Spam score

Definition
A number estimating how likely a given email is to land in the junk folder instead of the inbox.

Why it matters

A message that lands in junk looks the same as one the recipient ignored. A low reply rate then gets blamed on the offer, the subject line or the list, when the real cause is that nobody saw the email.

Most causes are quick to fix: authentication, wording and list quality. Checking before a large send is one of the cheapest ways to protect results, because a bad send also harms the sender's domain reputation, which affects every later email.

The score itself is an estimate from one checking tool, not what Gmail or Outlook will do. Use it to catch obvious problems before sending, and use real placement tests and reply rates to confirm.

How to apply it

  • Set up SPF, DKIM and DMARC for the sending domain. Gmail and Yahoo require them from bulk senders, and missing records are a common cause of poor placement.
  • Run each template through a spam checker and remove flagged patterns, such as heavy capital letters, many links or pushy phrases.
  • Keep complaint rates very low. Google asks bulk senders to stay under 0.3 per cent and aim well below it.
  • Offer an easy unsubscribe link.
  • Remove people who have not engaged for months. See List hygiene.
  • Warm up a new domain gradually before raising volume.

What it is

Mailbox providers such as Gmail and Outlook decide where each message goes. They do not publish a single score. Spam filters and checking tools do produce one, for example by adding points for suspicious wording, broken formatting or missing authentication. The open-source filter SpamAssassin, for instance, marks a message as spam once its points pass a set threshold. A score from a checking tool is therefore an estimate, useful as a warning light and not a verdict.

SEO tools also use "spam score" for something different: how risky a website looks to search engines. This entry covers email.

Common mistakes

  • Treating the score as the goal. A clean score on a bad list still fails.
  • Buying lists, which brings spam traps and complaints.
Worked example

Suppose a small recruitment firm sends an outreach campaign of 600 emails from a newly registered domain. Replies come in at a third of the usual level, and the team assumes the offer is weak. A spam checker run on the template finds a missing DMARC record, a run of capitals in the subject line and three links in the first sentence. The team fixes the records and the wording, cuts the links and holds the volume back while the domain builds a record.

The new mailboxes are set up with Lemwarm, which warms them up gradually and monitors sender reputation before the volume rises. The second campaign goes to a smaller, cleaner list with an easy unsubscribe link, and replies return to the usual level within two weeks. Checking the template took under an hour, which is far cheaper than a month of sending an offer that never reached the inbox.

Tools in the example

Some links are affiliate links: we may earn a commission at no cost to you. It never decides a ranking. How we work with partners

  1. Article

    Email deliverability

    The outcome the score protects.

  2. Article

    Domain reputation

    The longer-term trust record behind it.

  3. Article

    Warm-up

    Lowers the risk on a new domain.

  4. Article

    Reply rate

    The metric that drops when mail is filtered.