Dynamic segments auto-filter by rules — if the customer no longer matches (e.g. their tag changed), they drop out. That’s by design. For a fixed membership use a static segment.
The email-send / send-step nodes reference templates by name/ID, NOT by snapshot. Editing the template means every node using it picks up the new version on the next render. To freeze the wording, copy the template into a v2.
Reply detection matches on the email In-Reply-To / Message-ID headers. Some clients strip headers → analytics misses it. Workarounds: use a campaign-specific reply-to address, or track via the inbox.
The LinkedIn li_at cookie usually lives 30–90 days, but LinkedIn invalidates it sooner when it sees something suspicious. Re-login in the browser → copy the new cookie → update the credential. Make sure the user-agent matches the browser you logged in with.
Two levels:
  1. Per-customer → customer drawer → toggle “Autoreply bot” off.
  2. Workspace-wide → Settings → AI Models → unset the Chat slot (global flag coming soon).
Emergency stop: revoke the AI credential → the bot can’t call the model → it halts instantly.
Two independent campaigns can overlap their audience and send twice. How to avoid:
  • Add a Path node at the start of campaign B that checks for a tag like campaign:A:active.
  • Or use Exit — Cooldown in campaign A to push the customer into a “currently in campaign A” segment, then exclude that segment from campaign B’s audience.
Each import processes ~50–100 rows/second. 5000 rows ≈ 1–2 minutes. The CSV Imports :id page shows progress in real time. You can keep using the app — no need to wait.
  • A Wait node → waits for the configured duration.
  • A messaging node with a large replyTimeoutMs → waits for that window.
  • Analytics → node detail → “in flight” lists every stuck customer plus where they are.
  • A config error (e.g. dead credential) → the node retries on backoff. Analytics reports “Errors”. Fix the credential and runs resume automatically.
Not required, but their API permissions only refresh on the next request. The UI may need a reload to update the menu correctly.
Not today (AlignUI design system is light-only). On the roadmap.
Other questions? Email support@edgr.ecvhub.net.