# Period Comparison Investigation

Discover what changed between time periods and why

# Period Comparison Investigation

Compare any two time periods to understand what changed and why.

***

## The Question

> **"What changed this quarter compared to last quarter?"**

Or variations:

* "Compare Q3 to Q2"
* "What's different about this month vs last month?"
* "How did performance change year over year?"
* "What improved and what declined?"

***

## What Scoop Investigates

Scoop performs a comprehensive comparison:

```
Investigation Plan:
├── Probe 1: Identify all metrics
│   └── What metrics exist in the data?
├── Probe 2: Calculate changes
│   └── How did each metric change?
├── Probe 3: Find significant changes
│   └── Which changes are statistically meaningful?
├── Probe 4: Analyze by dimension
│   └── Break down changes by segment, region, etc.
├── Probe 5: Identify drivers
│   └── ML analysis: What factors drove the changes?
└── Synthesis: Executive summary
```

***

## Example Output

```
Investigation Results: Q3 vs Q2 Comparison

SUMMARY:
Overall performance improved with revenue up 12%,
but customer acquisition costs increased 23%.

KEY IMPROVEMENTS:
├── Revenue: +12% ($4.2M → $4.7M)
├── Win Rate: +8 points (32% → 40%)
├── Customer Satisfaction: +0.4 points (8.1 → 8.5)
├── Deal Size (avg): +15% ($24K → $28K)
└── Time to Close: -5 days (45 → 40 days)

KEY DECLINES:
├── Customer Acquisition Cost: +23% ($850 → $1,045)
├── Lead Volume: -18% (1,200 → 984)
├── Trial Conversion: -3 points (28% → 25%)
└── Support Response Time: +2 hours (4h → 6h)

BREAKDOWN BY SEGMENT:
├── Enterprise: Revenue +24%, Win Rate +12 pts
├── Mid-Market: Revenue +8%, Win Rate +2 pts
└── SMB: Revenue -3%, Win Rate -1 pt

ROOT CAUSE ANALYSIS:
├── Revenue increase driven by 3 large enterprise deals
├── Win rate improvement correlated with new demo process
├── CAC increase due to 40% higher ad spend (CPL up 35%)
├── Lead volume decline from paused content marketing
└── Support slowdown linked to 2 team departures

RECOMMENDED FOCUS AREAS:
1. Investigate high CAC - ROI may not justify spend
2. Resume content marketing for lead volume
3. Address support capacity before it impacts CSAT
4. Document enterprise success factors to replicate
```

***

## Sample Prompts

### Basic Comparison

```
"What changed this quarter vs last quarter?"
```

### Specific Metrics

```
"How did revenue and costs change month over month?"
```

### With Focus Area

```
"What changed in our enterprise segment this quarter?"
```

### Year Over Year

```
"Compare this Q3 to Q3 last year"
```

### Root Cause Focus

```
"Why did performance change between Q2 and Q3?"
```

***

## Follow-Up Questions

| Follow-Up                                | What It Reveals              |
| ---------------------------------------- | ---------------------------- |
| "Why did CAC increase?"                  | Deep dive on specific metric |
| "What drove the enterprise improvement?" | Success factor analysis      |
| "Show me the weekly trend"               | More granular timing         |
| "Which team members improved most?"      | Individual performance       |
| "What predicts these improvements?"      | ML pattern analysis          |

***

## Data Requirements

Any dataset with:

| Field           | Purpose                                        |
| --------------- | ---------------------------------------------- |
| Date Field      | For period grouping                            |
| Numeric Metrics | What to compare                                |
| Dimensions      | For breakdown analysis (segment, region, etc.) |

The more metrics and dimensions, the richer the comparison.

***

## Types of Comparisons

### Sequential Periods

* This month vs last month
* This quarter vs last quarter
* This week vs last week

### Year-Over-Year

* Q3 2025 vs Q3 2024
* This month vs same month last year
* Rolling 12 months vs prior 12

### Custom Ranges

* Before vs after product launch
* Pre-campaign vs post-campaign
* First half vs second half

***

## Understanding the Analysis

### Significant vs Noise

Scoop identifies which changes are meaningful:

* Statistical significance testing
* Filters out random variation
* Highlights changes worth investigating

### Dimension Breakdowns

Shows WHERE changes happened:

* By customer segment
* By product line
* By region/territory
* By team member

### Root Cause Analysis

Uses ML to find WHY changes happened:

* Correlation analysis
* Factor importance
* Pattern detection

***

## Tips for Better Comparisons

1. **Ensure date coverage** - Both periods need sufficient data
2. **Include dimensions** - Enables breakdown analysis
3. **Add context fields** - Campaign, rep, source for richer analysis
4. **Consistent definitions** - Same metrics calculated the same way

***

## Related Patterns

* [Pipeline Investigation](pipeline-investigation.md) - Pipeline-specific comparison
* [Revenue Investigation](revenue-investigation.md) - Revenue-focused analysis
* [Churn Investigation](churn-investigation.md) - Compare churned vs retained