← Back to Blog

Cold Email SPAM Trigger Words: 47 Phrases That Kill Your Deliverability in 2026

Published June 10, 2026 — 16 min read

How We Tested This

Before I dump a list of words at you, let me explain the methodology. There is a lot of bad information online about spam trigger words. Some articles list words that were triggers in 2015. Others list words that trigger consumer spam filters but do not affect B2B cold email at all. We wanted real data, 2026 edition.

We tested 50,000 cold emails across three environments:

• Mail-Tester spam score: 15,000 emails, each with one trigger word variant. Measured spam score on a 0-10 scale. Anything below 7 is considered at risk.
• GlockApps inbox placement: 15,000 emails sent to seed lists across Gmail, Outlook, Yahoo, and Apple Mail. Measured inbox placement rate (IPR).
• Real campaign data: 20,000 emails sent to verified Shopify store owner lists over 6 months. We tagged emails containing specific trigger words and tracked deliverability, open rate, and spam complaint rate.

All tests used the same sending infrastructure: 4 dedicated IPs, properly warmed, SPF/DKIM/DMARC configured. The only variable was the email copy. This isolates the trigger word effect from infrastructure issues.

One important note: a single trigger word usually does not tank your deliverability. Spam filters look at combinations. Your email with "free" might land in inbox. Your email with "free + guaranteed + act now + limited time" is going to spam. The more triggers you stack, the faster your score drops.

Category 1: The Sales-y Hype Words (Worst Offenders)

These are the most common and most damaging. They come from a time when email was a direct response channel and every subject line screamed at you. Modern spam filters have been trained on millions of these and they are the fastest way to get flagged.

The biggest lesson here: urgency words are the fastest path to spam. "Act now" alone dropped inbox placement by 22%. Combine it with "free" and "guaranteed" and you are looking at a 40%+ delivery failure. Cold email to business owners does not need manufactured urgency. The offer should stand on its own.

Category 2: The Money Words

Any word that promises wealth, income, or financial transformation triggers spam filters aggressively. These filters have seen enough "make $10K/month from home" emails to last several lifetimes.

Notice a pattern here: B2B cold email should talk about business outcomes, not personal wealth. "Increase revenue" lands in inbox. "Make money" lands in spam. The filter sees "make money" and assumes you are selling a course, not a legitimate B2B service.

Category 3: The Clickbait Subject Line Words

These are words that scream "I am marketing email and I do not care about you." Subject lines containing these words get deleted without being opened -- if they even reach the inbox.

The fake RE:/FWD: trick is the single most damaging thing you can do to a cold email subject line. It dropped inbox placement by 24% in our tests. Beyond the spam filter issue, it damages trust. The moment someone realizes you faked a reply thread, you have burned that contact permanently. Do not do it.

Category 4: The Body Copy Killers

These are phrases that appear in email body copy and trigger filters even when the subject line is clean. Body copy triggers are less aggressive than subject line triggers, but they accumulate. Five body triggers in a 200-word email can push you over the threshold.

The phrase "this is not spam" is a special case of stupid. It decreased spam score by 1.6 points on its own. Spam filters have been trained to associate this phrase with... spam. Saying "this is not spam" in an email is like saying "I am not a liar" on a first date. It does not help your case.

Category 5: The Formatting Triggers (Non-Text)

Spam filters do not just read your words. They analyze your HTML structure, image-to-text ratio, link patterns, and code quality. These formatting triggers often fly under the radar because people focus exclusively on word choice.

Image-only emails are the worst formatting offender by a wide margin. A 3.1-point spam score drop is catastrophic. This is why you should never send a "designed" HTML email template for cold outreach. Plain text, or minimal HTML with a 60/40 text-to-HTML ratio, is still the standard for cold email in 2026.

Link shorteners are a close second. Bit.ly, TinyURL, and similar services are heavily used by phishers and spammers. Using one signals "I have something to hide about where this link goes." Always use your full domain URL.

What Actually Works: A Practical Rewrite Example

Here is a before-and-after of a real cold email we tested. The "before" version used 11 trigger words. The "after" version used zero. Same offer, same audience, same sending infrastructure.

The difference is not just word choice. The "after" version does four things right: (1) it mentions something specific about the recipient's business, (2) it leads with a genuine observation, not a sales pitch, (3) it asks a simple question instead of making demands, and (4) it reads like a human wrote it to another human.

The "before" version sounds like a template that got blasted to 10,000 people. Because it is.

The Complete 47-Word Checklist

Here is the full list of 47 trigger words and phrases, ranked by impact. Run your cold email copy against this list before every send. If you find more than three, rewrite.

The Stacking Effect: Why 3 Triggers Is Worse Than 1+1+1

A single LOW severity trigger word in your email body will probably not affect deliverability at all. A single HIGH severity trigger in your subject line might drop inbox placement by 10-15%. But the stacking effect is not additive. It is exponential.

In our tests, emails with zero triggers had an average IPR of 94%. Emails with one trigger: 89%. Two triggers: 79%. Three triggers: 61%. Four triggers: 38%. Five or more triggers: 22% or lower. The filter does not care that each individual word is minor. It sees the pattern and says "this is marketing spam."

This is why I recommend a hard rule: zero trigger words in every cold email you send. Not "try to avoid." Zero. It is not that hard once you internalize the replacements, and the deliverability difference is the difference between a campaign that works and one that does not.

How to Test Your Copy Before Sending

Before you send any cold email campaign, run your copy through these three checks:

1. Mail-Tester.com -- Send your email to the address they provide, get a spam score out of 10. Aim for 9.0+. If you score below 7.5, rewrite.
2. GlockApps Inbox Placement Test -- Send to their seed list and see exactly which inbox providers deliver to inbox vs spam. Worth the $59/month if you send volume.
3. Manual trigger word check -- Copy-paste your email into a text file and search for every word on the 47-word list above. Count the hits. If more than zero, rewrite until zero.

I also recommend sending a test batch of 100-200 emails to a small segment of your list before running the full campaign. Track deliverability, open rate, and spam complaint rate on the test batch. If anything looks off, fix the copy before scaling.

The Relationship Between Copy Quality and List Quality

One thing I need to mention: even perfect copy cannot save a bad list. If your email addresses are unverified, outdated, or scraped from public pages, your bounce rate will be high. High bounce rate damages your sender reputation regardless of how clean your copy is.

Spam filters evaluate your email on two dimensions: content quality (what you say) and sender reputation (who you are). Content quality is the copy. Sender reputation is driven by bounce rate, spam complaint rate, engagement rate, and domain/IP history. List quality affects sender reputation directly.

This is why using a verified email list matters as much as clean copy. If you buy a list of Shopify store owner emails from a provider that verifies every address before delivery, your bounce rate stays under 2%. Low bounce = good sender reputation = even decent copy lands in inbox. If you scrape unverified addresses from the internet, your bounce rate hits 15-30%. No amount of clean copy can fix that.

Last updated: June 10, 2026. All deliverability data from B2BRepurpose campaigns run January 2025 through May 2026. Test methodology described in detail above. Individual results will vary based on sending infrastructure, list quality, and offer relevance.

← Back to Blog