Ghost Newsletters: Publish, Email, and Choose the Right Audience

A developer-writer has finished a tutorial. The website needs the full explanation, but the newsletter might work better as a short roundup linking to it. Sending the tutorial itself is another option. Decide what belongs in the inbox before opening the publish menu.

Ghost's publishing mode is only the first decision. Website access needs its own check, as does the email audience. This guide uses Ghost's documentation to connect those checks. The tutorial and roundup are hypothetical examples; this is not an account walkthrough or a report of a real send. No subscriber data was accessed.

Choose Publish only, Email only, or Publish and email

Match the publishing mode to the reader's task

Start with the output you want. Ghost's email delivery guide defines these publishing modes, each with a different website and email outcome.

Choosing a mode does not settle website access or guarantee that a selected recipient will receive or read the message. Email only should not be treated as a confidential channel.

Use a tutorial and a roundup differently

In the hypothetical example, Publish only puts the tutorial on the website without emailing it. The writer can then prepare a separate roundup using Email only. That means reviewing two drafts, but it also leaves room to write an email that stands on its own.

Use Publish and email if the complete tutorial belongs in subscribers' inboxes. The editorial choice comes first. A shorter email is not automatically better, and duplicating the website post is not automatically useful.

Separate website access, newsletter subscriptions, and labels

Who may read the website post?

Set the tutorial's reading boundary explicitly. Public access and member restrictions answer a different question from who receives the email. Ghost's publishing guide describes previews for public visitors and free or paid members. Inspect the states that matter for this post.

For example, a roundup sent to free subscribers should explain if its main tutorial requires paid membership. Otherwise, change the tutorial's access to match the promise. Ordinary post access still matters when sharing a URL. Ghost separately supports publisher gift links: anyone using one can read that specific restricted post without signing up or paying. Treat a gift link as access-granting, not confidential. This example uses an ordinary public URL, not a gift link.

Who subscribes to this newsletter?

Check the specific newsletter. Ghost's recipient-selection documentation combines newsletter subscription with access-level and label constraints. Separate newsletters have their own subscriptions, settings, and templates, as the setup guide explains. A paid member's status alone does not tell you whether they subscribe to this one.

Suppose the roundup is intended for free subscribers to a developer newsletter who have a tutorial-interest label. Compare these hypothetical members against that intent.

During newsletter creation, check the existing-subscriber opt-in setting. Enabling it automatically subscribes existing members. Treat that as an enrollment decision, not a shortcut for fixing a recipient selection.

Where labels and member filters fit

Ghost's member-management guide documents saved filtered views and applying labels to filtered groups. Those views help administrators revisit a group. They are not evidence that every saved filter is available directly as a send segment.

An email unsubscribe does not itself remove website membership. Email disabled is different again. Ghost documents disabled addresses as unreachable or associated with spam-marked messages; inactivity alone is not the explanation given for that status.

Run the web-access-email preflight before sending

Write a delivery intent before opening the publish menu

The Optijara Web-Access-Email Preflight is an editorial checklist created for this guide. It is not an official Ghost feature or a tested automation. Begin with a plain-language delivery intent covering the website outcome and its access boundary. Specify the email audience separately. Name a correction owner and record the timing.

For this example, publish the tutorial publicly without email. Send the separate roundup to developer newsletter subscribers who match the free-member and label constraints. Public access avoids a mismatch here because the roundup promises a tutorial those readers can open. It is a choice for this example, not a rule for every publication.

The diagram shows the review sequence. It does not describe Ghost's internal execution.

Keep the hypothetical intent in a planning note like this JSON. It is not an API payload or evidence of an applied configuration.

{
  "exampleStatus": "hypothetical-unexecuted",
  "website": {
    "access": "public",
    "mode": "publish-only",
    "content": "tutorial"
  },
  "email": {
    "mode": "email-only",
    "content": "roundup"
  },
  "recipients": {
    "newsletter": "developer",
    "accessLevel": "free",
    "label": "tutorial-interest"
  },
  "timing": "site-timezone-review-required",
  "correctionOwner": "publication-editor"
}

If AI helps draft the roundup, approve its facts before approving delivery. Optijara's brand-safe editorial workflow guide covers that earlier review. Good copy still needs a recipient check.

