KPIs and Metrics: How to Define and Track What Matters

What Are KPIs and Why Do They Matter?

Key Performance Indicators (KPIs) are quantifiable measurements that evaluate how effectively an organization, team, or individual is achieving their objectives. The "key" in KPI is important — not every metric is a KPI. KPIs are the small set of metrics that matter most for assessing progress toward strategic goals. Choosing the right KPIs is one of the most impactful things a data analyst can do for an organization.

Without clearly defined KPIs, teams optimize for the wrong things, measure activity rather than outcomes, and struggle to align around shared goals. With the right KPIs, everyone from the CEO to the frontline team understands what success looks like and can see whether they're moving toward it.

The Difference Between Metrics and KPIs

All KPIs are metrics, but not all metrics are KPIs. A metric is any quantifiable measure — page views, number of support tickets created, average session duration, lines of code written. A KPI is a metric that is directly tied to a strategic objective and used to evaluate performance against that objective.

Monthly Recurring Revenue (MRR) is a KPI for a SaaS business because it directly reflects the health and growth of the subscription model. The number of support tickets created is a metric — it's useful context, but it doesn't directly tell you whether the support team is performing well or whether customers are succeeding. Customer satisfaction score (CSAT) and time to resolution are more likely to be KPIs for a support team.

Vanity Metrics vs. Actionable Metrics

Vanity metrics are numbers that look impressive but don't meaningfully connect to business outcomes. Total registered users, cumulative downloads, and social media follower counts are classic vanity metrics — they tend to go up over time regardless of whether the product is actually succeeding, and they don't tell you what to do differently.

Actionable metrics are directly connected to decisions and behaviors. Daily Active Users (DAU), week-over-week retention, conversion rate from trial to paid, and net revenue retention all tell you something meaningful about product health and point toward specific actions when they move in the wrong direction. The test: if the metric changed significantly, would it change what your team does? If the answer is no, it's probably a vanity metric.

Frameworks for Defining Good KPIs

The SMART framework is the most widely used guide for defining KPIs: Specific (clearly defined and unambiguous), Measurable (quantifiable with available data), Achievable (realistic given constraints), Relevant (aligned with strategic goals), and Time-bound (tracked over a defined period with a clear target). A KPI that fails any of these criteria will create confusion rather than clarity.

The North Star Metric framework identifies a single metric that best captures the core value delivered to customers and predicts long-term business growth. For Airbnb, it's nights booked. For Spotify, it's time spent listening. For Slack, it's messages sent. Aligning an entire organization around one primary metric focuses effort and prevents teams from optimizing for local metrics that don't improve the overall business.

The OKR (Objectives and Key Results) framework pairs qualitative objectives with 2–5 measurable key results that define what success looks like. Key results are typically binary (achieved or not) or measured as progress toward a specific target. OKRs cascade from company level to team level, ensuring alignment across the organization.

Leading vs. Lagging Indicators

Lagging indicators measure outcomes that have already occurred — revenue, customer churn, annual recurring revenue, net promoter score. They tell you how you've done but don't give you early warning of problems. Leading indicators are metrics that predict future outcomes — sales pipeline size, trial signups, feature adoption rate, support ticket volume. They give you time to act before a problem shows up in the lagging metrics.

A balanced KPI framework includes both types. Lagging indicators confirm results; leading indicators enable proactive management. If your trial-to-paid conversion rate (a leading indicator of revenue) drops in week 1, you can intervene before the impact shows up in monthly revenue three or four weeks later.

Building a Metrics Hierarchy

Large organizations benefit from a structured metrics hierarchy that connects team-level metrics to company-level KPIs. At the top sits the company's primary business objective — grow revenue, expand into new markets, achieve profitability. Below that sit departmental KPIs — for a growth team, new user acquisition; for a product team, activation rate; for a revenue team, ARR growth. Below departmental KPIs sit team-level operational metrics that feed into them.

This hierarchy makes accountability clear: every team can see how their metrics connect to company success, and leadership can trace underperformance in top-line KPIs down to the specific team or operational driver responsible.