diff --git a/README.md b/README.md
index 66a3ac99..d5a79599 100644
--- a/README.md
+++ b/README.md
@@ -461,7 +461,7 @@ in isolation.
 We generally recommend testing this uniqueness condition by either:
 * generating a [surrogate_key](#surrogate_key-source) for your model and testing
 the uniqueness of said key, OR
-* passing the `unique` test a coalesce of the columns (as discussed [here](https://docs.getdbt.com/docs/building-a-dbt-project/testing-and-documentation/testing/#testing-expressions)).
+* passing the `unique` test a concatenation of the columns (as discussed [here](https://docs.getdbt.com/docs/building-a-dbt-project/testing-and-documentation/testing/#testing-expressions)).
 
 However, these approaches can become non-perfomant on large data sets, in which
 case we recommend using this test instead.