What KPIs prove that recovery is stable?

Still have a question, spotted an error, or have a better explanation or a source we should cite?

You've done the hard work. Complaints are down, inbox placement is climbing, and things feel better. But "feels better" isn't the same as "stable." Before you start scaling back up or relaxing your sending discipline, you need the numbers to back it up over time, not just one good week.

Here's how to read them.

Postmaster signals (your most direct window)

Gmail Postmaster Tools is the closest thing to an official reputation report card. You're looking for your domain reputation to hold at Medium or High for at least four consecutive weeks with no dips into Low or Bad. One good week doesn't count. Four stable weeks with zero yellow or red periods is the floor. Microsoft's Microsoft 365 Smart Network Data Services (SNDS) should show a consistently green status across your sending IPs for the same window.

If you see reputation bounce between Medium and Low, or SNDS flicker between green and yellow, recovery isn't stable yet. Variance matters as much as the average.

Engagement metrics (what real people are doing)

Stable doesn't mean improving. It means not declining. You want open rates and click rates that are consistent week over week, not spiking one send and tanking the next. Complaint rates should be sitting below 0.1% per send. If you're seeing complaints creep above that threshold, something in your list or content is still off (and mailbox providers are definitely noticing).

Watch the trend line, not the individual sends. A slow, steady open rate is far healthier than a volatile one that averages out to the same number.

Delivery metrics (the structural checks)

Your bounce rate should be stable and low, with no policy bounces showing up. Policy bounces are a signal that a mailbox provider is actively rejecting your mail at the policy level, not just filtering it. That's a red flag even if overall delivery looks fine. Inbox placement (measured through seed testing or placement tools) should be consistently landing where it belongs, not shifting between inbox and spam folder across sends.

The absence signals (the quiet ones that matter)

No blocklist listings. No delivery emergencies. No inbox-related escalations coming from your team or customers. These sound obvious, but they're worth checking systematically. You can run a quick blocklist check with our free blocklist checker to confirm nothing's crept up quietly.

What "stable" actually means in practice

But a rough working definition: all of the above metrics, held steady across a minimum of four weeks, with no individual send triggering a reversal. That's the signal that reputation has genuinely rebuilt, not just temporarily recovered after a single clean campaign.

Once you're there, the next question is usually how to avoid slipping back once you start scaling again. And setting up the right alerts means you'll catch any wobble before it becomes a crisis.

Not sure if your numbers are telling you what you think they are? Drop into our SOS hotline and we'll take a look with you. Free, no pitch, just eyes on your actual data.

Contributors

Who worked on this answer

Every name links to their profile. Every company links to their site. Real people, real accountability.

Ask an AI · tailored to your setup

Analyze my recovery KPIs

I'm recovering my domain's email reputation and I want to know when it's truly stable. Based on my situation below, tell me: which of these KPIs am I missing or misreading, what thresholds should I be watching, and whether my numbers suggest I'm ready to scale back up or should keep holding. My current metrics: [paste Gmail Postmaster reputation level and how many weeks it's held], complaint rate, bounce rate and any policy bounces, SNDS status, inbox placement % if known.

Edit the yellow boxes, then send to the AI of your choice.