diff --git a/crates/atuin-server-postgres/migrations/20240614104159_idx-cache.sql b/crates/atuin-server-postgres/migrations/20240614104159_idx-cache.sql
new file mode 100644
index 00000000000..af70bff7e72
--- /dev/null
+++ b/crates/atuin-server-postgres/migrations/20240614104159_idx-cache.sql
@@ -0,0 +1,43 @@
+-- using group by and max is slow. as the store grows, latency is creeping up.
+-- get a handle on it!
+-- Really this sort of workload (lots of aggregation) is better suited to OLAP,
+-- but let's just stick with postgres until we can't any more
+
+create table store_idx_cache(
+	id bigserial primary key, 
+	user_id bigint,
+
+  host uuid,
+  tag text,
+  idx bigint
+);
+
+
+create or replace function cache_store_idx()
+returns trigger as 
+$func$
+begin
+	if (TG_OP='INSERT') then
+		update store_idx_cache set idx = (select max(idx) from store where user_id=new.user_id and host=new.host and tag=new.tag) where user_id=new.user_id and host=new.host and tag=new.tag;
+
+		if not found then
+			insert into store_idx_cache(user_id, host, tag, idx) 
+			values (
+				new.user_id, 
+        new.host,
+        new.tag,
+				(select max(idx) from store where user_id = new.user_id and host=new.host and tag=new.tag)
+			);
+		end if;
+	end if;
+
+	return NEW; -- this is actually ignored for an after trigger, but oh well
+end;
+$func$
+language plpgsql volatile -- pldfplplpflh
+cost 100; -- default value
+
+create or replace trigger tg_cache_store_idx
+	after insert on store
+	for each row 
+	execute procedure cache_store_idx();