@useatlas/create 0.0.2 → 0.0.4

package/templates/nextjs-standalone/data/demo-semantic/entities/accounts.yml

@@ -1,170 +1,162 @@
 name: Accounts
 type: fact_table
 table: accounts
-grain: one row per account record
-description: Auto-profiled schema for accounts (80 rows). Contains 7 columns, linked to companies.
+grain: one row per subscription account
+description: |
+  Subscription accounts representing our billing relationship with each customer.
+  Each account belongs to one company and has a plan tier (Free, Starter, Pro,
+  Enterprise), a billing status (Active, Inactive, Suspended, Churned), and a
+  monthly_value representing MRR contribution. Contract dates track the lifecycle.
+  A company can have multiple accounts (e.g. separate subscriptions for different teams).
+
 dimensions:
   - name: id
     sql: id
     type: number
-    description: Primary key
+    description: Primary key — unique account identifier
     primary_key: true
-    unique_count: 80
-    sample_values:
-      - '1'
-      - '2'
-      - '3'
-      - '4'
-      - '5'
-      - '6'
-      - '7'
-      - '8'
+
   - name: company_id
     sql: company_id
     type: number
-    description: Foreign key to companies
-    unique_count: 50
-    sample_values:
-      - '1'
-      - '2'
-      - '3'
-      - '4'
-      - '5'
-      - '6'
-      - '7'
-      - '8'
+    description: Foreign key to companies — which customer owns this account
+
   - name: plan
     sql: plan
     type: string
-    unique_count: 4
-    sample_values:
-      - Enterprise
-      - Free
-      - Pro
-      - Starter
+    description: |
+      Subscription tier. Determines pricing and feature access.
+      Free = $0, Starter = ~$49-89, Pro = ~$299-500, Enterprise = ~$5K-15K.
+    sample_values: [Enterprise, Pro, Starter, Free]
+
   - name: status
     sql: status
     type: string
-    unique_count: 4
-    sample_values:
-      - Active
-      - Churned
-      - Inactive
-      - Suspended
+    description: |
+      Current billing status. Active = paying, Inactive = paused,
+      Suspended = payment issue, Churned = cancelled.
+    sample_values: [Active, Churned, Inactive, Suspended]
+
   - name: monthly_value
     sql: monthly_value
     type: number
-    unique_count: 41
-    sample_values:
-      - '0'
-      - '49'
-      - '59'
-      - '69'
-      - '79'
-      - '89'
-      - '299'
-      - '350'
+    description: |
+      Monthly Recurring Revenue (MRR) from this account in dollars.
+      Free accounts have $0. This is the core revenue metric.
+
   - name: contract_start
     sql: contract_start
     type: date
-    unique_count: 49
-    sample_values:
-      - Sat Jun 01 2019 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Thu Aug 01 2019 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Fri Nov 01 2019 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Wed Jan 01 2020 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Sat Feb 01 2020 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Sun Mar 01 2020 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Wed Apr 01 2020 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Fri May 01 2020 00:00:00 GMT-0400 (Eastern Daylight Time)
+    description: Date the subscription began
+
   - name: contract_end
     sql: contract_end
     type: date
-    unique_count: 10
-    null_count: 70
-    sample_values:
-      - Fri Jan 01 2021 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Sat May 01 2021 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Wed Feb 01 2023 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Wed Mar 01 2023 00:00:00 GMT-0500 (Eastern Standard Time)
-      - Sat Apr 01 2023 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Mon May 01 2023 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Thu Jun 01 2023 00:00:00 GMT-0400 (Eastern Daylight Time)
-      - Tue Aug 01 2023 00:00:00 GMT-0400 (Eastern Daylight Time)
+    description: |
+      Date the subscription ended. NULL means the account is still under contract.
+      Set when an account churns or is terminated.
