What actions count as positive signals (opens, clicks, replies, moves to inbox)?

Not all positive signals are created equal. Filters weigh your subscribers' actions differently, and some of what you're measuring in your ESP dashboard is not quite what you think it is.

Here's how the main positive signals stack up, roughly from strongest to weakest:

Replies are the gold standard. A reply is a human decision that no proxy server or bot can fake reliably. When someone replies to your email, Gmail, Outlook, and most other providers treat that as a very strong vote of trust. It says: this person wanted a real conversation. Filters prioritize senders who get replies, and that signal is personal to the individual inbox, not just your aggregate reputation.

Moving from spam to inbox is explicit feedback. When someone opens the spam folder and moves your email to their inbox (or clicks "Not Spam"), they're directly telling the filter it got it wrong. That correction carries real weight, especially at Gmail, which is known to factor individual user corrections into future filtering for that sender.

Adding to contacts or address book is a declaration of trust. It tells the provider's system that this sender belongs here. It also often unlocks image loading by default, which is a quiet but meaningful signal of an established relationship.

Clicks are strong, but they're not all equal either. A click on a real link you cared about is very different from an automated security scanner pre-clicking every URL in your email. Some corporate mail environments, particularly those running Microsoft 365 with Safe Links enabled, will click your links before the human even reads the message. Your ESP may count those as clicks. The filter knows the difference.

Opens are the trickiest signal of all. Since Apple Mail launched Mail Privacy Protection (MPP) in 2021, every Apple Mail user on iOS 15+ appears to open every email you send them. That's not a real open. It's Apple's proxy server pre-fetching your tracking pixel before the recipient decides anything. Gmail and Outlook process opens through their own caching layers too, though less aggressively. The point is: your ESP's open rate is increasingly a proxy for delivery, not genuine interest.

Operationally, this matters a lot. If you're segmenting your list by "opened in the last 90 days" and a big chunk of your list is Apple Mail users, you may be keeping people active who haven't actually engaged in months. Your sender reputation is built on real signals, and inflated open data can mask a real engagement problem until it's too late.

Forwarding and starring are positive, but minor. They're nice to see, but they won't move the needle on reputation the way replies and rescues from spam will.

The practical takeaway: design your emails to earn replies and clicks from real humans. Opens are still worth tracking as a directional metric, but treat them with skepticism, especially if Apple Mail is a big slice of your audience. And if you have subscribers who haven't done anything clickable in six months, they're probably not helping your reputation even if your ESP says they're opening everything.

Not sure how healthy your engagement signals actually look? You can run a quick check with our free Email Header Analyzer to see what filters are actually seeing, or reach out on our SOS hotline if you're worried about a reputation drop.