Testing and Previewing a Klaviyo Email Before BFCM: The Quality Check

A Black Friday email goes out to a large share of your list. An empty variable, a broken link or a message cut off by Gmail is then seen thousands of times. Klaviyo offers several checking tools, and they do not all verify the same thing. This article puts them in the order in which we use them.

TL;DR. Preview checks layout and personalization, a test send shows the rendering in a real inbox, and a campaign sent to a small internal list is the check closest to a real send. An email over 95 KB is exposed to Gmail clipping. For flows, manual mode lets you review each message before it goes out. In our view, this check should be scheduled several days before the operation, because test sends have monthly caps.

Want an outside look before you build? We offer a Q4 readiness audit: 30 minutes, free, no strings attached.

Step 1: preview in the editor

From the template editor, the Preview button at the top right opens the preview. The desktop and mobile toggles let you check the layout on each format: content, links, images, spacing and conditional content. The exact options depend on what you are editing: a template, a campaign email or a flow email (Klaviyo, preview and send test emails).

Preview with real data

A layout preview shows only part of the result. If the email uses personalization, you need the preview with a real profile, to see how profile variables display. For a flow email that supports event data, the page advises using a recent event, to check the variables tied to the event. If the preview is incorrect, check that the chosen profile or event actually contains the data referenced in the message.

For flows, the Preview & test button presents the 10 most recent qualifying profiles or events (Klaviyo, test and preview flow messages). For a flow triggered by a metric, such as an abandoned cart, these are the 10 latest trigger events. For a flow triggered by a list or a segment, you can search for any profile in the account, even one that is not in the list.

In our view, before BFCM, preview with at least three different profiles: a customer with a full history, a new subscriber with no purchase, and a profile where a personalization field is empty. The third one reveals the "Hello ," greetings with no first name.

Step 2: the test send

The test send is for seeing the email in a real inbox, which the page describes as the best way to check spacing, fonts and the rendering specific to each email client. It specifies that some dynamic content does not behave the same way in a preview as in a real send, and that to validate real links or unsubscribe behavior, you need a test under real conditions.

Preview limits

A preview can be sent to 30 recipients at most at a time, and each account has a monthly cap depending on its plan: 100 preview emails for the free plan from 0 to 250 contacts, 500 for 251 to 500 contacts, and one tenth of the plan's sending capacity above 500 contacts (Klaviyo, preview and send test emails). A team that tests in a loop during the week before BFCM can therefore hit the cap. The page on preview emails classified as spam indicates that you can then preview in the editor, or create a preview list and send a campaign to it.

Why a test email lands in spam

The dedicated page gives several explanations (Klaviyo, why some preview emails land in spam):

• Placeholder content: filler text can trigger the email client's spam filter. The page recommends test content that looks as much as possible like the real campaign.
• Gmail learns per user: if you often send previews to yourself without opening them or clicking, they can end up in spam. Opening and clicking those emails, or dragging them to the primary tab, helps reverse the trend.
• Sending address identical to the recipient: if Google manages your company's address and the sending or reply-to address is the recipient's own, Gmail can classify the test as spam, mistaking it for a phishing attempt.
• Corporate firewall: sending several tests to several colleagues can trigger strict filters.

If tests sent from your own address cause problems, the page suggests using the preview from the templates tab, which is sent from Klaviyo and not from your address. If tests sent from the template editor cause problems in turn, it suggests the preview from the flow or campaign editor.

Step 3: the internal preview list

For the most realistic check of a campaign email, the page advises sending the campaign to a small internal list: you create a list of reviewers, add the people concerned, and send the campaign to that list instead of the real audience. This is the best option for checking behavior specific to sending.

One limit to keep in mind: a message sent as a campaign does not contain the dynamic event content available in metric-triggered flows. To send a real flow email with all the event details, you have to trigger the flow yourself by performing the trigger action.

Step 4: email size and Gmail clipping

Gmail cuts off a message that exceeds a certain size. On desktop the limit is generally 102 KB, on mobile it varies from about 20 KB on iOS to about 75 KB on other devices, the 20 KB limit of the Gmail iOS app being applied inconsistently. This limit applies to the code (text, styles, links, HTML) and not to the weight of images, which load from an external source (Klaviyo, why is my email clipped).

