The table has one identifier, six categorical dimensions, and nine numeric metrics, but only four of those metrics are stored as numbers. The only date/time column is a pipeline timestamp. This classification comes from the column profiles (`inspect_columns`) on `raw.AI_Usage_and_Impact_on_Students_and_Professionals`. **Identifier** - `User_ID` (VARCHAR) is a near-key, with about 287 distinct values across 300 rows, ranging from U0001 to U0288. It is not perfectly unique, so some IDs repeat. **Categorical dimensions** - `User_Type` has 2 values (Professional, Student). - `Gender` has about 12 distinct values, which suggests inconsistent spellings or casing. - `Education_Level` has about 17 distinct values, also likely inconsistent labels. - `AI_Tool` has about 16 values. - `AI_Purpose` has about 7 values. - `Would_Recommend` has about 4 values, with Yes among them. - `Country` has about 20 values and `Profession` about 110. The profiler tags both as "spatial", but they are really geographic and occupational categories. - `Tasks_Performed` is free text, with about 288 distinct values. It is descriptive rather than a usable dimension. **Numeric metrics** - Stored as numbers: - `Productivity_Score` (DOUBLE) ranges from 1.6 to 10.0 and has 16 nulls. - `Accuracy_Rating` (BIGINT) ranges from 2 to 5 and has 15 nulls. It behaves like an ordinal rating. - `Satisfaction_Score` (BIGINT) ranges from 3 to 10 and has 7 nulls. - Stored as text (VARCHAR) and needing cleaning before analysis: - `Age` - `Monthly_Income`, which is personally identifying, so aggregate it only. - `Monthly_AI_Cost` - `AI_Usage_Hours_Per_Day`, whose values include unit text such as "9.9 hrs". - `Time_Saved_Hours_Per_Week` - `Work_or_Study_Hours_Per_Day` **Dates and times** - There is no business date, such as a survey date or signup date, so time-trend analysis is not possible. - `_ingestion_timestamp` (TIMESTAMP) is the only temporal column. It holds a single value (2026-10-02 09:58:27), so it only records when the data was loaded. **Pipeline metadata, not analytical** - `_batch_id`, `_source_file` and `_source_system` each have one distinct value and can be ignored.