Power BI Relationship Cardinality and Ambiguity Check
Examine the table relationships in a Power BI model for wrong cardinality, many-to-many risks and ambiguous paths that inflate or hide totals.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.
Please check the relationships in my Power BI model for the kinds of design problems that give wrong totals. Mark all Power BI and DAX statements [VERIFY]. Inputs: the tables, with row counts and the columns that act as keys [TABLES_AND_KEYS], every relationship with its columns, cardinality, filter direction and active status [RELATIONSHIP_LIST], the grain of each fact table in words [FACT_GRAINS], and symptoms such as totals that double or categories showing blank [SYMPTOMS]. Return: (1) a relationship review table with columns From, To, Cardinality, Direction, Active, Key unique on the one side? (Yes / No / Unknown), Risk, Suggested change; (2) the model shape: whether it is a star or a snowflake, the facts, the dimensions, and any table acting as both; (3) specific risks: many-to-many relationships, bidirectional filters, two paths between tables, relationships on non-unique keys, blank members from unmatched rows, fact tables joined to each other directly; (4) a diagnostic test list with expected results, for example comparing the distinct count of a key to the row count, showing the arithmetic; (5) for each symptom, ranked hypotheses and one check each; (6) a suggested redesign described in words as a diagram list, to be tried on a copy. Guardrails: do not make changes; propose. Do not assume the key is unique; ask for evidence. Be careful of recommending bidirectional filters; explain the trade-offs. Reconcile any change with source totals before publishing. Add a worked example with invented figures: Customers table with 1,000 rows but only 940 distinct IDs. Show the 60 duplicate rows, 6 percent, and why sales could be double counted for those customers.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.