Choosing the Right KPIs: Stop Tracking Vanity Metrics

You're tracking 47 metrics. Congratulations, you're measuring everything and understanding nothing.

Let's fix that.

The Vanity Metrics Trap

These Numbers Look Great But Mean Nothing

Total Registered Users

Users: 100,000 🎉
Active users: 847
Revenue: $4,200

99% of your users don't use your product. That's not success.

Page Views

Page views: 1,000,000
Bounce rate: 94%
Time on page: 8 seconds
Conversions: 12

Bots and accidental clicks don't pay bills.

Social Media Followers

Followers: 50,000
Engagement rate: 0.2%
Clicks to website: 23
Customers from social: 1

Vanity metrics make you feel good but don't make you money.

What Makes a Good KPI?

The SMART Framework

Specific - Clear and well-defined

❌ "Improve customer satisfaction"
✅ "Increase NPS from 45 to 60"

Measurable - Can be quantified

❌ "Better product quality"
✅ "Reduce bug reports by 30%"

Actionable - You can influence it

❌ "Market conditions"
✅ "Customer acquisition cost"

Relevant - Tied to business goals

❌ "Number of meetings"
✅ "Revenue per employee"

Time-bound - Has a deadline

❌ "Eventually grow revenue"
✅ "Grow revenue 20% by Q2"

The Three Questions Test

For each metric, ask:

1. What decision does this inform?

   • If none, it's vanity

2. What action do we take if it moves?

   • If unclear, it's not actionable

3. Who owns this metric?

   • If nobody, it won't improve

KPIs by Business Type

SaaS Companies

North Star Metric:

Annual Recurring Revenue (ARR)
or
Monthly Recurring Revenue (MRR)

Supporting KPIs:

| Category | Metric | Why It Matters | |----------|--------|----------------| | Growth | New MRR | Revenue expansion | | Retention | Churn Rate | Customer satisfaction | | Efficiency | CAC Payback Period | Unit economics | | Usage | Daily Active Users (DAU) | Product engagement | | Quality | Net Promoter Score (NPS) | Customer advocacy |

Calculation:

-- MRR
SELECT
  SUM(subscription_amount) as mrr
FROM subscriptions
WHERE status = 'active'
  AND billing_period = 'monthly';

-- Churn Rate
SELECT
  COUNT(CASE WHEN status = 'cancelled' THEN 1 END) * 100.0 /
  COUNT(*) as churn_rate
FROM subscriptions
WHERE cancelled_date >= DATE_TRUNC('month', CURRENT_DATE);

-- CAC Payback Period
SELECT
  AVG(customer_acquisition_cost) /
  AVG(monthly_revenue_per_customer) as payback_months
FROM customers;

E-Commerce

North Star Metric:

Gross Merchandise Value (GMV)

Supporting KPIs:

| Category | Metric | Formula | |----------|--------|---------| | Revenue | Average Order Value | Total Revenue / Orders | | Conversion | Cart Abandonment | Carts Created / Orders | | Retention | Repeat Purchase Rate | Repeat Customers / Total | | Marketing | ROAS | Revenue / Ad Spend | | Operations | Order Fulfillment Time | Ship Date - Order Date |

Example Dashboard:

📊 Weekly E-Commerce Metrics

GMV: $147,000 (↑ 12% vs last week)
├─ Orders: 1,234 (↑ 8%)
├─ AOV: $119 (↑ 4%)
└─ Conversion: 2.4% (↓ 0.2%)

⚠️ Action Item: Conversion dropped
   Investigate checkout flow

B2B/Sales-Driven

North Star Metric:

Revenue

Supporting KPIs:

| Stage | Metric | Target | |-------|--------|--------| | Lead Gen | SQLs Generated | 100/month | | Sales | Win Rate | 25% | | Efficiency | Sales Cycle Length | 45 days | | Retention | Gross Revenue Retention | 100% | | Growth | Net Revenue Retention | 120% |

Sales Pipeline:

Leads: 400
  ↓ (25% qualify)
SQLs: 100
  ↓ (50% progress)
Opportunities: 50
  ↓ (25% close)
Customers: 12-13/month

Content/Media

North Star Metric:

Engaged Time
(not just page views)

Supporting KPIs:

• Active Users (daily, weekly, monthly)
• Content Engagement Rate
• Subscriber Growth
• Revenue per User (ads or subscriptions)
• Content Production Efficiency

Marketplaces

North Star Metric:

Gross Transaction Value (GTV)

Supporting KPIs:

Two-sided metrics:

Supply Side:

• Active sellers
• Listings created
• Seller retention

Demand Side:

• Active buyers
• Purchase frequency
• Buyer lifetime value

Platform Health:

• Take rate (commission %)
• Repeat transaction rate
• Time to first transaction

Building Your KPI Framework

Step 1: Define Your North Star

The ONE metric that matters most.

