Data Preparation and Cleansing Tools
The Core Cleansing Set
Select (choose, rename, reorder, retype fields), Formula (create/replace fields), Data Cleansing (trim, remove null rows, replace nulls, change case), Filter (split rows on a condition), Sort, Sample, Unique, and Record ID are the workhorses. Know which tool changes rows versus which changes columns: Select and Formula change schema; Filter, Sample, and Unique change row counts.
Multi-Row and Multi-Field
Multi-Row Formula references other rows (Row-1, Row+1) to compute running totals or deltas; it requires a grouping field to reset per group. Multi-Field Formula applies one expression across many fields using the CurrentField placeholder. Traps ask whether these tools change the record count - Multi-Row Formula does not; it only adds or updates fields.
Imputation and Deduplication
Unique keeps the first record per key combination after an optional sort, so sort order determines which duplicate survives. Data Cleansing can replace nulls with a constant or with a group average via a separate aggregation step.
Worked example: To keep the latest record per customer: Sort by Customer then by Date descending, then Unique on Customer. Reversing the sort keeps the oldest instead - a frequent wrong-answer option.
Rule of thumb: Clean schema first (Select, Formula), then clean rows (Data Cleansing, Filter), then shape (Unique, Sort).