What emails do insurance agents actually open? The plain ones.
Across about 6,000 emails we sent to insurance agents for two clients in 2026, the best performers were a thank-you for attending, a short reminder of who the firm is, and a worked example of one policy. Each opened above 55%. Subject lines that led with a number were the weakest we sent, and an email sent in a different person's name opened about 12 points lower.
The list mattered as much as the words. On an old list of agents, almost one address in ten was dead, and most of the dead ones were carrier email addresses. The agents had not quit. They had changed carriers.
What opened, and what got clicked.
Open rates are inflated by mail apps that open messages on their own, so read the click column first.
| Opened | Clicked | What it was | |
|---|---|---|---|
| Thank-you for attending | 56.4% | 16.8% | Plain thanks to webinar attendees, with the replay. The best email in either campaign. |
| First two emails in the agent sequence | Above 55% | Not split out | A reminder of who the firm is, then one term policy worked through as an example. |
| The podcast email | Not split out | 12.8% | Built around a podcast conversation. The best click rate in its campaign. |
| Attendees, whole sequence | 34.9% | 12.0% | Average over 903 sends to people who showed up. |
| No-shows, whole sequence | 19.2% | 3.75% | Average over 801 sends to people who registered and missed it. |
| Offer notice, both lanes | 28.2% and 31.0% | 7.4% and 6.8% | Two emails to about 610 people each, attended and no-show together. |
Four things the numbers taught us.
- Lead with who you are, not what you want. The strongest agent email of the summer was a plain reminder of who the firm is. Agents on an old list do not remember you, and the first email has to fix that before any offer can land.
- Show one real case. The other email above 55% walked through a single term policy, start to finish. Agents open a worked example because it is something they can use with a client on Monday.
- Keep the numbers out of the subject line. The two weakest emails in the software campaign both led their subject with a figure. The next sequence led with plain curiosity instead.
- Keep one sender. The one email sent in a different person's name opened about 12 points below the usual sender's average. Agents open mail from a person they recognize.
One pattern held across both campaigns: clicks barely moved when opens did. That says the bodies were doing their job, and the subject line was the thing worth testing.
Why a tenth of one agent list was dead.
Worked example: cleaning an old list in the first week
In the first week of the life settlement campaign, we ran 1,946 agent addresses through verification. 1,702 came back deliverable, 87.5%. With the contacts already flagged as bad, 219 addresses on a list of 2,279 were silenced, 9.6%, so they could not drag down the sender's reputation.
The surprise was who the dead addresses belonged to. They were not agencies that had closed. They were carrier email addresses, and the carriers are still in business. The agent had moved on, and was still selling somewhere else. These are the domains with more than one dead address.
| Knights of Columbus (kofc.org) | 5 |
| MassMutual (financialguide.com) | 4 |
| Farmers (farmersagent.com) | 4 |
| Western and Southern (wslife.com) | 3 |
| American Family, Lincoln Financial, Thrivent | 2 each |
So a dead carrier address is a phone call, not a delete. We turned off email for those contacts only and left text and calls on, and the bad-email list became a queue for a person to call. A list from a different client, checked in September, ran 3.3% bad.
Two mistakes that cost more than a weak subject line.
Judging the campaign by its open rate.
Apple Mail and others open messages automatically, so open rates run high and tell you little about people. Clicks are the honest number. For scale, the all-industry average click rate in Klaviyo's 2026 email benchmarks is 1.69%; the attendee sequence above averaged 12.0%.
"Fixing" a typo in an agent's address.
Correct an address only when the original domain does not exist at all. We found one that looked like a typo for a big provider, and the corrected address verified as deliverable. But the original domain was a real internet provider, so the "fix" would have sent an agent's details to a stranger. Two tests, every time: the old domain is bogus, and the new address verifies on its own.
Questions people ask
What is a good open rate for emails to insurance agents?
In our 2026 campaigns, whole sequences averaged 19% to 35% opens, and the best single emails opened in the mid 50s. Opens are inflated by mail apps, so compare click rates too: our sequences ran from 3.75% for no-shows to 12% for attendees.
Why did so many of my agent emails bounce?
On an older list, usually because the agent changed carriers and the carrier address died with the appointment. In one list of 2,279 agents, 9.6% of addresses were silenced as bad, and the dead domains were mostly carriers. Verify before the first send, and call the ones that fail.
Should emails to agents come from the company or a person?
A person, and the same person every time. In our data the one email sent under a different name opened about 12 points lower than the usual sender's average.
Is it worth emailing people who missed the webinar?
Yes, with lower expectations. No-shows opened at 19.2% and clicked at 3.75%, about a third of the attendees' click rate. When attendees and no-shows later got a time-limited offer together, the two emails opened at 28% and 31%.
Founder of ElevAIte CRM and a licensed insurance producer in Pennsylvania and Maryland. He built and ran both campaigns this article draws on.
Published · Updated
Keep reading
- Case studyRe-engaging dormant agent leads: 32.9% opensThe valuation tool, the cadences and the list cleanup.
- Case studyWebinar follow-up for WealthLogixThree lanes, 22 templates, and what we changed after the step stats.
- IndustryCRM for insurance agentsThe leaks in a book of business, and what runs to close them.
Want your emails read by the step stats, not by guesswork?
We build the sequences, verify the list before bad addresses hurt the sender, and change the next email because of what the last one did.
Figures are from two clients' CRM email reports and workflow step stats, July to September 2026, and are specific to those campaigns. Open rates are affected by automatic opens in mail apps. Nothing here is a prediction for any other list.