How do spam filters decide if an email is spam?

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

You hit send. Somewhere between your server and your recipient's inbox, a decision gets made. Was that a legitimate email or spam? Here's how that decision actually happens.

Spam filters don't flip a single switch. They run a weighted calculation across multiple signals, and the combined score determines whether your email lands in the inbox, the spam folder, or gets rejected outright. No single factor wins or loses the game on its own.

What gets checked before your content is even read

The first checks happen at the SMTP connection, before your message body arrives. Filters look at your sending IP's history, whether it's on any blocklists, and whether your SPF record and DKIM signature pass. If these fail, many filters reject the message before scanning a single word. You never even get to the content stage.

This is why authentication matters so much. A domain that fails SPF or sends without DKIM starts the scoring process already in the red.

What gets checked once the message arrives

Now once the message is accepted for inspection, filters scan the content itself. They look for known spam phrase patterns, suspicious or mismatched links, unusual HTML structure, and poor text-to-image ratios. They also check whether your links resolve to domains with bad reputations.

Beyond content, filters look at your sender reputation. This is built from your IP address history, your domain's history, and how recipients have interacted with your mail over time. High complaint rates, low open rates, and frequent deletions without reading all push the score toward spam. Replies, clicks, and moves to primary inbox pull it the other way.

The engagement layer most senders underestimate

Mailbox providers like Gmail and Outlook pay close attention to what recipients do after they receive your mail. Not just whether they open it, but whether they delete it immediately, mark it as spam, click through, or even reply. This engagement history shapes how your future emails are scored. It's one reason a sudden burst of complaints can damage deliverability for weeks afterward.

How to work with the scoring, not against it

The practical takeaway is that spam filtering rewards consistency. Authenticate your domain properly. Send to people who actually want your emails. Keep complaint rates low by making unsubscribes easy to find. Clean your list regularly so you're not mailing addresses that no longer exist or engage.

If you're not sure whether your authentication setup is passing those early checks, you can run your domain through our free SPF Checker or DKIM Record Lookup in seconds. And if something feels broken, our SOS hotline is free.

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

Get a personalized filter-score breakdown

I want to understand how spam filters score my emails so I can fix what's holding me back. Here's my situation: [describe your sending setup, ESP, domain age, recent bounce or complaint trends, and any authentication already in place]. Walk me through which signals are most likely hurting my score and what I should prioritize fixing first.

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