Test the message after lemlist has substituted personalization, including the signature, custom text and destination URLs. A clean template with variable tokens is not the content recipients receive. Also verify the sending account: lemlist's documented editor test uses the logged-in user's connected email provider, which may differ from the campaign sender you intend to examine.
Choose representative personalization cases
Prepare non-sensitive fixtures for a complete record, an absent optional field and an unusually long value. Add a case for any condition that changes the message substantially, such as inserting a personalized image or a different landing-page URL. Do not use a real customer's private data simply to make the preview look realistic.
Keep a list of the fixture values and the expected rendered outcome. For the missing-field case, define a readable fallback rather than deciding after the send whether an empty greeting is acceptable. This turns the exercise into a check of actual behavior instead of a subjective review of one attractive example.
Preview the saved campaign step
The official preview guide directs you to the Sequence tab, an email step and Preview. Check that saved variables appear as real values, including those in the subject. Review every relevant step because a correct opening message does not establish that a follow-up has the same data or formatting.
Inspect the resulting links as carefully as the visible sentences. A personalized URL can look plausible while pointing to another record's page or including an empty query value. A display label is not proof of the destination. Record the final hostname and intended path without copying private tracking identifiers into a public report.
Confirm which account sends the test
lemlist states that test emails are sent from the logged-in user's address and require that user's connected provider. If the campaign is assigned elsewhere, an editor test can verify rendering while examining a different mailbox route. Keep this as an explicit boundary of the test.
For placement through the intended route, create an LDM test for the actual sender, obtain its current recipients and use an isolated permitted send that preserves the rendered content and relevant settings. Do not claim the campaign route was verified unless it was actually exercised. Forwarding a personal copy is insufficient because it adds a different sender and context.
Compare rendering and placement separately
Save the submitted or received copy so that the values tested can be reconstructed. Record the sender account, step, fixture, timestamp and available message identifier beside LDM's observed folder. A content check can pass while placement needs investigation; a copy in Inbox can still contain a broken greeting or wrong calendar link.
When an optional field changes the result, first compare the final messages. Did the fallback remove the entire sentence, change the amount of HTML, or introduce another domain? Those differences help frame the next experiment. They do not prove that the recipient provider penalizes personalization itself.
Stop at defensible conclusions
A small panel cannot measure how a specific prospect will react to a personalized opening. It does not validate facts in generated icebreakers or authorize using sensitive personal information. Review the accuracy and relevance of personalized claims independently before any real outreach.
If one rendered variant is observed in Spam, keep the remaining conditions stable and investigate that variant's actual differences. Preserve required unsubscribe features. Removing required elements to obtain a better test result would create an unrepresentative message.
Frequently asked questions
Is a preview enough for image personalization?
It checks the displayed result in that view. A real received copy is still needed to examine image loading and folder placement for the tested recipient and time.
Can I report a logged-in-user test as a campaign-sender result?
Only if that user and route are the intended campaign sender. Otherwise label it as rendering or editor-test evidence and perform the missing route check separately.
Should every record produce identical copy?
No. The goal is correct, intentional variation. Test the branches that change meaning, links or structure, rather than forcing every fixture into the same sentence.
Official references
Reviewed 2026-11-05; workflow guidance, not a completed test report.