Benchmark Analyzer

Establish baselines, set targets, and track progress — turn raw metrics into benchmarks you can measure against.

When to Use

• "What's our baseline conversion rate?"
• "Set a target for checkout completion — and track us against it"
• "How does this quarter compare to Q2?"
• "Are we on track to hit our retention target?"
• "What's a good rage-click rate? Are we above or below average?"
• "Benchmark our error rate against industry standards"

Mental Model

A metric without context is just a number. A benchmark gives it context:

• Internal baseline: What's normal for us? (12-month average, last quarter's average)
• Target: Where do we want to be? (goal set by the team, industry standard)
• Comparison: How do we compare? (to ourselves over time, to peers if data available)

Workflow

Step 1: Define what to benchmark

Ask the user:

• "What metrics matter most?" (conversion, retention, error rate, etc.)
• "What time period should be the baseline?" (last quarter, last 12 months, since launch)
• "Do you have specific targets, or do you want me to suggest them?"

Step 2: Compute the baseline

For each metric, compute over the baseline period:

fullstory:build_metric(query="checkout conversion rate", output_type="single_number")
fullstory:compute_metric(metric_id, time_range="last_12_months")
→ Baseline: 19.8% (12-month average)

Also compute the standard deviation or range:

fullstory:build_metric(query="checkout conversion rate", output_type="trend")
fullstory:compute_metric(metric_id, time_range="last_12_months")
→ Range: 16% - 24%, typical weekly variance: ±3pp

Step 3: Compare current to baseline

Current (last 30 days): 21.2%
Baseline (12-month avg): 19.8%
→ +1.4pp above baseline (+7%)
→ Within normal range (baseline range: 16-24%) ✅

Step 4: Set targets

Based on the baseline, suggest targets:

• Conservative (achievable): 21-22% (slightly above baseline)
• Ambitious (stretch): 23-24% (top of historical range)
• Moonshot (breakthrough): 25%+ (above historical max)

Ask: "Where do you want to set the target?"

Step 5: Track progress

For each target, build a comparison:

Progress toward Q3 target (22% checkout conversion):
Current: 21.2% (96% of target)
On track: ✅ (if trend continues, should hit 22% by end of Q3)

Step 6: Industry context (with caveats)

Fullstory doesn't provide industry benchmarks. But you can provide general context from public research (with appropriate caveats):

## Conversion Rate Benchmark — E-Commerce Checkout

Your rate: 21%
Industry average (public data): 15-25%
→ You're in the top half, but not exceptional

Top quartile: >25%
→ If you hit your ambitious target (24%), you'd approach top quartile

⚠️ Industry benchmarks are directional, not precise. Different products, different audiences. Use as context, not comparison.

Step 7: Report

## Benchmark Report — Q3 2026

### Checkout Conversion
- Baseline (12-month): 19.8% (range: 16-24%)
- Current (Jul): 21.2% (+1.4pp above baseline)
- Q3 Target: 22%
- Progress: 96% of target, on track ✅
- Industry context: 15-25% for e-commerce (directional)

### Error Rate
- Baseline: 0.9% (range: 0.5-2.1%)
- Current: 0.8% (-0.1pp below baseline)
- Q3 Target: <1.0%
- Progress: Exceeding target ✅

### Rage Clicks
- Baseline: 1,400/month (range: 1,100-2,200)
- Current (Jul): 1,850 (↑32% above baseline) 🔴
- Q3 Target: <1,500/month
- Progress: Off track — promo code issue is driving the spike

### Summary
- 2 of 3 metrics on or exceeding targets
- 1 metric off track (rage clicks) — fix in progress

Baseline Heuristics by Metric Type

Guidelines

• Baselines need enough data. <3 months of data is too noisy. Use at least 6 months, ideally 12.
• Seasonality matters. "Conversion is up vs last month" is meaningless if last month was December (holiday retail spike). Compare to same month last year.
• Targets should be achievable. A target 2x above historical max is demotivating. Suggest a range: conservative → ambitious → moonshot.
• Industry benchmarks are directional. "The average e-commerce conversion rate is 15-25% (per Baymard Institute, 2024)" — always cite the source and caveat the comparison.
• Don't benchmark rage clicks or dead clicks against industry — there's no public data. Benchmark against your own baseline.
• Re-baseline annually. If you've improved the product, the old baseline is stale. "Baseline updated to 12-month average ending Q3 2026."