AI email design
Each variant starts from a one-line brief, designed in your brand, ready to test against the others.
See AI email designSUBJECT LINE TESTING, DONE FAIRLY
We test every subject line. You crown the real winner.
We send a few designed variants of each campaign — different subject lines, openers, offers or layouts — rotated evenly across your whole list, not a sample. You see each one's results, human opens apart, and you pick the winner.
THE OLD WAY OF TESTING
Most A/B tests crown the wrong winner
A typical split test sends two versions to a small slice of your list, waits an hour, then calls a winner and blasts it to everyone else. That slice might be a hundred people out of five thousand.
The scoreboard moves within minutes, before a single human has opened a thing. Security scanners open test emails automatically and click links automatically, and they get counted right alongside real interest, so the "winner" is often just whichever subject line a scanner liked best.
You end up sending the rest of your campaign on a guess dressed up as data, to people who never got a say in it.
HOW FAIR TESTING WORKS
One campaign, several honest shots
Here's what happens between the send and your real winner.
01
We design your variants
Our team creates two, three or more versions of your campaign: different subject lines, openers, offers or whole layouts.
02
We split your whole list evenly
Every variant goes to an even share of your entire audience, not a small sample, so the test runs at full size from the start.
03
Results come in per variant
Opens, clicks and replies are tracked for each version on its own line, with human opens kept apart from automated ones.
04
You pick the winner
Compare the real numbers and choose which version worked. Nothing gets crowned automatically on noisy early results.
THE DETAILS
Three ways fair testing protects your results
Here's exactly what sits behind that one headline: more versions, a fair split, and results you can trust.
VARIANTS, YOUR WAY
Several designed versions, not just two subject lines
We launch your campaign with multiple variants: different subject lines, different openers, different offers, even different layouts entirely, each one designed by our team.
- Varied subject lines, openers, offers or full layouts
- Every variant designed by our team, in your brand
- As many versions as the test calls for
ROTATED EVENLY, NO SHORTCUTS
Rotated across your whole audience, not a slice
Every variant goes out to an even share of your entire list at the same time, not a test group of a few hundred. A list of five thousand testing three subject lines means close to seventeen hundred people see each one.
- Every variant reaches an even share of everyone
- No small sample standing in for your whole list
- Fair at full size, not fair in theory
RESULTS YOU CAN TRUST
See human opens, separate from the noise
Each variant's results show opens, clicks and replies on their own line, with human opens reported separately from the automated opens that come from security scanners and preview tools.
- Per-variant results, not one blended number
- Human opens separated from automated opens
- You decide when enough people have replied
WHAT IT MEANS FOR YOU
What fair testing means for your business
Five ways a fair test changes what goes out next.
You stop guessing and start knowing
Every send becomes a real experiment, run at full size, not a coin flip dressed up as strategy. You see exactly which subject line, offer or layout your audience responded to, with the numbers to back it.
Every subscriber counts, not a sample
With the whole list split evenly, your test reflects your actual customers, not a guess about a slice of them.
Real interest, not scanner noise
Human opens and automated opens sit in separate columns, so a scanner opening a test email is never mistaken for a customer who cared.
You call the winner, every time
Nothing gets crowned on an hour of early, noisy numbers. You choose when the results are in and which version goes forward.
Your mail looks less like one blast
Different subject lines and layouts going out together keep your campaign from landing as one identical message to every inbox provider at once.
COMPARED TO THE USUAL SPLIT TEST
A fairer fight than the usual split test
A typical split test
- Small sample. A slice of your list decides for everyone else.
- Early call. A winner gets crowned within the hour.
- One scoreboard. Opens and automated opens blend into one number.
- Two options. Usually just subject line A or B.
- One look. The same email lands in every inbox at once.
Web Presence Websites
- Full audience. Every variant reaches an even share of everyone.
- Your call. Nothing is crowned until you say so.
- Separate scoreboards. Human opens counted apart from automated ones.
- As many as you need. Subject lines, openers, offers or layouts.
- Several looks. Different versions go out together, not one identical blast.
QUESTIONS
Questions about fair testing
Is this the same as A/B testing?
It's a fairer version of it. Every variant goes to an even share of your whole audience, not a slice, and results count real people, not scanners opening mail automatically.
How many variants can one campaign have?
As many as your test needs, whether that's two subject lines or several combinations of subject line, opener, offer and layout. Our team designs every one.
How do you stop a scanner from deciding the winner?
Every open is classified as human or automated the moment it happens, and the two are reported in separate columns, so an automated open never gets mistaken for a customer.
When do I find out which version won?
Whenever you decide the results are in. Nothing is crowned automatically on early numbers, so you can wait until enough of your audience has had a chance to open and click.
Does splitting my list hurt my results?
No. Every variant still reaches its even share of your whole audience, so you're never testing on a small group and guessing about the rest.
Can I test more than subject lines?
Yes. We vary the opening line, the offer or the whole layout, not just the subject line, and compare all of it the same way.
Will sending several versions make me look spammy?
The opposite. Sending different versions to different parts of your list keeps your mail from looking like one identical blast to every inbox provider at once.
Do I need a big list to run a fair test?
Any list works. The split is even whatever size your list is, so the share each variant reaches simply scales with your audience.
WORKS WITH
Works alongside everything else we send
Fair testing is one part of getting campaigns right. Here's what pairs with it.
Honest reporting
See every variant's human opens, clicks and bounces live, the moment your campaign goes out.
See honest reportingEvery address checked
Every address is checked with its mail server first, so your test results reflect real inboxes, not dead ones.
See address verificationSegments and follow-ups
Once you've got a winner, we save the people who clicked it and follow up with a resend to the rest.
See segments and follow-ups
Pick winners your audience actually chose
Every campaign, a few designed versions, one fair fight — and you decide who takes it.