What happened when we sent cold email under seven sender names.

The short version.

• The two male names got 1.7 to 2 times the reply rate of the five female names. Bob and Gerald each had an overall reply rate of about 5%. Elizabeth, Olivia, Mia, Rachel and Anna had 2.5 to 3%.
• Gerald's inboxes brought in about 3 of the roughly 5 positive replies a day. Bob's brought in 1, and the five female names together brought in 1, though every name had the same planned volume.
• We had expected the female names to win. A lot of cold email advice says female sender names get more replies, especially from male decision-makers. In this test, with this offer and these buyers, they got fewer.
• Every inbox had a profile picture: young women on the female names, older men on Bob and Gerald. Gender and apparent age changed together with the name, so we can't separate the name from the picture, or from any other difference between the inboxes.
• In 2018, one sender in a 30-person UK campaign landed in spam far more often than the rest. All 30 used the same templates and setup. The one whose name is more commonly associated with an Asian background had the spam problem, for much of a year. One account can't separate the name from that account's own reputation.

How we ran the 2026 test.

We set out to test a common piece of cold email advice: that female sender names get more replies, particularly from male decision-makers, and that a profile picture strengthens the effect. The campaign offered B2B lead generation to small businesses in the US with 1 to 50 employees.

Each of the seven names had 20 sending inboxes, and each inbox sent 15 emails a day. That gave every name the same planned volume of 300 emails a day, and the test a capacity of 2,100 a day, or 126,000 over the 60 days if every inbox sent its full quota.

Bob and Gerald got 1.7 to 2 times the reply rate.

Bob and Gerald each had an overall reply rate of about 5%, counting every reply. The five female names had 2.5 to 3%. That puts the male names' reply rate at about 1.7 to 2 times the female names', or 67 to 100% higher.

Gerald's inboxes brought in 3 of every 5 positive replies.

We saw about five positive replies a day: about three from Gerald's inboxes, one from Bob's and one from the five female names together. Every name had the same number of inboxes and the same daily target, so Gerald's inboxes were set to send about one email in seven, and they brought in about 60% of the positive replies. The five female names were set to send about 71% of the email and brought in about 20%.

Why Gerald may have done so well.

We don't have a definitive answer. These are the explanations we think are worth testing.

• He may have come across as experienced. Gerald is a traditional name that may suggest an older professional, and his profile picture showed an older man, while the female names' pictures showed young women. The test can't tell age apart from gender. For a small-business owner weighing a lead generation offer, an impression of experience and credibility could affect whether they reply.
• The buyers were owners of small companies. The owner of a business with 1 to 50 employees may respond to a sender differently from a marketing manager at a large company. Some sender identities may feel more familiar or relevant to one audience than another.
• Credibility may have mattered more than approachability. Female names are often recommended on the assumption that they seem more approachable. Depending on the offer, recipients may respond more to someone they see as knowing the work.
• The name came with a picture and an email address. Together they may shape the first impression. We didn't change them one at a time, so we can't tell which one moved the result.

In 2018, one sender's emails landed in spam far more often.

Sender names first caught our attention in 2018, on a large UK campaign Elena managed for a client. Thirty agents sent about 30,000 first emails a month, or about 120,000 a month with follow-ups, for almost a year. They used the same email templates, the same technical setup and the same general approach. Twenty-nine of them used traditional British male names. One used a male name more commonly associated with an Asian background.

The 29 agents with British names got consistently strong results. The one agent's emails landed in spam far more often, and the gap persisted through much of the year. We tracked inbox and spam placement with third-party placement tests and with manual sends to seed addresses, much like seed-list testing today.

Both campaigns side by side.

What it means for your sender names.

One 60-day test of one offer is too little to rename every inbox Gerald, so we test sender identity the way we test a subject line.

• Split inboxes evenly across the identities you test, and judge each one on positive replies against the emails it sent, not on total replies.
• Test the picture apart from the name. Run one name with two different pictures, such as a younger and an older face, or one picture under two names, so you know which one moved the result.
• Retest when the market, company size or offer changes. A sender that works with US small-business owners buying lead generation may not work with enterprise buyers or in another country.
• Treat advice about gender, familiarity or credibility as a hypothesis. That includes the advice that female names get more replies, which this test didn't bear out.
• Check spam placement for each sender, not only replies. Seed tests run sender by sender show placement separately from replies, as they did for us in 2018.

How to cite this study.