ESP-Aware Routing and Sending Health Gate

Why Does ESP-to-ESP Matching Matter?

Gmail and Microsoft filter inbound email differently, and they treat cross-ecosystem sends with extra suspicion. A Gmail-hosted sending domain with a perfect reputation inside Google's ecosystem often underperforms when sending to Microsoft 365 recipients — and vice versa. The same mailbox can land 80% in primary inbox for Gmail recipients and 30% for Outlook recipients on the same day.

Most sequencers ignore this and push sends round-robin across all connected mailboxes, which means every campaign is leaking deliverability on the cross-ESP half of its list. Superkabe instead classifies each lead's ESP at ingest time and routes the send through the mailbox with the strongest recent performance against that specific ESP.

How Does the Health Gate Decide What to Send?

Every mailbox in your fleet carries a live health score derived from bounce rate, deferral rate, reply rate, and recent complaint signals. The health gate runs as a pre-send check: before any message is handed to the SMTP layer, the gate verifies that the assigned mailbox is currently GREEN for the target ESP.

If the best-matched mailbox is YELLOW, the send is queued briefly and retried against the next-best candidate. If no mailbox is currently GREEN for that recipient's ESP, the lead is held in the routing queue rather than forcing a send through a compromised sender. This is the difference between "send now and hope" and "send only when the match is safe."

The health gate integrates with Superkabe's 5-phase recovery state machine. A mailbox in warning, paused, quarantine, restricted-send, or warm-recovery state is automatically removed from the routing pool for the affected ESP until the recovery protocol promotes it back to healthy.

What Does GREEN / YELLOW / RED Scoring Actually Measure?

GREEN means the mailbox has a bounce rate under the warning threshold, a reply rate within normal range, and no recent ISP-side negative signals against the recipient's ESP. YELLOW means one or more metrics are drifting — not enough to pause the mailbox, but enough to deprioritize it in the routing queue. RED means the mailbox has breached a safety threshold and has been automatically removed from active sending.

Scoring is recomputed continuously as new send telemetry arrives. A mailbox that handled 500 Gmail recipients cleanly this morning and then produced 3 bounces to Outlook recipients this afternoon will be GREEN for Gmail and YELLOW for Outlook at the same time. Routing decisions use the ESP-specific score, not a global average.

How Does This Change the Economics of Outbound?

Teams that switch to ESP-aware routing typically see 15–25% higher inbox placement on cross-ESP sends without adding a single mailbox. The volume you already have starts performing closer to its theoretical ceiling because the right sender is chosen for every recipient. Combined with the health gate holding risky sends, your overall bounce rate drops while your inbox placement rises — the two metrics that most directly govern how much pipeline your domains produce.