How to Name Your Secondary Domains
Last updated: November 17, 2026
Your messaging and audience matter far more than your domain name, but a few naming choices consistently cost you replies. Here's what to pick and what to avoid, based on reply rates across hundreds of Amplemarket customers.
For setting up the domain itself (alias domains, registrars, authentication), see About Setting up a New Domain.
Choose a name
Keep your brand in the name so your emails don't seem to come from a third party. These patterns perform equally well, so choose whichever is available and reads best for your brand:
Your brand on a different TLD:
acme.co,acme.ioA short prefix before your brand:
tryacme.com,getacme.com,useacme.com
Keep it short and simple: getacme.com rather than acmeoutboundsalestool.com.
Choose a TLD
The top-level domain (TLD) is the part after the dot, such as .com or .io.
Use .com, .co, .io, .net or .org. They all perform about the same. There's more recognition and trust around a .com, so if you're in doubt, we recommend .com.
Avoid .us and .info. Domains on these TLDs get noticeably fewer replies than the same team's other domains.
Names to avoid
Adding both a prefix and a suffix, such as
tryacmehq.comorgetacmeloans.com. This is the pattern that underperforms most clearly.joinandmeetprefixes, such asjoinacme.com. They get fewer replies than other prefixes.Misspellings of your brand, such as
acm.comoracmr.com. To a recipient, a domain that's almost your brand looks like phishing.Hyphens, when you have a choice.
tryacme.comdoes slightly better thantry-acme.com. The difference is small, so a hyphenated name is fine if it's the only one available.
What to expect
Secondary domains usually get around half the replies of your main domain, whatever you name them. That's normal, and it's why the name alone shouldn't decide where you send from.
The data behind these recommendations
How we measured it
We looked at 28.8 million sequence emails sent from Amplemarket workspaces in 2025 and 2026, and labelled each sending domain by how its name relates to the brand (tryacme.com, acme.io, acmehq.com and so on).
Sorting domains by reply rate doesn't work. It mostly measures the team behind the domain: the workspace explains 74% of the differences in reply rate, the domain name only 6%. So we compared each domain with the other secondary domains in the same workspace, using a Poisson regression with a fixed effect per workspace. A score of 1.00 means "replies like getacme.com".
Naming patterns
Pattern | Replies vs. getacme.com |
|---|---|
| about 2.2× |
| 1.96× (small sample) |
| 1.17× |
| 1.05× |
| 1.04× |
| 1.01× |
| 1.00× (baseline) |
| 0.86× |
| 0.86× |
| 0.82× |
| 0.60× (clearly lower) |
Most of these differences are within noise. Prefix + suffix is the only pattern that's clearly worse. join also stands out on a second measure: it gets about a quarter of the main domain's replies, against roughly half for try and get.
TLDs
TLD | Replies vs. .com |
|---|---|
.com | 1.00× (baseline) |
.co, .io, .net, .org | 0.89× to 1.02× |
.us | 0.75× (clearly lower) |
.info | 0.66× (clearly lower) |
The TLD scores exclude the 10 largest senders, so a few big teams can't drive the result.
Hyphens and domain age
Hyphenated domains got about 15% fewer replies than unhyphenated ones in the same workspace. That's a small effect, right at the edge of what the data can detect. Domain age made no clear difference once a domain was warmed up and sending.
Caveats
Reply rate reflects both inbox placement and how well the message works, so it isn't a pure deliverability measure. Teams also don't assign campaigns to domains at random. Treat these results as patterns to avoid, not guarantees.