Spam

Spam is unsolicited electronic communication sent in bulk - email, SMS, social DMs, comments, push notifications - that the recipient didn’t request and generally doesn’t want. In marketing contexts, spam is both a legal problem (CAN-SPAM, GDPR, CASL, and similar legislation all regulate it) and a deliverability problem (ISPs and platforms increasingly filter any messaging that looks spam-adjacent, even legitimate marketing mail caught in the net). The working definition for a marketer is narrower than the legal one: any message the recipient would consider unwanted counts, whether or not it’s technically compliant.

The four common spam categories

Unsolicited commercial email. The classic. Sent to purchased or scraped lists, no prior relationship, generally pushing a product or offer. Most targeted by anti-spam law.

Phishing and fraud. Messages impersonating trusted brands to extract credentials, payment info, or other sensitive data. A criminal subset of spam, addressed by different legal frameworks.

Comment and forum spam. Bot-posted links in blog comments, forum threads, and review sections. Mostly aimed at gaming link signals for SEO; mostly ignored by modern algorithms, which is why it’s declining.

Legitimate-but-unwanted marketing. Real brands sending to real subscribers who don’t remember opting in, sending too frequently, or sending irrelevant content. Not legally spam, but treated as such by recipients and increasingly by spam filters. The category marketers have the most control over.

The legal landscape (simplified)

Three jurisdictions worth understanding:

CAN-SPAM (US, 2003). Allows unsolicited commercial email under strict conditions: accurate sender info, non-deceptive subject lines, clear unsubscribe, physical mailing address. Opt-in is not required. Relatively permissive.

GDPR and PECR (EU/UK). Requires explicit prior consent for most commercial electronic messaging. Opt-out is not enough; recipients must actively opt in. Fines up to 4% of global turnover.

CASL (Canada). Similar to GDPR in requiring express consent. Strict in enforcement. Penalties can reach $10M per corporate violation.

Operating globally means complying with the strictest applicable regime, which in practice means explicit opt-in for anyone you can’t verify is outside EU/UK/Canada jurisdiction. “Legal” spam in one market can be illegal in another.

What makes legitimate email get classified as spam

Four technical signals filters use:

Sender reputation. Accumulated score of your sending domain and IP, based on bounce rates, complaint rates, spam-trap hits, and engagement patterns. Low reputation = filter to spam folder regardless of content quality.

Content signals. Spammy phrases, excessive exclamation marks, image-heavy/text-light emails, URL patterns associated with known spam. Content filtering is softer than it was a decade ago but still active.

Authentication compliance. SPF, DKIM, and DMARC records. Mail that fails authentication is treated with heavy suspicion; mail that passes looks legitimate to the receiving server.

Engagement patterns. Low open rates and high delete-without-read rates degrade reputation over time. Inboxes use these signals to filter sends that subscribers technically receive but don’t read.

How to not be spam-classified when you’re not spamming

Six disciplined moves:

Use double opt-in. Confirming opt-in via a follow-up email ensures the subscriber actually wanted the list. Protects against typos, hostile third-party signups, and list-quality decay.

Respect unsubscribes immediately. Not within 10 days (CAN-SPAM allowance), not within 48 hours - within minutes. Complaint rates compound.

Monitor soft-bounce and hard-bounce rates. Rising bounce rates signal list-quality issues that affect reputation.

Segment and personalise. A 20% open rate on a segmented send signals a healthier relationship than a 5% open rate on a blasted send, even if the absolute numbers are smaller.

Authenticate with SPF, DKIM, DMARC. Non-negotiable in 2026. Gmail and Yahoo now require full authentication for bulk senders, and non-compliant senders see inbox placement collapse.

Prune inactive subscribers. Subscribers who haven’t engaged in 6–12 months are hurting deliverability. A re-engagement campaign, followed by removal of the still-unengaged, improves overall list performance.

Comment and forum spam

For site owners, two defensive moves:

Moderation and CAPTCHA. Stops most bot-posted comment spam. Modern invisible reCAPTCHA adds minimal friction for legitimate users.

nofollow or ugc attributes on user-generated links. Strip ranking-signal value from comment and forum links, removing the incentive for spammers to target the site.

See email marketing for the broader context in which email spam avoidance sits.

Related terms

• Email Marketing - the channel where spam discipline matters most
• Soft Bounce - bounce rate directly affects spam classification
• Hard Bounce - bounce-driven reputation damage is common
• Subscribers - list health is the root of spam-filter performance
• Lead Generation - bad lead-capture practices produce spam-likely lists