Vanity Metrics

What It Is

Vanity metrics are measurements that look impressive and reliably trend upward but do not help anyone make a decision or reflect genuine value. Classic examples include lines of code written, number of commits, raw page views, total registered users, or story points completed. They feel good in a status report yet say nothing actionable about whether the product or team is actually succeeding.

The term comes from the lean-startup world but applies broadly to engineering and management dashboards.

Why It Happens

• Easy to measure. Vanity metrics are simple to collect, so they default into dashboards.
• They flatter. Numbers that only go up make reports look good and stakeholders feel reassured.
• Activity mistaken for outcome. Counting effort (commits, hours) is easier than measuring results (value, retention).
• Pressure to show progress. Teams reach for any positive-looking number when asked how things are going.

Why It Hurts

Vanity metrics create false confidence: a dashboard glowing green while the product fails. Effort gets misallocated toward whatever moves the visible number rather than what creates value. Because the metric is a poor proxy, optimizing it produces gamed behavior — more commits, smaller and more numerous, with no improvement in software. Decisions made on vanity data are decisions made blind.

Warning Signs

• Every reported metric only goes up and to the right.
• No one can name a decision that a given metric would change.
• Numbers are cherry-picked for reports rather than chosen for insight.
• Activity counts (lines of code, commits, hours) stand in for outcomes.

Better Alternatives

• Actionable metrics. Track numbers that, when they move, tell you to do something different.
• Outcome-based OKRs. Tie measurement to results (retention, revenue, reliability), not activity.
• A north-star metric. Pick one metric that genuinely captures delivered value and align around it.
• DORA metrics. For delivery health, use deployment frequency, lead time, change-failure rate, and recovery time.

How to Refactor Out of It

For every metric on a dashboard, ask: "What decision does this change?" If the answer is none, remove it. Replace activity counts with outcome measures tied to user and business value. Prefer metrics that can move in both directions and that expose problems, not just successes. Pair counts with rates and ratios that resist gaming. For engineering delivery specifically, adopt the DORA metrics, which correlate with real performance. The goal is a small set of honest, actionable numbers rather than a wall of reassuring ones.