diff --git a/README.md b/README.md
index f4a902b..6664559 100644
--- a/README.md
+++ b/README.md
@@ -71,10 +71,16 @@ function getCachedUser(id, cb) {
 Second, node-cache-manager features a built-in memory cache (using [node-lru-cache](https://github.com/isaacs/node-lru-cache)),
 with the standard functions you'd expect in most caches:
 
-    set(key, val, ttl, cb)
+    set(key, val, {ttl: ttl}, cb) // * see note below
     get(key, cb)
     del(key, cb)
 
+    // * Note that depending on the underlying store, you may be able to pass the
+    // ttl as the third param, like this:
+    set(key, val, ttl, cb)
+    // ... or pass no ttl at all:
+    set(key, val, cb)
+
 Third, node-cache-manager lets you set up a tiered cache strategy.  This may be of
 limited use in most cases, but imagine a scenario where you expect tons of
 traffic, and don't want to hit your primary cache (like Redis) for every request.