+
   - name: monthly_value_bucket
     sql: |-
       CASE
-        WHEN monthly_value < (SELECT PERCENTILE_CONT(0.33) WITHIN GROUP (ORDER BY monthly_value) FROM accounts) THEN 'Low'
-        WHEN monthly_value < (SELECT PERCENTILE_CONT(0.66) WITHIN GROUP (ORDER BY monthly_value) FROM accounts) THEN 'Medium'
+        WHEN monthly_value = 0 THEN 'Free'
+        WHEN monthly_value < 100 THEN 'Low'
+        WHEN monthly_value < 1000 THEN 'Mid'
         ELSE 'High'
       END
     type: string
-    description: monthly value bucketed into Low/Medium/High terciles
+    description: MRR tier bucket — Free / Low / Mid / High
     virtual: true
-    sample_values:
-      - Low
-      - Medium
-      - High
+    sample_values: [Free, Low, Mid, High]
+
   - name: contract_start_year
     sql: EXTRACT(YEAR FROM contract_start)
     type: number
-    description: Year extracted from contract_start
+    description: Year the contract started — useful for cohort analysis
     virtual: true
+
   - name: contract_start_month
     sql: TO_CHAR(contract_start, 'YYYY-MM')
     type: string
-    description: Year-month extracted from contract_start
+    description: Year-month the contract started
     virtual: true
-  - name: contract_end_year
-    sql: EXTRACT(YEAR FROM contract_end)
-    type: number
-    description: Year extracted from contract_end
-    virtual: true
-  - name: contract_end_month
-    sql: TO_CHAR(contract_end, 'YYYY-MM')
-    type: string
-    description: Year-month extracted from contract_end
+
+  - name: is_churned
+    sql: CASE WHEN status = 'Churned' THEN true ELSE false END
+    type: boolean
+    description: Whether this account has churned
     virtual: true
+
 measures:
   - name: account_count
     sql: id
     type: count_distinct
-  - name: total_monthly_value
+    description: Number of unique accounts
+
+  - name: total_mrr
     sql: monthly_value
     type: sum
-    description: Sum of monthly value
-  - name: avg_monthly_value
+    description: Total Monthly Recurring Revenue
+
+  - name: avg_mrr
     sql: monthly_value
     type: avg
-    description: Average monthly value
+    description: Average MRR per account
+
 joins:
   - target_entity: Companies
     relationship: many_to_one
     join_columns:
       from: company_id
       to: id
-    description: Each account belongs to one company
+    description: Each account belongs to one company (customer)
+
 use_cases:
-  - Use for segmentation analysis by plan, status
-  - Use for aggregation and trends on monthly_value
-  - Use for time-series analysis using contract_start, contract_end
-  - Join with companies for cross-entity analysis
+  - MRR analysis — total, by plan, by status, over time
+  - Churn analysis — churned vs active accounts, churn rate by plan
+  - Revenue segmentation — which plans drive the most revenue
+  - Contract lifecycle — average duration, upcoming renewals
+  - Cross-entity joins to companies for firmographic revenue analysis
+
 query_patterns:
-  - description: Accounts by plan
+  - name: mrr_by_plan
+    description: Total MRR and account count by plan tier
     sql: |-
-      SELECT plan, COUNT(*) as count
+      SELECT plan,
+             SUM(monthly_value) AS total_mrr,
+             COUNT(*) AS account_count,
+             AVG(monthly_value) AS avg_mrr
       FROM accounts
+      WHERE status = 'Active'
       GROUP BY plan
-      ORDER BY count DESC
-  - description: Accounts by status
+      ORDER BY total_mrr DESC
+
+  - name: churn_rate_by_plan
+    description: Churn rate by plan type
     sql: |-
-      SELECT status, COUNT(*) as count
+      SELECT plan,
+             COUNT(*) FILTER (WHERE status = 'Churned') AS churned,
+             COUNT(*) AS total,
+             ROUND(100.0 * COUNT(*) FILTER (WHERE status = 'Churned') / COUNT(*), 1) AS churn_pct
+      FROM accounts
+      GROUP BY plan
+      ORDER BY churn_pct DESC
+
+  - name: accounts_by_status
+    description: Account distribution by billing status
+    sql: |-
+      SELECT status, COUNT(*) AS count
       FROM accounts
       GROUP BY status
       ORDER BY count DESC
-  - description: Total monthly_value by plan
+
+  - name: mrr_by_contract_year
+    description: MRR cohort analysis by contract start year
     sql: |-
-      SELECT plan, SUM(monthly_value) as total_monthly_value, COUNT(*) as count
+      SELECT EXTRACT(YEAR FROM contract_start) AS cohort_year,
+             SUM(monthly_value) AS total_mrr,
+             COUNT(*) AS accounts
       FROM accounts
-      GROUP BY plan
-      ORDER BY total_monthly_value DESC
+      WHERE status = 'Active'
+      GROUP BY cohort_year
+      ORDER BY cohort_year

package/templates/nextjs-standalone/data/demo-semantic/entities/companies.yml

@@ -1,207 +1,143 @@
 name: Companies
-type: fact_table
+type: dimension_table
 table: companies
-grain: one row per company record
-description: Auto-profiled schema for companies (50 rows). Contains 8 columns.
