How to Improve Cold Email Open Rates
Cold email open rates follow inbox placement first, then sender name, subject line and preview. Why opens mislead, what to fix first, what to measure instead.

On this page
Cold email open rates depend on inbox placement first. An email that lands in spam is rarely seen, however good its subject line. Once the email is in the inbox, three lines decide whether it gets opened: the sender name, the subject line and the preview text.
The open rate itself is a weak metric for cold email. It is counted by a tracking image that privacy features load on the reader's behalf and that blocked images never load, and Google says low open rates are not a reliable sign of spam placement. Fix placement, then the three lines the reader sees, and judge the campaign by replies.
We make Outreach2day, cold email infrastructure with domains, mailboxes, warmup and campaigns. Where a step below uses one of our features, we say so.
Order | What to work on | Why |
|---|---|---|
1 | Inbox placement: authentication, warmup, volume, list, plain copy | An email in spam is rarely opened |
2 | Sender name | It is shown next to the subject in the inbox list |
3 | Subject line | It tells the reader what the email is about |
4 | Preview text | In a plain-text email it is the first line of the body |
5 | Measurement | Reply counts show what people did; open counts include machines |
What an Open Rate Measures and Why It Misleads
Sending tools count an open when a tracking pixel loads: a tiny, invisible image with a unique address, added to the HTML of each email. When the image is requested from the tool's server, the tool marks that recipient as opened. Open rate is unique opens divided by emails delivered.
Anything that loads the image without a person reading the email adds an open. Anything that stops the image from loading removes one.
Apple Mail Privacy Protection
Apple describes Mail Privacy Protection as a feature that "helps protect your privacy by preventing email senders from learning information about your Mail activity." Its Protect Mail Activity setting "downloads remote content in the background by default — regardless of whether you engage with the email" (Apple).
For a sender, every recipient who reads mail in Apple Mail with this setting on can show as opened whether or not they read the email. The share of such recipients differs from list to list, so the inflation differs too, and two campaigns to different lists can't be compared on opens.
Image Proxies
Gmail loads images through Google's servers: "When your users open email messages, Gmail uses Google's secure proxy servers to serve images that might be included in these messages" (Google Workspace Help). Gmail's help for users says senders "may know whether you've opened an email that has an image," and that they "can't use image loading to get information about your computer or location" (Gmail Help).
So an open from a Gmail reader tells you the image was requested through Google's servers, and nothing about the reader's device or location.
Blocked Images and Plain Text
When a reader or their company's mail settings block images, the pixel never loads and a read email counts as unopened. A plain-text email carries no pixel at all, so it has no open rate. We recommend plain text for cold email, for the reasons in the tracking section below.
What happened | Counted as an open |
|---|---|
The reader opened the email in a client that loads images | Yes |
Apple Mail downloaded the remote content in the background | Yes, whether or not the reader read it |
The reader opened the email with images blocked | No |
The email was sent as plain text with no tracking image | No data |
What Google Says About Open Rates
Google's sender guidelines address opens directly (Google sender guidelines):
"Google doesn't track open rates"
"Google can't verify the accuracy of open rates reported by third parties"
"Low open rates aren't necessarily an accurate indicator of deliverability or spam classification issues"
If you track opens at all, use them to compare two variants sent to the same list, from the same mailboxes, on the same days. Don't read a low open rate as proof of spam placement or a high one as proof of interest.
Fix Placement Before Copy
An email in the spam folder is rarely opened, so check placement before you rewrite the subject line. Copy can cause spam placement too; the paired test at the end of this section shows whether it does. Work through these in order.
Authenticate Every Sending Domain
Authentication proves to mailbox providers that mail from your domain comes from you. Google requires every sender to "Set up SPF or DKIM email authentication for your sending domains," and senders of more than 5,000 messages a day to set up SPF, DKIM and DMARC (Google sender guidelines). Set up all three on every cold domain at any volume.
Protocol | What It Does | Why You Need It |
|---|---|---|
SPF (Sender Policy Framework) | Lists the servers allowed to send email for your domain. | Receivers can reject mail from servers not on the list. |
DKIM (DomainKeys Identified Mail) | Adds a cryptographic signature to every email. | Proves the message was not altered in transit and ties it to your domain. |
DMARC (Domain-based Message Authentication, Reporting & Conformance) | Tells receivers what to do with mail that fails SPF and DKIM alignment (none, quarantine or reject) and sends you reports. | Protects your domain from spoofing and shows who sends in your name. |
Check a domain with the SPF, DKIM and DMARC checker. It reads the records in your browser, and each finding says what breaks and how to fix it.