For every business:

North Star = Value Delivered to Customers

Examples:

• Spotify: Time spent listening
• Airbnb: Nights booked
• Slack: Messages sent by teams
• Amazon: Purchases per customer

Step 2: Map the Customer Journey

Awareness → Interest → Consideration → Purchase → Retention → Advocacy

One KPI per stage:

Awareness: Website visitors
Interest: Content engagement
Consideration: Product demos
Purchase: Conversion rate
Retention: Churn rate
Advocacy: NPS

Step 3: Identify Input vs Output Metrics

Output Metrics (lagging indicators):

• Revenue
• Profit
• Customer count

Input Metrics (leading indicators):

• Sales calls made
• Trials started
• Feature adoption

Focus on inputs to drive outputs.

Step 4: Set Targets

Use historical data:

-- Historical performance
SELECT
  DATE_TRUNC('month', date) as month,
  AVG(conversion_rate) as avg_conversion,
  STDDEV(conversion_rate) as std_dev
FROM daily_metrics
WHERE date >= CURRENT_DATE - INTERVAL '12 months'
GROUP BY month;

Set realistic targets:

Current: 2.4%
3-month average: 2.2%
Target: 2.7% (realistic 12% improvement)
Stretch: 3.0% (ambitious 25% improvement)

Advanced KPI Concepts

Ratio Metrics

More informative than raw numbers:

❌ Revenue: $100,000
✅ Revenue per employee: $250,000

❌ Support tickets: 500
✅ Tickets per customer: 0.5

❌ Marketing spend: $50,000
✅ CAC: $125

Cohort Analysis

Track metrics by cohort:

-- Retention by signup month
SELECT
  DATE_TRUNC('month', signup_date) as cohort,
  DATE_TRUNC('month', activity_date) as month,
  COUNT(DISTINCT user_id) * 100.0 /
    FIRST_VALUE(COUNT(DISTINCT user_id))
      OVER (PARTITION BY cohort ORDER BY month)
    as retention_rate
FROM user_activity
GROUP BY cohort, month
ORDER BY cohort, month;

Result:

| Cohort  | Month 0 | Month 1 | Month 2 | Month 3 |
|---------|---------|---------|---------|---------|
| Jan 2024| 100%    | 45%     | 38%     | 35%     |
| Feb 2024| 100%    | 52%     | 46%     | 43%     | ← Better!
| Mar 2024| 100%    | 48%     | 42%     |         |

Unit Economics

Understand profitability at the customer level:

Customer Lifetime Value (LTV)
= (Average Purchase Value × Purchase Frequency × Customer Lifespan)

Customer Acquisition Cost (CAC)
= (Sales + Marketing Costs) / New Customers

LTV:CAC Ratio
= LTV / CAC

Healthy ratio: 3:1 or higher

Example:

Average purchase: $100
Purchases per year: 4
Customer lifespan: 3 years
LTV = $100 × 4 × 3 = $1,200

CAC = $400

LTV:CAC = 3:1 ✅ Healthy

Common Mistakes

1. Too Many KPIs

❌ Tracking 50 metrics
✅ 1 North Star + 5-7 supporting metrics

You can't optimize everything.

2. Not Segmenting

❌ Overall conversion: 2.4%
✅ Mobile: 1.8%
   Desktop: 3.2%

Now you know where to improve!

3. Ignoring Statistical Significance

Conversion this week: 2.42%
Conversion last week: 2.38%

Difference: 4%! 🎉

But sample size = 50 visitors
Not statistically significant ❌

Wait for more data.

4. Not Setting Thresholds

Define when to act:

🟢 Green: > 3.0% conversion
🟡 Yellow: 2.5-3.0% conversion
🔴 Red: < 2.5% conversion

Automatic alerts when crossing thresholds.

5. Metrics Without Ownership

For each KPI:
- Owner: Who's responsible?
- Goal: What's the target?
- Actions: What levers can they pull?
- Review: How often do we check?

The KPI Dashboard

Executive View (Weekly)

┌─────────────────────────────────┐
│ Revenue: $2.4M (↑ 12%)         │
│ New Customers: 247 (↑ 8%)      │
│ Churn: 3.2% (↓ 0.5%)           │
│ NPS: 58 (↑ 3 points)           │
└─────────────────────────────────┘

High-level, actionable.

Operational View (Daily)

┌─────────────────────────────────┐
│ Today's Metrics (vs yesterday)  │
├─────────────────────────────────┤
│ Signups: 23 (↑ 4)              │
│ Trials Started: 12 (↓ 2)       │
│ Conversions: 3 (→ 0)           │
│ Churn: 1 (↓ 1)                 │
│ Support Tickets: 14 (↑ 3)      │
└─────────────────────────────────┘

Detailed, diagnostic.

Analytical View (Monthly)

Deep dive:

• Trends over time
• Segment breakdowns
• Cohort analysis
• Correlation analysis

Conclusion

The best KPI framework is:

• Simple - Few metrics, easy to understand
• Actionable - Can influence through specific actions
• Aligned - Tied to business goals
• Owned - Clear responsibility
• Reviewed - Regular cadence

Stop tracking everything. Start tracking what matters.

Your dashboard should answer three questions:

1. Are we winning?
2. Why or why not?
3. What should we do about it?

If your metrics can't answer these, you're measuring the wrong things.

Choose wisely. Measure ruthlessly. Act decisively.