The Analytics Metrics Hierarchy: From Vanity Metrics to Business Outcomes
By Emily Redmond, Data Analyst at Emilytics · April 2026
TL;DR: Metrics pyramid: business outcomes (revenue) → leading indicators (what predicts outcomes) → engagement (activity) → vanity (feel-good numbers). Build reporting from the top down. Track vanity metrics only if they predict leading indicators.
Why Most Teams Track the Wrong Metrics
You send out a report that says "pageviews are up 200%!" and nobody cares.
Then you say "revenue is up 8%" and everyone suddenly pays attention.
The difference is simple: pageviews are a vanity metric. Revenue is a business outcome. And your job is to report on outcomes, not metrics that feel good.
The Metrics Hierarchy
Think of metrics like a pyramid. Each level supports the ones above it.
Level 1: Business Outcomes (Top of Pyramid)
These are what your business actually cares about:
Examples:
- Revenue
- Profit
- Market share
- Customer lifetime value
- Enterprise value
Why it matters: This is the only level that directly impacts business success.
Questions it answers: Are we making money? Are we growing? Is the company worth more?
Level 2: Leading Indicators
These are metrics that predict business outcomes:
Examples:
- Conversion rate (predicts revenue)
- Churn rate (predicts lifetime value)
- Feature adoption (predicts engagement, retention)
- Customer satisfaction (predicts retention)
- Cost per acquisition (predicts profitability)
Why it matters: Leading indicators let you predict outcomes before they happen. If churn is rising, you know revenue will fall in 60 days.
Questions it answers: What's likely to drive our outcomes? Are we setting ourselves up for success or failure?
Level 3: Engagement Metrics
These show activity but don't directly predict outcomes:
Examples:
- Sessions
- Pageviews
- Time on site
- Email opens
- Video plays
- Clicks
Why it matters: Engagement shows that people are interested, but doesn't prove they'll convert or stay.
Questions it answers: Are people paying attention? Are we getting engagement?
The trap: High engagement without conversion (or retention) is noise.
Level 4: Vanity Metrics (Bottom of Pyramid)
These feel good but usually don't drive decisions:
Examples:
- Total website visits
- Total social followers
- Total downloads
- Total newsletter subscribers
- Total podcast listens (without engagement or conversion)
Why it matters: They don't. They feel good. That's the opposite of useful.
Questions it answers: Does this make us feel good? (That's it.)
The warning: If a metric doesn't drive a decision or predict a business outcome, it's probably vanity.
How to Know if a Metric Matters
Ask three questions:
1. If this metric moved up 30%, would we change strategy?
If no, it's vanity. Don't report it.
If yes, it might matter. Keep asking.
2. Does this metric predict a business outcome?
Conversion rate predicts revenue. High churn rate predicts revenue decline. Feature adoption predicts retention.
Session count alone? Doesn't predict much of anything.
3. Can we act on it?
If a metric is up but you have no idea what to change, it's not useful.
"Feature adoption is down" means something. "Bounce rate is up" might mean something. "Total visits is up" means... traffic happened.
Real Examples by Industry
E-commerce Metrics Pyramid
Level 1 (Outcomes): Revenue, profit margin, customer lifetime value
Level 2 (Leading indicators): Conversion rate, average order value, repeat purchase rate, cost per acquisition
Level 3 (Engagement): Sessions, pages per session, time on site, add-to-cart rate
Level 4 (Vanity): Total page views, total visitors, total add-to-carts (without conversion)
SaaS Metrics Pyramid
Level 1 (Outcomes): Monthly recurring revenue, profit, enterprise value
Level 2 (Leading indicators): Trial-to-paid conversion, churn rate, net revenue retention, feature adoption
Level 3 (Engagement): Monthly active users, feature usage, session duration
Level 4 (Vanity): Total signups (without conversion), total logins, total feature clicks
Content/Publisher Metrics Pyramid
Level 1 (Outcomes): Revenue (ads, subscriptions, sponsorships), audience size
Level 2 (Leading indicators): Return visitor rate, average session duration, email signup rate, conversion rate to paid
Level 3 (Engagement): Sessions, pageviews, time on page, scroll depth
Level 4 (Vanity): Total pageviews, total sessions, bounce rate (without understanding intent)
When Vanity Metrics Actually Matter
Vanity metrics become useful when they predict leading indicators:
Example 1: Total followers doesn't matter. But if you track total followers AND know that 3% convert to customers, now it's useful. Total followers becomes a proxy for business outcome.
Example 2: Total pageviews doesn't matter. But if you know 2% of pageviews convert to leads, total pageviews becomes a predictor. Now it matters.
The rule: Vanity metric + conversion rate = leading indicator.
Building a Metrics Hierarchy for Your Business
Step 1: Define your business outcome.
What does success look like? Revenue? User growth? Engagement? Pick one primary outcome.
Step 2: Identify leading indicators.
What metrics predict that outcome? What needs to happen before revenue grows? What causes churn?
Step 3: Map engagement metrics.
What activities suggest people are moving toward the outcome? What do converters do before converting?
Step 4: Identify vanity metrics (and avoid them).
What metrics feel good but don't predict anything? What should you NOT report on?
Step 5: Build your reporting hierarchy.
Top of the report: Level 1 (outcomes) Next: Level 2 (leading indicators) Next: Level 3 (engagement, only if it predicts level 2) Never: Level 4 (vanity)
Frequently Asked Questions
Q: Is bounce rate a vanity metric?
A: Depends. If 80% of your traffic bounces from search but they're high-intent and convert, bounce rate is misleading. If your bounce rate increases AND leading indicators drop, it matters. Usually, it's a distraction metric.
Q: What about user satisfaction (NPS)?
A: NPS is a leading indicator. It predicts churn and revenue. Report it.
Q: Is "total users" a vanity metric?
A: Yes, if it's all users. "Active users" or "monthly active users" is better (it predicts engagement). "Paying users" is even better (it predicts revenue).
Q: Should I report metrics I can't act on?
A: No. If you can't answer "what should we do about this?", don't report it.
Q: How do I know which metrics to report to which audience?
A: Executives see outcomes. Managers see leading indicators. Practitioners see engagement. Nobody sees vanity metrics.
The Bottom Line
Stop reporting pageviews, total users, and bounce rate. Start reporting what drives revenue: conversion rate, churn rate, customer lifetime value, cost per acquisition.
The best reports have an inverse pyramid: biggest numbers at the top (revenue), supported by smaller numbers below (conversion, retention, cost).
Build your metrics hierarchy. Report from top to bottom. Everything else is noise.
For more on building reports, see analytics reporting guide or KPI vs. analytics dashboards.
Emily Redmond is a data analyst at Emilytics — AI analytics agent watching your GA4, Search Console, and Bing data around the clock. 8 years experience. Say hi →