What spintax is, and when it helps
Spintax is a way of writing several options for one phrase so each send picks one of them. It is a cheap hygiene habit with three real uses, and it is not the reason your mail is or is not landing. Seven tools write it seven different ways, and the table is first.
By Rahul Bageria, co-founder · Updated August 2026 · 12 min read
Seven tools, seven syntaxes
Almost every page on this topic teaches one syntax and calls it the standard. Read from the tools themselves, and no two of these seven agree.
| Tool | How you write it | What its own docs add |
|---|---|---|
| Instantly | {{RANDOM | Hi | Hello | Hey}} | Variables nest inside a spin, and fallbacks chain: {{firstName | lastName | for you}}. |
| Smartlead | {Hi|Hello|Hey} | The classic single brace. Its guide puts spin tags in the greeting and the call to action. |
| Woodpecker | {{SPINTAX | "Hi" | "Hello"}} | Values go in quotes. Works in the subject line. Snippet fallbacks fail inside the formula. |
| lemlist | {% spin %} {% variation %} Hi {% variation %} Hello {% endspin %} | Not spintax at all, but Liquid spin tags. No indentation inside the syntax. |
| Saleshandy | {spin} Hi | Hello | Hey {endspin} | There is a button in the editor. Its own advice is three to four spin tags per email. |
| Salesforge | #{{Hi|Hello|Hey}} | The hash prefix is part of it. Works in the body and in the subject line. |
| GMass | {{spin}}Hi{{variation}}Hello{{end spin}} | It picks options in order, not at random, until you turn on an account-wide setting. |
Sources, each read from the tool's own documentation in August 2026: Instantly (July 2026); Smartlead; Woodpecker (December 2024); lemlist; Saleshandy (modified August 2026); Salesforge; GMass (February 2026). Apollo's knowledge base is closed to outside readers, so it is out of the table either way.
Your tool decides the syntax, and a good deal else besides, which is why we keep a running roundup of the best cold email tools.
What spintax is, and where it came from
Two short answers, and the second explains why the word makes deliverability people wince.
Spintax is a way of writing several options for one phrase so each send picks one of them. Its classic form is {cat|feline|kitty}: three words in one slot, one chosen per message, and the braces never reach the reader.
The idea came from search, not email. Plagiarism Today's history of article spinning dates the first tool that reached a wide audience to 2004, and Google's Panda update in February 2011 ended it as a way to rank.
Email inherited it on the way out, and the same history records email spammers picking it up to get past filters. That is the whole provenance.
Spintax works on the last thing a filter looks at
A filter asks three questions in order. Spintax touches only the third, and the third is the weakest of them.
Can it tell who sent this?
Authentication, a real domain, a sending history it can attach to you. Nothing you write in the email moves this answer, and it is checked first.
What happened last time you sent?
Complaints, deletions, replies. This layer settles most outcomes, and it is a running record of how people treat your mail, not of how you word it.
And only now, the message itself
Links, hidden content, the shape of the message. Spintax lives here, which is why it cannot rescue a problem sitting two layers above it. The ranked triage is in why cold emails go to spam.
Every founder who has asked me about spintax had a problem in the first two layers, and none of them had a spintax problem. Turn the spin tags off in your greeting, then go and check whether your domain authenticates. Same afternoon, completely different outcome.
The thing spintax is meant to beat was built to ignore it
Spintax exists to defeat bulk-mail detection. Bulk-mail detection has published its answer to that for twenty-five years.
The oldest of these systems works on a checksum, which is a short fingerprint of the message body, deliberately built to stay the same when small things change. Rotated text whose only job is to change that fingerprint has a name in that world: a hash buster.
That is exactly what spintax is, and the system answers the question by name in its own FAQ. Here is what each layer of the machinery actually reads, and whether spinning your words moves any of it.
Built to survive your variations
Distributed Checksum Clearinghouses, in production since the winter of 2000, says its checksums stay constant across common variations in bulk messages, personalizations included. Its FAQ adds that they ignore many personalizations and hash busters. Spinning does not move this.
No rule about repeated content exists
Google's Email sender guidelines cover authentication, spam-complaint rate, message format and one-click unsubscribe. Read end to end in August 2026, they say nothing about identical, duplicate or templated content. There is no rule here for spinning to move.
A complaint grade, not a text grade
Microsoft's bulk complaint level runs 0 to 9, and its own definitions describe each band by how many complaints a sender generates. What people do with your mail is the graded thing. Word choice is not.
Every bulk check at once still falls short
Apache SpamAssassin publishes its rule scores. Its three collaborative bulk-mail checks, added together, land under its own spam threshold, so being caught by all of them at once does not condemn a message. It is open-source scoring, not what Gmail or Outlook run.
None of this makes spintax harmful. It makes it small. The layer that decides whether your mail lands is reputation, and holding it is the weekly discipline in our cold email deliverability guide.
Where variation genuinely earns its place
Three jobs, none of them about spam filters. Each takes about ten minutes to set up and then never needs touching again.
