How to isolate high-risk segments?

Your sender reputation is under pressure. You know the problem is somewhere in your list, but you can't pause everything forever. The goal right now is to cut the risky parts away from the healthy ones so you can keep sending to people who actually want your email while the rest sits in quarantine.

Here's how to do that in practice.

Step 1: Define what "high-risk" actually means

High-risk isn't a gut feeling. It maps to real thresholds that mailbox providers use to judge you. A bounce rate above 2% is a red flag. A spam complaint rate above 0.1% (one complaint per 1,000 emails) is what Gmail and Yahoo Mail use to start throttling or blocking senders. Anything above 0.3% is a crisis. If segments in your list are driving those numbers, they're high-risk by definition.

Address characteristics also signal risk before you've even sent. Unknown or purchased sources, addresses collected before you had a confirmed opt-in process, and addresses from legacy systems you inherited are all candidates for quarantine. So are any addresses that have never opened or clicked a single email from you.

Step 2: Pull the data from your ESP

Most email platforms let you filter contacts by last engagement date, acquisition source, and historical bounce or complaint history. Export that data into a spreadsheet or use your ESP's built-in segmentation tools. The filters you want to run are:

• No opens or clicks in 180+ days. This is your primary inactivity threshold. Some senders use 90 days during a crisis because the stakes are higher than normal.
• Hard bounced at any point. These addresses are dead. They should already be suppressed, but confirm they are. A hard bounce means the address is permanently undeliverable and continuing to send to it actively harms your reputation.
• Sourced from a list purchase, third-party scrape, or unknown origin. You did not get permission from these people. Flag every one of them.
• Previous complaint history. If someone marked you as spam once, the odds of them doing it again are high. Some ESPs track this. Use it.

Step 3: Create three buckets, not two

A simple safe/risky split misses nuance. Try three buckets instead.

• Safe to send. Confirmed opt-in, engaged in the last 30 to 90 days, known acquisition source, no bounce or complaint history. This is your recovery list. Keep sending to these people during your throttled recovery sends.
• Monitor and pause. Last engaged 90 to 180 days ago, acquisition source is known but signup confirmation is unclear, or mild historical issues. Don't send to this group right now. Revisit them in 30 days once your reputation metrics are trending up.
• Suppress now. Inactive for 180+ days, unknown source, hard bounces, complaint history, or purchased. These addresses come off every active campaign immediately. You can decide later whether to try a careful re-engagement or suppress permanently.

Step 4: Set exclusion rules in your ESP, not just a one-time export

But a spreadsheet cleanup is a point-in-time fix. What you want is a live exclusion segment so that as contacts age into riskier territory, they're automatically kept out of future sends. Build a suppression segment or exclusion list in your ESP that refreshes dynamically. That way you're not doing this manually again in three months.

Step 5: Reintroduce cautiously, with proof

Once your core metrics stabilize (complaint rate back below 0.08%, bounce rate below 1.5%), you can think about reintroducing the "monitor and pause" group. Send a single re-engagement campaign to a small test portion, watch the numbers for 48 hours, and only expand if things hold. The "suppress now" group should stay suppressed unless you have a clear plan to validate those addresses before sending again.

If you want help identifying exactly which addresses are salvageable before you try to reintroduce them, we clean lists at RME and you'll get back a clear keep/suppress/monitor split. Take a look at RME Clean if the segmentation is feeling overwhelming.