Two consequences for a promotional email:

• A clipped email shows a View entire message link, which opens in the browser a version created by Gmail without the style tags: alignment, columns and tables can disappear or change appearance.
• The tracking pixel, which is used to calculate the open rate, is cut off too, so opens are miscounted. The page advises moving it to the top of the email if size is a concern.

To check, open Preview & test and look at the Email Size section: Klaviyo estimates the message size and the clipping risk. Under 85 KB, the email is not at risk. From 85 to 95 KB, it may be. Above 95 KB, clipping is likely. The same page nonetheless cites mobile limits of about 20 KB on iOS and about 75 KB on other devices, without saying which of the two these bands refer to. In our view, "under 85 KB" is therefore best read as a margin against the 102 KB desktop limit, not as an assurance on mobile. The page recommends adding the template to a campaign or a flow before measuring, so that UTM parameters are included in the estimate, and previewing with several profiles if the email contains a lot of dynamic content. This ties in with our article on UTMs in Klaviyo.

To reduce size without removing content, the page cites several levers: limit blocks displayed only on desktop or only on mobile (their code is included in every send), avoid the header block with a links bar, which is very heavy in code, enable embedded styles in Settings, Email, CSS Optimization, remove section background images, remove superfluous styles inherited from a copy and paste out of a word processor (by pasting as plain text), and group blocks and sections with similar styles. A Black Friday email with a dynamic product grid and many conditional blocks is typically affected, in our view.

Step 5: testing in several inboxes

Klaviyo's inbox tests, run in partnership with Mailgun, are reserved for paid accounts. The email subscription includes 100 tests per user per month, and each selected inbox counts as one test: choosing five inboxes uses five tests. The page recommends choosing a small sample based on the most common inboxes among your subscribers (Klaviyo, use inbox testing).

You start the test from Preview & test, then Inbox testing, choosing a profile or an event, then the inboxes from the Dark mode, Mobile, Desktop and Web groups. While the previews are being generated, do not edit the template, or you will get previews that no longer match. A share link can be enabled for reviewers: it expires after 6 days, or immediately if you disable it, and reviewers see all the selected versions.

The page warns that each inbox interprets HTML slightly differently: a dark mode can invert colors, and Outlook generally offers the most limited support. Our guide to email dark mode details these cases.

Step 6: testing flow logic

For a flow with complex filters, the page offers two methods (Klaviyo, test and preview flow messages). The first is the trigger preview tool, which shows how contacts enter the flow and how the filters evaluate them. The second puts a message in manual mode: it is scheduled as if it were live, but instead of going out, it is marked for review. Contacts wait in Waiting, then move to Needs Review at send time if they pass the filters. You can then send one message, send everything or cancel.

If nobody enters the flow in manual mode, the filters are probably too restrictive. The page suggests creating a segment that reproduces the flow's configuration to see whether any contacts fill it. For a simple flow, or one built from an idea built into Klaviyo, such as a welcome series, it indicates that you can set all the messages live without testing the logic.

In our view, manual mode is worth using for flows modified for BFCM, such as a last-week flow or a waitlist flow, before making them live.

Checklist before BFCM

These points are Deliver advice.

1. Each email is previewed on desktop and mobile with at least three profiles, including one with incomplete data.
2. A test send is done for each template, with content that looks like the real campaign.
3. Important campaigns are sent to an internal list of reviewers before scheduling.
4. The Email Size section is checked for each email, with a margin under 85 KB.
5. A few inboxes representative of your list are tested, without using up the 100 monthly tests.
6. Modified flows go through manual mode before being made live.
7. No change is made to a template while an inbox test is generating.

FAQ

What is the difference between preview and test send?

The preview displays inside Klaviyo and is used to check layout and personalization. The test send arrives in a real inbox and shows the actual rendering of the email client.

How many recipients can you add to a preview?

Up to 30 at a time, with a monthly cap depending on the plan.

At what size does Gmail cut off an email?

The general limit on desktop is 102 KB of code. Klaviyo estimates the risk as likely above 95 KB.

Do heavy images cause clipping?

No. The weight of image files is not part of the calculation, only the code that displays them counts.

How many inbox tests does the subscription include?

100 per user per month, with each selected inbox counting as one test. The feature is reserved for paid accounts.

Want an outside look at your quality check? We run Q4 readiness audits, or you can talk directly with the Deliver team.