"Hi ," is a worse email than any amount of repeated phrasing, and it is the most common visible defect in cold outbound. In most tools the fallback lives in the same syntax family as the spin tag, which is the part actually worth setting up.
When you write to several people at one account, they forward to each other. Three word-for-word identical notes is a credibility problem long before it is a deliverability one. lemlist's docs name this use, and make no deliverability claim while doing it.
If every message in your sent folder is the same message, you cannot review your own outbound. Small, unglamorous, and the only benefit on this list that survives being looked at closely.
Want your outbound fixed at the layer that actually decides it?
Book a Fit CheckSpinning is not testing, and nothing tells you which won
The platform files a reply against the prospect, not against the wording. So the more of the email you spin, the less you learn.
Saleshandy's own FAQ says it plainly: it tracks opens and replies at the prospect level, not at the text level. That is reassurance about tidy reporting, and a wall in front of learning anything.
Five spin tags of three options each is 243 versions of one email. You wrote one and read one, and all 243 report a single reply rate. GMass makes the trade explicit: turn on A/B testing and it overrides randomization, sorting your options into test groups instead.
Spin the ask
{spin} Worth a quick call? | Open to 15 minutes? | Shall we find a time? {endspin}
- ✕Three asks, one reply rate
- ✕No version is ever credited
- ✕You never chose the one that sent
Ship one ask, test the next
Worth a quick call next week? Then run "Open to 15 minutes?" as step two's variant.
- ✓One sentence, one meaning
- ✓The variant is tracked and reported
- ✓You find out which one won
How many sends it takes before that difference is real is arithmetic, and it is set out in our cold email A/B testing guide.
Nobody sets up spin tags at 3am on a domain they bought yesterday with no authentication on it. Spintax correlates with care, and care is what moves the number. When a tactic and a result travel together, ask what kind of person does the tactic.
How spun copy actually fails
A spun email rarely reads broken. It reads slightly off, which is worse, because nobody can tell you why they did not reply.
"Cheers | Kind Regards | Hey there" in one sign-off tag, going to the whole list.
- ✕Half your sends wear the wrong clothes
- ✕"Hey there" lands on a general counsel
- ✕The joins read fine, the tone does not
"Best | Thanks | Regards" in one sign-off tag, three options in one register.
- ✓Every option you would have written alone
- ✓Nothing that changes who you sound like
- ✓Nothing that carries your reason for writing
The discipline that catches this takes two minutes. Hit preview five times and read every option out loud against the sentence around it. If a tag holds an option you would not have written on its own, delete the option, not the tag.
Greetings and sign-offs carry no argument, so spin those all you like. The claim, the reason you are writing and the ask are the whole email. Spin those and you have shipped a version of your own argument that you never chose.
Running this on a two-person team
Four decisions. None of them takes an afternoon, and the fourth is the one that matters.
- 1 Set a fallback on every merge field you use.
- 2 Spin the greeting and the sign-off, and nothing else.
- 3 Run the sentence that matters as a tracked variant.
- 4 Then go and check your authentication, where the real problem is.
Four ways spintax costs you
None of these turn up in a report, which is exactly why they can run for months.
Every spun option is untracked. The more of the email you spin, the less any reply can tell you about which words earned it.
You wrote one version and previewed one version. The rest of the combinations went out to real people without anyone reading them once.
Rotating Hi, Hello and Hey tells the reader nothing about themselves. Spinning ten variants of a bland email gives you ten bland emails. Relevance beats rotation.
The afternoon that went into spin tags was the afternoon your authentication and your list needed, and it does not come back.
Never hide a div to make a conditional fire
Instantly's help center documents pasting a hidden div at the foot of your email so a conditional has the field it needs. Google's sender guidelines name hiding content with HTML and CSS as a reason messages get marked as spam. That is a bad trade.
Questions founders ask
What is spintax?
Does spintax actually improve deliverability?
What spintax syntax does my tool use?
Is spintax the same as A/B testing?
Where should I never use spintax?
What should I do instead of spinning more?
Co-founder of Real Good GTM. He has been the first business hire and Chief of Staff at seed-stage B2B startups, building outbound pipeline before any playbook existed. He wrote this one because every page on the topic asserts the deliverability claim, and not one of them opened a mailbox provider's documentation to check it.
Connect on LinkedInGo and fix the layer that decides it
The triage when mail stops landing, the discipline that keeps it landing, and the tools that send it.
Why cold emails go to spam
The nine causes ranked by how likely each one is yours, and the ten-minute check that clears half the list.
Read the triageThe deliverability guide
Authentication, warmup, complaint rates and recovery: the machinery that decides whether any of your words get read.
Read the guideThe best cold email tools
Seven syntaxes, seven products. What each one is actually good at, and what it costs to find out.
See the roundupWant outbound that lands without the tricks?
Book a fit check. We'll look at how your sending is set up, where your replies are actually coming from, and tell you straight if outbound is not the right motion for you yet.
Book a Fit CheckNo hard sell. No fake numbers. Real good work speaks for itself.