Configuring
Split output
The Joiner can emit three separate tables rather than one: matched, left-unmatched, right-unmatched.
That gives you the same diagnostic Alteryx and Prep provide by default — you can see exactly which keys failed to match instead of noticing later that a total looks low.
Row multiplication
If the right table has duplicate keys, each left row matches several times and the output grows.
Check the row count. If it is higher than the left input and you did not expect that, add a GroupBy on the right branch first, grouping by the join key so there is one row per key.
Same fix as pre-aggregating in SQL, groupby before merge in pandas, Summarize before Join in Alteryx, or an aggregate step in Prep.
Duplicate column names
When both tables have a column of the same name, the dialog offers to append a suffix, or to filter one out. Set it deliberately — the default suffixes make downstream references confusing.
Joiner versus Cell Replacer
For a simple lookup — replacing a code with its description from a small table — the Cell Replacer node is lighter and clearer than a full join. It takes a dictionary table and swaps values in place.