Warm Up and Cap Volume
New mailboxes have no sending history. We recommend 14 days of warmup before the first cold email, full volume from day 14, and warmup running for as long as the mailbox sends cold email.
Keep each mailbox at up to 25 cold emails a day and each domain at up to 5 mailboxes. To send more, add mailboxes in batches and check the warmup inbox rate before each new batch.
Clean the List Before Every Send
Invalid addresses bounce, and a high bounce rate hurts your reputation. Verify addresses before each send and remove the ones that fail; our guide to email verification programs compares the tools.
Remove anyone who asks you to stop, and keep complaints low. Google asks senders to keep the spam rate "below 0.10%" and to "avoid ever reaching a spam rate of 0.30% or higher." A complaint costs more than a missed reply: it tells the provider the email was unwanted.
Keep the First Email Plain
Links, images and tracking pixels add content that filters judge. Google's Postmaster Tools help says: "Some types of message content, such as Internet links, increase the possibility of messages being treated as spam" (Postmaster Tools help).
Send the first email as plain text with no links, including the website in the signature. Mail apps turn a bare domain into a link. Paste each step into the cold email copy checker: it flags links, bare domains, images, spam-trigger phrases and fake Re: or Fwd: subjects before you send.
Test Placement With a Control Email
A placement test sends your email to seed inboxes at Google and Outlook and reports inbox or spam for each one. A single test is noisy, so pair it: send a neutral, non-sales email and your campaign email from the same mailboxes in the same minute.
Result | What it points to | Next step |
|---|---|---|
Control and campaign email both in spam | The domain, the mailboxes or the sending IPs | Check authentication, warmup inbox rate and volume; repeat on another day |
Control in the inbox, campaign email in spam | Copy or test variation | Remove links and images, rewrite the subject and opening line, test again |
Both in the inbox | These seed inboxes received both messages | Work on the sender name, subject line, list and offer |
Read Google and Outlook results separately: the same email can land in the inbox at one and in spam at the other. Repeat a result on a second day before you change domains or mailboxes. The full method is in our inbox placement test guide, and how reliable placement tests are explains the noise.
Outreach2day includes Google and Outlook placement tests for mailboxes that are 21+ days old, with up to 100 mailboxes per test and an inbox or spam result for each seed.
Sender Name, Subject Line and Preview Text
Once the email is in the inbox, the reader sees three things before opening it: the sender name, the subject line and the first words of the body. A cold email comes from someone the reader does not know, so all three should read like a normal message from one person to another.
Sender Name
Use the name of the person the email comes from, as in "Anna" or "Anna from Outreach2day". Keep it the same in every email from that mailbox and in the signature. A company name alone or a team name such as "Sales Team" reads like a newsletter.
Subject Line
What to aim for:
Short and specific: the reader's company, role or a problem they are likely to have
Written the way people write to colleagues, in sentence case
A question when the email asks one
A first email without "Re:" or "Fwd:": a fake reply prefix misleads the reader
What to leave out: ALL CAPS, several emojis, exclamation marks, and promises the email does not keep. These annoy readers and can trigger spam filters.
At volume, the same subject line can go out thousands of times a day. Write it as spintax, so few exact subjects repeat, and group the variations into a small number of families you can measure. The subject line spintax generator skill writes families in this format, and our cold email subject lines post has examples and patterns.
Send follow-ups with an empty subject so they stay in the same thread as the first email.
Preview Text
Email clients show the start of the body next to or under the subject. Marketing emails set it with hidden preheader text; in a plain-text cold email the preview is the first line of the body. Write that line for the inbox list:
Start with the reason for the email: who it is for or what you noticed about their company
Keep the greeting short ("Hi Sarah,") so the first sentence shows in the preview
Avoid repeating the subject line word for word
Should You Track Opens in Cold Email?
We recommend turning open tracking off for cold email. Compare what each setup puts in the email and what it tells you:
With open tracking | Without open tracking | |
|---|---|---|
Format of each email | HTML with an image loaded from the tracking domain | Plain text |
Domains in the email | Your sending domain and the tracking domain | Your sending domain |
What you learn | An open count inflated by Apple Mail and reduced by blocked images | Replies, auto-replies and bounces, the same as with tracking |
The tracking image goes into every email, including the first one from a mailbox the recipient has never heard from, and the tracking domain's reputation becomes part of yours. In return you get an open rate with the gaps described above.
Send plain text with no pixel and no links in the first email, and share links after the prospect replies. What Google says about links and URL-domain rate limits is in do open tracking and links hurt deliverability.