+grain: one row per customer company
+description: |
+  Customer companies (our B2B clients). Contains firmographic data including
+  industry vertical, headcount, founding year, HQ country, annual revenue,
+  and valuation. Each company can have multiple accounts (subscriptions)
+  and multiple people (contacts). This is the central dimension table for
+  customer-level analysis.
+
 dimensions:
   - name: id
     sql: id
     type: number
-    description: Primary key
+    description: Primary key — unique company identifier
     primary_key: true
-    unique_count: 50
-    sample_values:
-      - '1'
-      - '2'
-      - '3'
-      - '4'
-      - '5'
-      - '6'
-      - '7'
-      - '8'
+
   - name: name
     sql: name
     type: string
-    unique_count: 50
-    sample_values:
-      - Acme Corp
-      - AdTech Global
-      - AutoForge
-      - BioGenix
-      - BrightMinds
-      - CartFlow
-      - CastMedia
-      - CleanJoule
+    description: Company name
+    sample_values: [Acme Corp, DataFlow Inc, CloudSync, RetailHub]
+
   - name: industry
     sql: industry
     type: string
-    unique_count: 8
-    sample_values:
-      - Education
-      - Energy
-      - Finance
-      - Healthcare
-      - Manufacturing
-      - Media
-      - Retail
-      - Technology
+    description: Business sector / vertical
+    sample_values: [Education, Energy, Finance, Healthcare, Manufacturing, Media, Retail, Technology]
+
   - name: employee_count
     sql: employee_count
     type: number
-    unique_count: 46
-    sample_values:
-      - '50'
-      - '55'
-      - '60'
-      - '65'
-      - '70'
-      - '75'
-      - '80'
-      - '85'
+    description: Total headcount at the company
+
   - name: founded_year
     sql: founded_year
     type: number
-    unique_count: 19
-    sample_values:
-      - '2004'
-      - '2005'
-      - '2006'
-      - '2007'
-      - '2008'
-      - '2009'
-      - '2010'
-      - '2011'
+    description: Year the company was founded
+
   - name: country
     sql: country
     type: string
-    unique_count: 10
-    sample_values:
-      - AU
-      - BR
-      - CA
-      - DE
-      - FR
-      - IN
-      - JP
-      - SG
+    description: ISO 2-letter code for company HQ location
+    sample_values: [AU, BR, CA, DE, FR, IN, JP, SG, UK, US]
+
   - name: revenue
     sql: revenue
     type: number
-    unique_count: 50
-    sample_values:
-      - '2000000'
-      - '2500000'
-      - '3500000'
-      - '4000000'
-      - '5000000'
-      - '6000000'
-      - '6500000'
-      - '7000000'
+    description: Annual revenue in dollars (the company's own revenue, not what they pay us)
+
   - name: valuation
     sql: valuation
     type: number
-    unique_count: 46
-    sample_values:
-      - '16000000'
-      - '20000000'
-      - '25000000'
-      - '28000000'
-      - '30000000'
-      - '32000000'
-      - '35000000'
-      - '40000000'
-  - name: employee_count_bucket
+    description: Company valuation in dollars
+
+  - name: company_size
     sql: |-
       CASE
-        WHEN employee_count < (SELECT PERCENTILE_CONT(0.33) WITHIN GROUP (ORDER BY employee_count) FROM companies) THEN 'Low'
-        WHEN employee_count < (SELECT PERCENTILE_CONT(0.66) WITHIN GROUP (ORDER BY employee_count) FROM companies) THEN 'Medium'
-        ELSE 'High'
+        WHEN employee_count < 100 THEN 'Small'
+        WHEN employee_count < 300 THEN 'Mid-Market'
+        ELSE 'Enterprise'
       END
     type: string
-    description: employee count bucketed into Low/Medium/High terciles
+    description: Company size segment based on headcount
     virtual: true
-    sample_values:
-      - Low
-      - Medium
-      - High
-  - name: founded_year_bucket
+    sample_values: [Small, Mid-Market, Enterprise]
+
+  - name: revenue_tier
     sql: |-
       CASE
-        WHEN founded_year < (SELECT PERCENTILE_CONT(0.33) WITHIN GROUP (ORDER BY founded_year) FROM companies) THEN 'Low'
-        WHEN founded_year < (SELECT PERCENTILE_CONT(0.66) WITHIN GROUP (ORDER BY founded_year) FROM companies) THEN 'Medium'
