coherent KPI system aligns thousands of daily choices with strategy; a poor one produces motion without progress
Measure	Any single quantitative observation about the business
Metric	A measure tracked over time and used to monitor or communicate performance
Key Performance Indicator	A metric deliberately selected because it materially affects a strategic or operational objective
Characteristics of a Good KPI
Specific	Refers unambiguously to one defined behaviour or outcome
Measurable	Can be quantified with an agreed definition and source
Achievable	Target can plausibly be reached through actions the team controls
Relevant	Genuinely linked to strategy or to a defined operational objective
Time-Bound	Carries a frequency and a time horizon
Actionable	A change in the KPI implies a change in someone's behaviour
Aligned	Consistent with the KPIs above and below it in the cascade
Auditable	Source data, calculation, and interpretation can be inspected
Types of KPIs
Lagging KPI	Measures the result after the period has closed
Leading KPI	Measures an early signal that predicts a future result
Strategic KPI	Reviewed at board and executive level over months and quarters
Tactical KPI	Reviewed by senior managers over weeks and months
Operational KPI	Reviewed by line managers daily or weekly
Financial KPI	Currency-denominated outcomes such as revenue, cost, EBITDA
Non-Financial KPI	Customer, process, people, or quality outcomes
Outcome KPI	Measures what happened
Driver KPI	Measures what causes the outcome
Quantitative KPI	Numerical measure such as average handling time
Qualitative KPI	Structured rating or category such as customer-effort score
Balanced Scorecard
Balanced Scorecard	Kaplan and Norton framework spanning four perspectives linked by causal hypotheses
Financial Perspective	How do we look to shareholders
Customer Perspective	How do customers see us
Internal Process Perspective	What must we excel at
Learning and Growth Perspective	Can we continue to improve and learn
OKR and North Star
Objectives and Key Results (OKR)	Cycle of qualitative ambitions paired with three to five quantitative key results
North Star Metric	Single high-level metric representing the long-term value created for customers
Guardrail Metric	Companion metric that flags damage caused by chasing the North Star
AARRR Pirate Metrics
AARRR Pirate Metrics	Customer-funnel framework for product and growth analytics
Acquisition	How do users find us
Activation	Do they have a good first experience
Retention	Do they come back
Referral	Do they tell others
Revenue	Do they monetise
KPI Tree
KPI Tree	Decomposition of a top-line outcome into mathematical or logical drivers down to leaf KPIs
Designing a KPI System
Anchor in Strategy	KPIs measure progress against objectives; without objectives they measure activity
Identify Strategic Outcomes	Translate each strategic objective into the small set of outcome KPIs that would prove it is being achieved
Map Drivers and Leading Indicators	For each outcome KPI, identify the operational drivers believed to cause it
Define KPI Precisely	Document name, formula, source, owner, frequency, target, and threshold in a KPI catalogue
Cascade to Teams	Each strategic KPI translates into consistent departmental and team KPIs
Set Targets and Thresholds	Targets express ambition, thresholds express acceptable variation; both should be evidence-based
Build Dashboards and Cadence	Different audiences need different views; cadence determines the dashboard
Audit and Prune	Drop KPIs that no longer drive action; add only when a clear gap is identified
KPI Cascading
Line of Sight	Every employee can trace how their daily work connects to a corporate objective
Common Pitfalls
Vanity Metrics	Pitfall of impressive-looking numbers that have no relationship to outcomes
Goodhart's Law	When a measure becomes a target it ceases to be a good measure unless guarded
Too Many KPIs	Pitfall of long dashboards that crowd out attention; ten true corporate KPIs is usually plenty
Lagging-Only Dashboards	Pitfall of dashboards composed only of lagging financial outcomes with nothing to act on
Misaligned Cascading	Pitfall of local KPIs that contradict corporate KPIs and damage the level above
Definition Drift	Pitfall of the same KPI name carrying different definitions across functions
Unactionable KPIs	Pitfall of KPIs whose movement does not imply any action by anyone
Set-and-Forget	Pitfall of KPIs designed once and never revisited as the strategy evolves
Dashboards Without Decisions	Pitfall of dashboards reviewed but never producing a change in behaviour
Neglecting Qualitative Measures	Pitfall of over-quantifying until valuable qualitative signals disappear from the leadership conversation