Files
SecondBrain/10 Knowledge/CTO Academy/Data/KPI.md
T

2.8 KiB
Raw Blame History

MD file: KPI.md
Estimated reading time: 1113 minutes

KPI

Definition

A KPI, or Key Performance Indicator, is a metric selected because it indicates progress toward an important business objective.

Every KPI is a metric. Not every metric is a KPI.

Simple Mental Model

Metric = something we can measure. KPI = a measurement important enough to tell us whether we are succeeding.

How It Works

A strong KPI connects:

Business Objective
      ↓
Desired Outcome
      ↓
KPI
      ↓
Target
      ↓
Actions / Decisions

Example

If the strategic objective is streaming growth, executive KPIs might include:

  • Monthly Active Viewers
  • Total Watch Hours
  • Viewer Retention
  • Streaming Revenue

Other metrics can still matter operationally without being KPIs.

Leading vs Lagging Indicators

Lagging indicators measure outcomes after they happen, such as revenue or churn.

Leading indicators may provide earlier signals, such as weekly engagement or trial activation.

How It Fits Into the Bigger Picture

[[Fact Table]]
Raw measures
      ↓
[[Semantic Layer]]
Governed metrics
      ↓
[[Business Glossary]]
Agreed meaning
      ↓
KPI
      ↓
Power BI
      ↓
Management decision

My Company / Real-World Context

Potential KPIs could include:

  • Monthly Active Viewers
  • Total Watch Hours
  • Viewer Retention
  • Video Start Success Rate
  • Advertising Revenue
  • WCAG Compliance Rate
  • SLA Compliance

If everything is "key," nothing is key.

CTO Perspective

Bad executive dashboards display whatever is easy to measure.

Good dashboards display what is important to manage.

A CTO should challenge output metrics that do not demonstrate business value.

Questions to Ask

  • What business objective does this KPI represent?
  • Why is it "key"?
  • Who owns it?
  • What is the target?
  • What decision changes when this KPI changes?
  • Is it leading or lagging?
  • Can we trust the underlying Data Quality?
  • Are we measuring output or outcome?

Meeting Scenario

Situation: A department proposes 18 more metrics for the executive dashboard.

Possible response:

"These may all be useful metrics, but which of them directly indicate whether we're achieving our strategic objectives, and what decisions would leadership make differently based on them?"

Key Takeaways

  • Every KPI is a metric, but not every metric is a KPI.
  • KPIs should connect directly to business objectives.
  • Leading indicators provide earlier signals; lagging indicators measure outcomes.
  • Executive dashboards should focus attention.
  • CTOs should distinguish output from outcome.
  • Metric
  • Measure
  • Semantic Layer
  • Business Glossary
  • Data Governance
  • Data Quality
  • Data Lineage
  • Leading Indicator
  • Lagging Indicator