Open the email in Klaviyo's template editor, click Preview & test, type one or more addresses into the recipient field, and click Send Test. Klaviyo accepts up to 30 addresses at a time, separated by commas, and the mail arrives in a minute or two.
The procedure is not the interesting part. What matters is what that message is, because a Klaviyo test send is not a smaller version of your campaign. Klaviyo's own documentation states that preview emails do not render message variables such as manage preferences and unsubscribe links, do not render dynamic coupons, and carry neither Google Analytics tracking variables nor click tracking on their links. Those are not cosmetic omissions. They are close to a complete list of the things a mailbox provider inspects before it decides where to file your mail.
So a Klaviyo test send answers one question well and several questions not at all.
| What you want to know | Does a Klaviyo test send answer it |
|---|---|
| Does the layout hold up, and is the copy right | Yes. This is what the feature is for. |
| Do my merge tags resolve for a real profile | Yes, if you pick a profile in the recipient field. |
| Does my one-click unsubscribe header work | No. Unsubscribe and manage preferences are placeholders in a preview. |
| Do my tracked links resolve, and does the tracking domain look clean | No. Click tracking is not applied to a preview. |
| Will this land in the inbox or in spam | No. There is no scoring, no filter simulation and no seed testing in Klaviyo. |
| Is my DKIM signature aligned to my own domain | No. You need to read the headers of a message that went out the normal path. |
Send a test email in Klaviyo
Klaviyo puts the control in the same place for every kind of email, so the path does not change between a campaign and a flow.
- Open the email in the template editor. This works for a standalone template, a campaign, or a flow email triggered by a list, a segment or a date. Klaviyo notes that metric-triggered flow emails are the exception, because their dynamic event content has no event to draw from in a preview.
- Click Preview & test in the editor toolbar.
- Enter the recipients. Use the Search profiles field to pick a real profile if you want to see how merge tags resolve for an actual subscriber, or type plain addresses separated by commas for a straight layout check.
- Click Send Test.
- Click Done to return to the editor.
Klaviyo's documentation for this flow was last updated on 14 August 2025.
Klaviyo caps test sends at 30 recipients and meters them monthly
Two limits apply, and neither is announced in the interface at the moment you hit them. The recipient cap is 30 addresses in a single test send. The monthly allowance depends on your contact count: 100 preview emails on a free plan holding up to 250 contacts, 500 per month between 251 and 500 contacts, and one tenth of your plan's sending capacity above that.
The allowance is shared across the account rather than granted per user, which is worth knowing if a design review routinely copies six people on every iteration. A team that tests thoroughly on a small plan can exhaust a month of previews in a week of work on one campaign.
Preview & test removes the links a filter actually reads
The stripped elements are the point of this article, so they are worth stating individually rather than as a list you skim.
Unsubscribe and manage preferences render as placeholders. A working one-click unsubscribe is not a nicety. It is one of the bulk sender requirements Gmail and Yahoo enforce, and it depends on a correctly formed List-Unsubscribe and List-Unsubscribe-Post header pair that a preview never generates. Across messages tested with Unspam this year, only 14% send a one-click unsubscribe correctly, which makes it the most commonly broken requirement in the whole set, and a test send is structurally incapable of catching it.
Click tracking is not applied. Your campaign rewrites every link through a tracking domain. If that domain is unbranded, newly created, or shared with other senders, it is one of the first things a filter weighs. A preview sends the raw links, so the version you inspect is not the version your subscribers receive.
Dynamic coupons do not generate. A coupon block that renders empty in a test is expected behavior, not a bug, and chasing it wastes an afternoon roughly once per team.
Inbox testing shows rendering, not the spam folder
Klaviyo added an Inbox testing tab beside Preview & test, powered by Mailgun, and the name invites a misreading. It generates screenshots of your email as it appears across desktop, mobile, web and dark mode clients. It does not report whether a message reached an inbox or a spam folder, because it never sends anything through a mailbox provider's filters.
The feature is available on paid accounts only and metered at 100 inbox tests per user per month, where selecting five inboxes consumes five tests. As a rendering check it is genuinely useful. As an answer to "will this land in spam" it is the wrong instrument, and its documentation, updated on 1 October 2025, does not claim otherwise.
Klaviyo has no spam scoring anywhere in the product. Its published advice for finding content that trips filters is a manual bisect: send a blank email, add the HTML, add text one paragraph at a time, then add URLs one by one, with a test send between each step. For the scoring itself, Klaviyo's help center sends users to third-party tools and names Mail Tester and Google Postmaster Tools directly.
Your test lands in spam and the campaign does not
This is the single most common Klaviyo test result that sends people looking for a problem that does not exist, and Klaviyo has documented the cause. When the from address or the reply-to value matches the address you sent the test to, Gmail flags the message as possible phishing, because Gmail can see your sent folder and knows the message never left your account.
The fix is to stop testing against yourself. Send previews to an address on an unrelated domain. If the test still lands in spam from there, the signal is real and worth chasing.
A good Klaviyo deliverability score is not inbox placement
Klaviyo's Deliverability hub sits under Analytics and produces a 0 to 100 score, and it requires at least 1,000 emails sent in the past 30 days before it reports anything at all. The score is built from open rate, click rate, bounce rate, unsubscribe rate and spam complaint rate.
Read that list again with one question in mind: what does each input require in order to be recorded. Every one of them requires the message to have arrived somewhere a person could act on it. Mail filed directly into spam produces almost no opens, almost no clicks, no unsubscribes and, by definition, no complaints. It removes itself from the denominator. A program can hold a healthy score while a meaningful share of its audience never sees a send, which is why a good score and a spam problem coexist more often than the number suggests.
Placement is measured, not inferred. Of the campaigns run through Unspam so far this year, 65% of emails reach the primary inbox, and the Unspam deliverability benchmark tracks how that moves.
When to send the real thing instead
Once the layout is right, the remaining questions all need a message that travelled the normal path: through your sending domain, with your DKIM signature, your tracking links and your unsubscribe headers intact. That means a real campaign send to a seed address rather than a preview.
The Klaviyo deliverability guide walks that setup end to end, including the Smart Sending setting that silently skips repeat sends to the same profile inside 16 hours and will quietly swallow your second test of the afternoon. For the placement half of the question, an inbox placement test sends your campaign to seed mailboxes at Gmail, Outlook, Yahoo and five other providers and reports where each copy landed.
When you have the message you plan to send, run it through the free Unspam spam test first. It reads the authentication on the real send, scores the content, and shows you what a filter sees before a subscriber does.