Free guide19 min read
Send 100,000 cold emails a month
Domains and mailboxes you need, DNS, warmup, lists, copy and follow-ups, with a launch checklist, sent to your inbox.
What to Measure Instead
Replace opens with metrics that count what people and mail servers did.
Question | Metric | Where it comes from |
|---|---|---|
Is the copy working? | Human reply rate per mailbox and per subject family | Replies grouped by mailbox and subject family |
Are emails reaching working mailboxes? | Auto-replies (out-of-office) per mailbox | Incoming mail sorted by type |
Is a mailbox or domain losing trust? | Warmup inbox, spam and bounce rate per mailbox | Warmup stats |
Where does the email land? | Inbox or spam per seed, Google and Outlook separately | Paired placement tests |
Is the list clean? | Cold bounce rate, counted against cold sends | Bounces with cold and warmup split |
Is the offer working? | Positive replies and meetings per 1,000 emails sent | Your CRM |
Read them together:
Replies fall while auto-replies hold: the emails are arriving, so look at the list, the offer and the copy
Replies and auto-replies fall together: look at placement and run a paired test
One mailbox's warmup inbox rate drops while the others hold: act on that mailbox and its domain first
Cold bounces rise: check the list source and the verification step
To compare subject lines by replies, send each subject family from the same mailboxes to the same list segment over the same days. With the subject as the only difference, the gap in human replies is the subject's effect.
In Outreach2day, each mailbox shows its warmup inbox, spam and bounce rate for the last 14 days with a green, amber or red status. Incoming mail is split into human replies, out-of-office and bounces, and cold and warmup bounces are counted separately. Formulas for each rate are in email campaign performance metrics.
FAQ: Cold Email Open Rates
What Is a Good Cold Email Open Rate?
There is no benchmark you can rely on. The number depends on how many of your recipients use Apple Mail with Protect Mail Activity, how many block images, and whether you track opens at all. Email marketing benchmarks measure newsletters and promotions sent to subscribers, a different audience from cold email.
Compare your open rate only with your own earlier sends that used the same tracking settings, and judge campaigns by reply rate. Google says low open rates "aren't necessarily an accurate indicator of deliverability or spam classification issues," so a low number calls for a placement test before any conclusion.
Why Are My Cold Email Open Rates Low?
Check these in order:
Placement: run a paired placement test to Google and Outlook seed inboxes
Authentication: SPF, DKIM and DMARC on every sending domain
Mailbox readiness and volume: at least 14 days of warmup and up to 25 cold emails per mailbox a day
The list: bounces, outdated contacts and people outside your target market
The three lines the reader sees: sender name, subject line and preview text
Measurement: recipients who block images count as unopened, and plain-text emails have no opens to count
Does Open Tracking Hurt Deliverability?
It adds risk. Open tracking puts an image hosted on the tracking domain into every email, so that domain is part of every message filters scan and its reputation becomes part of yours. What you get back is an open rate that, in Google's words, isn't "necessarily an accurate indicator" of spam classification. We recommend sending cold email with open tracking off.
Is A/B Testing Subject Lines Worth It?
Yes, when the result is measured by replies. Send the variants from the same mailboxes to the same list segment over the same days, split the sends evenly, and compare human reply rates. Opens are a poor judge here, because Apple Mail's background downloads and blocked images change the count whatever the subject line says.
Small differences in replies are often noise. Run the winner against a new variant on the next send, and keep the subject that wins twice. Outreach2day sequences support up to four A/B variants and spintax.
Sources
Checked October 1, 2026:
Apple: Mail Privacy Protection & Privacy: background download of remote content
Google: Email sender guidelines: open rates, authentication, spam rate thresholds
Google Workspace Help: Set up an image URL proxy allowlist: Gmail image proxy
Gmail Help: Turn images on or off in Gmail: what image loading tells senders
Gmail Help: Postmaster Tools dashboards: links and spam classification
Outreach2day sets up the domains, DNS records, mailboxes and warmup for cold email, with per-mailbox warmup stats and Google and Outlook placement tests, at $2.50 per mailbox a month.
Keep reading
Send 100,000 cold emails a month
How to send 100,000 cold emails a month while keeping quality, engagement and conversion rates high: domains, mailboxes, warmup, leads, copy and scaling
Step-by-step playbook · 19 min readDeliverability8 min read
Microsoft 365 Cold Email After MC1454397: New-Tenant Limits
Deliverability6 min read
Spam Test Score: What It Means and How to Fix It
Deliverability9 min read
Cold Email Deliverability 2026: What Changed, What to Do
Answer3 min read
My Reply Rate Dropped. Will New Domains Fix It?
Agent skill
Cold email campaign metrics reader