diff --git a/TOOLING.md b/TOOLING.md
index dfc1e89..f5ba2ae 100644
--- a/TOOLING.md
+++ b/TOOLING.md
@@ -332,7 +332,7 @@ the old names with a pointer rather than a bare error.
 | mode | cadence | cost | why |
 |---|---|---|---|
 | `stories` | daily | ~6 requests | stories expire in 24h and **cannot be backfilled**; this is the only run that loses content if skipped |
-| `profiles` | monthly | ~40-60 requests | every surface, `--abort 50` — stops enumerating a profile once it reaches content already held. Catches everything **new** |
+| `profiles` | weekly, randomized day | ~40-60 requests | every surface, `--abort 50` — stops enumerating a profile once it reaches content already held. Catches everything **new** |
 | `full-sweep` | rarely, by hand | **~420 requests** | no abort; walks every profile to the end. The only run that notices posts **edited** after we archived them (test case 15) |
 
 The skip-archive means an infrequent `profiles` run costs barely more than a frequent
diff --git a/docs/gallery-dl.md b/docs/gallery-dl.md
index b389ee5..7380eff 100644
--- a/docs/gallery-dl.md
+++ b/docs/gallery-dl.md
@@ -358,7 +358,7 @@ means an infrequent full sync costs barely more than a frequent one, because it
 only fetches what is new.
 
 ```
-# monthly, everything
+# weekly, everything
 gdl-sync.py --index <viewer-url> --staging ~/gdl/staging \
     --publish <user>@<nas>:<archives> --archive-db ~/gdl/artms.db \
     --urls-file artms_account_links.txt --execute
diff --git a/scripts/gdl-cron.sh b/scripts/gdl-cron.sh
index a73cc5a..20dc081 100755
--- a/scripts/gdl-cron.sh
+++ b/scripts/gdl-cron.sh
@@ -2,7 +2,7 @@
 # Unattended wrapper around gdl-sync.py. One argument: the run mode.
 #
 #   stories     daily      ~6 requests; the only surface that cannot be backfilled
-#   profiles    monthly    every surface, --abort 50: stops enumerating a profile
+#   profiles    weekly     every surface, --abort 50: stops enumerating a profile
 #                          once it reaches content already held, so it costs
 #                          ~40-60 requests and catches everything NEW
 #   full-sweep  rarely     every surface, no abort: walks each profile to the end
diff --git a/scripts/systemd/gdl-sync@profiles.timer b/scripts/systemd/gdl-sync@profiles.timer
index 131573c..e085a2a 100644
--- a/scripts/systemd/gdl-sync@profiles.timer
+++ b/scripts/systemd/gdl-sync@profiles.timer
@@ -1,9 +1,17 @@
 [Unit]
-Description=Monthly Instagram profile sync (all surfaces, --abort 50)
+Description=Weekly Instagram profile sync (all surfaces, --abort 50)
 
 [Timer]
-OnCalendar=*-*-03 04:00:00
-RandomizedDelaySec=45m
+# Weekly, but deliberately not pinned to a day: OnCalendar anchors to Monday
+# 00:00:00 and RandomizedDelaySec spreads the actual fire up to 6 days later,
+# so which day it lands on varies week to week -- same "don't look like a
+# machine" reasoning as the stories timer's 45m jitter, just applied to the
+# day instead of the minute. See TOOLING.md's healthchecks.io section: the
+# matching instagram-archiver-posts check uses Simple Schedule (period +
+# grace) rather than Cron Schedule for exactly this reason -- a fixed cron
+# day/time would fight this jitter instead of tolerating it.
+OnCalendar=weekly
+RandomizedDelaySec=6d
 Persistent=true
 
 [Install]