SCHEMA.md

# Solstice SaaS Growth Pack — Schema

## Goal
A dashboard-ready synthetic SaaS metrics pack. Imports cleanly into any BI tool and immediately supports SaaS growth, acquisition, and retention dashboards — no cleanup, no modeling.

## Pack Contents

### `companies.csv`
Grain: `company_id`

| Column | Type | Description |
|---|---|---|
| `company_id` | string | Stable primary key for each company |
| `company_name` | string | Human-readable company name |
| `industry` | string | Industry classification |
| `growth_style` | string | Synthetic profile used to drive realistic trends |
| `founded_date` | date | Company founding date |
| `avg_revenue_per_customer` | decimal | Average monthly revenue per active customer |
| `gross_margin_pct` | decimal | Gross margin percentage used in LTV estimates |
| `initial_active_customers` | integer | Starting active customer base |

### `growth_metrics.csv`
Grain: `date x company_id`

| Column | Type | Description |
|---|---|---|
| `date` | date | Observation date |
| `company_id` | string | Foreign key to `companies.csv` |
| `company_name` | string | Convenience label for charting |
| `revenue` | decimal | Estimated recognized revenue for the day (≈ MRR / 30.44 with small daily variation) |
| `mrr` | decimal | Monthly recurring revenue estimate |
| `new_customers` | integer | Customers acquired on the day |
| `churned_customers` | integer | Customers lost on the day |
| `active_customers` | integer | Active customer count at day end |
| `cac` | decimal | Customer acquisition cost |
| `ltv` | decimal | Customer lifetime value estimate |
| `marketing_spend` | decimal | Marketing spend for the day |
| `churn_rate` | decimal | Daily churn rate as a share of previous active customers |

### `channel_performance.csv`
Grain: `date x company_id x channel`

| Column | Type | Description |
|---|---|---|
| `date` | date | Observation date |
| `company_id` | string | Foreign key to `companies.csv` |
| `company_name` | string | Convenience label for charting |
| `channel` | string | Acquisition channel |
| `impressions` | integer | Channel impressions |
| `clicks` | integer | Channel clicks |
| `conversions` | integer | New customers attributed to the channel |
| `cost` | decimal | Daily channel spend |
| `revenue_generated` | decimal | Revenue attributed to channel conversions |
| `conversion_rate` | decimal | `conversions / clicks` |
| `click_through_rate` | decimal | `clicks / impressions` |

### `customer_segments.csv`
Grain: `company_id x segment`

| Column | Type | Description |
|---|---|---|
| `company_id` | string | Foreign key to `companies.csv` |
| `company_name` | string | Convenience label for charting |
| `segment` | string | Customer segment (`SMB`, `Mid-Market`, `Enterprise`) |
| `avg_ltv` | decimal | Average LTV for the segment |
| `avg_cac` | decimal | Average CAC for the segment |
| `churn_rate` | decimal | Segment churn rate |
| `avg_revenue` | decimal | Average recurring revenue per customer in the segment |

### `metric_definitions.csv`
Grain: `metric_name`

| Column | Type | Description |
|---|---|---|
| `metric_name` | string | Name of metric |
| `definition` | string | Human-readable definition |
| `formula` | string | Formula reference |
| `table_name` | string | Source table |
| `grain` | string | Grain where the metric is valid |

### `dashboard_suggestions.csv`
Grain: `dashboard_name x chart_name`

| Column | Type | Description |
|---|---|---|
| `dashboard_name` | string | Suggested dashboard grouping |
| `chart_name` | string | Suggested chart title |
| `chart_type` | string | Suggested visualization type |
| `primary_table` | string | Main source table |
| `x_axis` | string | Recommended x-axis field |
| `y_axis` | string | Recommended y-axis field(s) |
| `filter_suggestion` | string | Suggested dashboard filters |

## Join Model

- `companies.company_id = growth_metrics.company_id`
- `companies.company_id = channel_performance.company_id`
- `companies.company_id = customer_segments.company_id`

The dataset is intentionally denormalized with `company_name` repeated in fact tables so dashboards can still work even if users only import one or two files.

## Metric Definitions

### `revenue`
- Formula: `(active_customers * avg_revenue_per_customer) / 30.44`
- Notes: Daily recognized revenue approximation. Summing a full month of `revenue` reconciles to `mrr` within ~5%.

### `mrr`
- Formula: `active_customers * avg_revenue_per_customer`
- Notes: Included directly in `growth_metrics.csv`

### `cac`
- Formula: `marketing_spend / new_customers`
- Notes: Protected from divide-by-zero by generator rules

### `ltv`
- Formula: `(avg_revenue_per_customer * gross_margin_pct) / max(churn_rate, 0.01)`
- Notes: Daily churn rate is floored at 0.01 to avoid unstable LTV spikes on low-churn days.

### `churn_rate`
- Formula: `churned_customers / previous_active_customers`

### `conversion_rate`
- Formula: `conversions / clicks`

### `click_through_rate`
- Formula: `clicks / impressions`

## Synthetic Profiles

The generator uses multiple company profiles so the dashboards show realistic differences:

- `steady_plg`: strong SEO/content/referral, efficient long-term growth
- `paid_accelerator`: aggressive paid acquisition, higher spend and growth
- `enterprise_lumpy`: quarter-end deal spikes and lower churn
- `seasonal_b2c`: seasonal demand swings
- `churn_recovery`: visible churn event followed by recovery
- `capital_infusion`: growth acceleration after a mid-period expansion phase

## Dashboard Recommendations

### SaaS Growth Overview
- Revenue Over Time
- MRR and Active Customers

### Acquisition Efficiency
- CAC vs LTV
- Channel Revenue Contribution

### Customer Health
- New vs Churned Customers (Clustered Column)
- Churn Rate Over Time (Line)

### Segment Economics
- Segment LTV/CAC (Grouped Bar)
- Segment Revenue Mix (Stacked Bar)

## Import Notes

- All dates are ISO-8601 (`YYYY-MM-DD`)
- Currency values are USD
- IDs are stable and consistent
- No null-heavy cleanup is required before dashboarding