-        ELSE 'High'
+        WHEN revenue < 10000000 THEN 'Under $10M'
+        WHEN revenue < 50000000 THEN '$10M-$50M'
+        WHEN revenue < 100000000 THEN '$50M-$100M'
+        ELSE '$100M+'
       END
     type: string
-    description: founded year bucketed into Low/Medium/High terciles
+    description: Revenue tier for segmentation
     virtual: true
-    sample_values:
-      - Low
-      - Medium
-      - High
-  - name: revenue_bucket
-    sql: |-
-      CASE
-        WHEN revenue < (SELECT PERCENTILE_CONT(0.33) WITHIN GROUP (ORDER BY revenue) FROM companies) THEN 'Low'
-        WHEN revenue < (SELECT PERCENTILE_CONT(0.66) WITHIN GROUP (ORDER BY revenue) FROM companies) THEN 'Medium'
-        ELSE 'High'
-      END
-    type: string
-    description: revenue bucketed into Low/Medium/High terciles
-    virtual: true
-    sample_values:
-      - Low
-      - Medium
-      - High
-  - name: valuation_bucket
-    sql: |-
-      CASE
-        WHEN valuation < (SELECT PERCENTILE_CONT(0.33) WITHIN GROUP (ORDER BY valuation) FROM companies) THEN 'Low'
-        WHEN valuation < (SELECT PERCENTILE_CONT(0.66) WITHIN GROUP (ORDER BY valuation) FROM companies) THEN 'Medium'
-        ELSE 'High'
-      END
-    type: string
-    description: valuation bucketed into Low/Medium/High terciles
-    virtual: true
-    sample_values:
-      - Low
-      - Medium
-      - High
+    sample_values: [Under $10M, $10M-$50M, $50M-$100M, $100M+]
+
 measures:
   - name: company_count
     sql: id
     type: count_distinct
-  - name: total_employee_count
-    sql: employee_count
-    type: sum
-    description: Sum of employee count
+    description: Number of unique companies
+
   - name: avg_employee_count
     sql: employee_count
     type: avg
-    description: Average employee count
-  - name: total_founded_year
-    sql: founded_year
-    type: sum
-    description: Sum of founded year
-  - name: avg_founded_year
-    sql: founded_year
-    type: avg
-    description: Average founded year
+    description: Average headcount per company
+
   - name: total_revenue
     sql: revenue
     type: sum
-    description: Sum of revenue
+    description: Combined annual revenue of all companies
+
   - name: avg_revenue
     sql: revenue
     type: avg
-    description: Average revenue
-  - name: total_valuation
-    sql: valuation
-    type: sum
-    description: Sum of valuation
+    description: Average annual revenue per company
+
   - name: avg_valuation
     sql: valuation
     type: avg
-    description: Average valuation
+    description: Average company valuation
+
 use_cases:
-  - Use for aggregation and trends on employee_count, founded_year, revenue, valuation
-  - Avoid for row-level accounts/people queries — use those entities directly
+  - Customer segmentation by industry, country, company size
+  - Revenue and valuation analysis across the customer base
+  - Firmographic profiling for ICP (ideal customer profile) analysis
+  - Join to accounts for revenue-per-company analysis
+  - Join to people for stakeholder density per company
+
+query_patterns:
+  - name: companies_by_industry
+    description: Customer count and avg revenue by industry
+    sql: |-
+      SELECT industry,
+             COUNT(*) AS company_count,
+             AVG(revenue) AS avg_revenue,
+             AVG(employee_count) AS avg_headcount
+      FROM companies
+      GROUP BY industry
+      ORDER BY company_count DESC
+
+  - name: companies_by_country
+    description: Customer distribution by country
+    sql: |-
+      SELECT country,
+             COUNT(*) AS company_count,
+             SUM(revenue) AS total_revenue
+      FROM companies
+      GROUP BY country
+      ORDER BY company_count DESC
+
+  - name: top_companies_by_revenue
+    description: Largest customers by annual revenue
+    sql: |-
+      SELECT name, industry, country, revenue, employee_count
+      FROM companies
+      ORDER BY revenue DESC
+      LIMIT 10