Preview the member state and the email version

Use Ghost's preview options to inspect desktop and mobile presentation, then the email and relevant member states. Read the subject as well. Ghost defaults to the post title; its delivery guide documents subject customization from the preview area.

Read the email without relying on the website draft beside it. The opening should explain the roundup, and the tutorial link should match its promise. Check image alternatives and whether the message makes sense without the illustration. AI-generated visuals also need a content provenance review. Delivery checks cannot establish authenticity.

An authorized reviewer can send a test email to inspect inbox presentation. That test cannot validate the final audience filters or establish rendering across email clients. No test email was sent for this guide.

Handle shareable preview URLs carefully. Anyone supplied with one can view it, so it cannot prove membership enforcement. The eventual post restriction is not a reason to assume a confidential draft's preview is protected.

Check recipients, timing, and the final summary

Before approving publication, compare the draft with the intent using these checks.

Check the actual site timezone. Ghost's scheduling documentation places it in Ghost Admin Settings, General. The writer's local clock may differ.

Read the publishing summary against the written intent before approving it. Ghost's summary guidance explains how to return to settings or the preview. Any unexpected audience or output is a reason to stop. This walkthrough ends with a reviewed decision; it does not perform a send.

Know what can change after an email is sent

A website edit is not an email recall

Once the email has been delivered, treat it as a separate copy. Editing the website will not rewrite the inbox subject or body. Ghost's documented repair concerns tracked link destinations. Do not mistake that for message recall.

Use tracked-link updates within their limits

Ghost's link-update instructions start in the post's analytics, opened through its graph icon. Find Newsletter clicks and select the affected link. Change the destination, then click Update.

This requires email click tracking. Disabling tracking also disables link updates. The new destination applies to subsequent clicks, and the affected link's click count resets to zero. Record that reset when reviewing analytics. Earlier visits remain earlier visits; the feature does not establish a repair path for untracked links.

Choose an editorial correction when the content is wrong

Match the repair to what went wrong.

For a correction message, identify the mistaken statement and give the accurate replacement. Choose its audience deliberately. Subscribers to an unrelated newsletter do not automatically need it.

Common mistakes that change the audience or the message

Confusing a label with a subscription

A label that looks right is not enough. Neither is paid status or a saved member view. Check the selected newsletter's actual recipient constraints, using the delivery and membership documentation above. Subscription to a different newsletter does not establish inclusion in this one.

If the audience summary looks wrong, inspect the publishing selection first. Changing real member preferences to make the count look right is the wrong fix.

Reviewing the wrong artifact

A public preview cannot answer every member-access question. A test email cannot tell you who the final send targets. Review the evidence that answers the question in front of you.

For custom website experiences, Optijara's accessibility state-testing guide explains the limits of checking only the starting page. Its scope is web interaction testing. It does not certify email accessibility.

Read the default subject before accepting it. Follow the tutorial link using the relevant access state, and check the schedule against the site timezone. After sending, keep website corrections separate from inbox repairs.

Delivery and measurement limits to keep in view

Configured sending does not establish inbox placement

Newsletter sending is assumed to be configured. Ghost's deliverability guidance describes managed delivery for Ghost(Pro); a self-hosted installation needs a separate readiness check. This guide does not cover installation or DNS changes.

An approved audience is not a delivery guarantee. Ghost documents limits on inbox placement, and a Promotions tab is not the same as spam. Even a message that reaches the inbox may go unread.

Set the tracking policy before a mistake happens. Click tracking enables the documented link repair; disabling it removes that option. Consider the privacy trade-off deliberately rather than assuming repair will always be available.

Read analytics as signals, not proof of understanding

Read Ghost's post analytics guide before comparing results. Sent differs from unique-member Opens and Clicks. Website unique visitors differ from total views. Open and click rates use recipients as the denominator; repeated clicks by the same member do not add unique-member clicks.

Choose signals that answer the editorial question, without inventing an engagement benchmark.

For the hypothetical roundup, ask whether readers could reach the promised tutorial and where they got stuck. Read their responses alongside the counts. An open rate cannot answer that question on its own. Use what you learn to change the next draft.