Scenario

Control fan-out before aggregating: diagnose child multiplication, pre-aggregate to the parent grain, use clear self-join roles, and keep range bands non-overlapping.

Fan-Out, Pre-Aggregation, Self Joins, and Non-Equi Joins / write query

One-sentence task

M20-A07 - Checkpoint - produce one row per customer with lifetime ordered and paid amounts across fan-out hidden variants. Detect independent one-to-many fan-out and pre-aggregate to a common grain.

Structured output contract

Result grain

one customer with independently staged lifetime ordered and paid amounts

Exact columns

customer_id; customer_name; order_count; lifetime_ordered_amount; lifetime_paid_amount; rows_per_customer_key

Source population

Use the prompt setup plus FROM, JOIN, WHERE, and subquery predicates as the source population. Visible rows are only examples.

Grouping

Group only at the requested output grain: one customer with independently staged lifetime ordered and paid amounts.

Ordering

return order_count; return lifetime_ordered_amount; order by customer_id

Validation

select-only; hidden deterministic variants.

Relevant tables

Time and difficulty

Estimated time

12 minutes

Difficulty

4/5

Objective and concepts

Debug the requested SQL output contract for fan-out, pre-aggregation, self joins, and non-equi joins using source grain, columns, ordering, and edge-case evidence.

SELECT <requested_columns>
FROM <source_table>
WHERE <source_population_filter>
GROUP BY <result_grain_columns>
ORDER BY <deterministic_tie_breakers>;

SELECT o.order_id, COALESCE(items.ordered_amount, 0)::numeric(10,2) AS ordered_amount, COALESCE(payments.paid_amount, 0)::numeric(10,2) AS paid_amount, COALESCE(refunds.refund_amount, 0)::numeric(10,2) AS refund_amount FROM orders o LEFT JOIN (SELECT order_id, SUM(quantity * unit_price)::numeric(10,2) AS ordered_amount FROM order_items GROUP BY order_id) items ON items.order_id = o.order_id LEFT JOIN (SELECT order_id, SUM(amount) FILTER (WHERE payment_status = 'captured')::numeric(10,2) AS paid_amount FROM payments GROUP BY order_id) payments ON payments.order_id = o.order_id LEFT JOIN (SELECT oi.order_id, SUM(r.quantity * oi.unit_price)::numeric(10,2) AS refund_amount FROM returns r JOIN order_items oi ON oi.order_item_id = r.order_item_id GROUP BY oi.order_id) refunds ON refunds.order_id = o.order_id ORDER BY o.order_id;

Assumptions, dialect notes, and common traps

Duplicate policy

Preserve duplicate facts unless the prompt explicitly asks for distinct tuples or set semantics.

Null policy

Preserve NULL, empty string, zero, and false as distinct values unless the contract says to display a fallback.

Tie-breakers

Use every ordering rule in the contract and end tied business metrics with deterministic secondary keys when needed.

Zero-related entities

Preserve requested zero-related, no-match, or complete-period entities and make display fallback explicit.

Numeric tolerance

Round only at the requested final stage; hidden checks use the contract precision rather than visible formatting luck.

• Independent children multiply: Items and payments are joined directly and the resulting SUM looks plausible but repeats facts. Repair: Predict child-count multiplication and pre-aggregate each child collection to the parent grain.
• DISTINCT after SUM cannot repair fan-out: Repeated equal payment or item amounts are deduplicated instead of preserving real repeated facts. Repair: Stage facts at the correct grain before SUM rather than wrapping SUM inputs in DISTINCT.
• Self joins need role names: The same table is joined without clear aliases, making referred and referring customers interchangeable. Repair: Alias each table instance by its role and qualify every selected column.
• Range bands must not overlap: A boundary amount can match two bands or no band because both endpoints are inclusive. Repair: Use lower-inclusive and upper-exclusive predicates for adjacent ranges.

Opened hints

No hints opened yet.