Substack Analytics Explained: The Metrics That Matter and the Ones That Mislead

Substack analytics are useful when you separate five different questions: Did people encounter the work? Did subscribers open it? Did readers engage? Did the post attract subscribers? Did those subscribers stay or pay? No single metric answers all five.

Start with the outcome you wanted from the post, then choose the metric that matches it. Use views and traffic sources to study discovery, opens and delivery to assess subscriber reach, clicks and discussion to understand response, subscriptions to measure acquisition, and retention or revenue to evaluate the paid relationship. A large number can still be irrelevant to the decision in front of you.

Narrareach’s Substack analytics workflow keeps these signals beside the content and the next publishing decision. Substack should remain your source of truth for native subscriber, billing, retention, and directly reported post results.

The short guide to Substack metrics

This separation prevents the most common analytics mistake: treating attention, engagement, subscriber growth, and revenue as interchangeable forms of success.

What counts as a Substack view?

Substack defines total views as the number of times a published post is viewed across the web, email, and its app. Repeat views count. If one person views the same post five times, the total can rise by five. A view is therefore recorded attention, not a count of unique people who finished the article.

That distinction changes how you interpret a traffic spike. Ten thousand views could represent ten thousand individual readers, fewer people returning more than once, or a mixture of both. The number tells you that the post travelled. It does not tell you whether the audience understood the work, wanted the next issue, or became part of the publication.

Use views for questions such as:

• Which posts attracted the most attention over the same period?
• Did an older article begin receiving search or referral traffic?
• Did a distribution effort increase recorded visits?
• Which source sent people to a particular post?

Compare posts at the same age. A three-day-old article should not be judged against the lifetime views of an article published six months ago. Choose a practical window—perhaps the first seven days and the first 30 days—and use it consistently.

What does Substack open rate measure?

Open rate is the percentage of recipients who viewed the post after receiving it through email or the Substack app. A recipient who views the post several times counts once in that rate.

The word “open” sounds more exact than the measurement can be. Substack explains that email opens are detected when images load in a message. Images can load in a preview, while privacy settings or text-only email can prevent a genuine reading session from loading the tracking image. Opening a post after an app notification also contributes to the metric. Substack’s explanation of open tracking documents these limits and the inclusion of app activity.

An open rate is still useful because Substack applies a consistent platform definition. It is better for comparing similar sends inside your publication than for claiming an exact number of humans who read every word.

When open rate falls, inspect the numerator and denominator before rewriting every subject line. Suppose 600 of 1,000 recipients open one issue: the rate is 60%. A later issue reaches 1,500 recipients and records 750 opens: the rate falls to 50%, even though 150 more recipients opened it. The percentage weakened, but absolute opens grew by 25%.

The next question is whether the new subscribers continue opening later issues. If they came from a broad Note, a giveaway, an outside mention, or a general-interest post, their first interaction may not represent the publication’s usual subject. Subscriber-source and activity data can help distinguish a colder new cohort from a decline among established readers.

Delivery rate is a prerequisite, not an engagement score

Delivery rate describes the share of intended recipients who received the post through email or the app. If delivery deteriorates, an open-rate diagnosis begins too late in the funnel.

Check delivery when:

• a post reached far fewer subscribers than expected;
• open rate changed abruptly across several otherwise comparable sends;
• readers report that issues are missing;
• an imported or inactive audience behaves differently from established subscribers.

A delivered post has reached the intended channel. That does not mean it was noticed, opened, or read. Keep deliverability and editorial response separate so you do not try to fix an inbox problem with a stronger introduction—or a weak subject with list maintenance alone.

Engagement explains response, not necessarily growth

Likes, comments, restacks, shares, and link clicks answer different questions.

A like is a low-friction response. A comment can reveal what the reader understood, challenged, or wants next. A restack or share can expand distribution. A link click shows that an opener took the action connected to that link. None of those signals automatically means the person subscribed.

The distinction matters most with Notes. A short observation can earn enthusiastic agreement from people who enjoy it in the feed but have little reason to read a weekly publication. Another Note can attract fewer likes while sending qualified readers to an article that answers a specific question.

Substack provides individual Note statistics that can include clicks, shares, new followers, free and paid subscriptions, and revenue where applicable. Its current Notes guide is the correct reference for what the platform reports. Judge each Note according to its job:

• A conversation Note should produce relevant discussion.
• An article-introduction Note should give readers a clear reason to open the article.
• A publication-positioning Note should attract people who are likely to value future issues.
• A paid-offer Note should be assessed against paid interest, not applause alone.

This does not make likes meaningless. It stops a useful social response from being mistaken for subscriber conversion.

Subscriptions matter more when you preserve their source and quality

Substack’s post reporting can show free and paid subscriptions attributed to a post. Its Growth reporting connects subscriber movement with publishing activity and breaks sources into categories such as Google, Instagram, Substack onboarding, trackbacks, Direct, Direct to App, and other Substack surfaces. Substack’s current metrics guide defines these categories and explains the available publication, post, traffic, and growth views.

Source categories are evidence about an identifiable path. They are not perfect histories of every touchpoint. “Direct,” for example, can include a typed address, a bookmark, or a visit for which referral data was unavailable. “Direct to App” means an outside link opened content in the app without preserving the exact external source. Do not rename either category “loyal readers” or credit it to a campaign without additional evidence.

Subscriber count also needs a quality check. A source that produces many subscribers can be valuable, but ask what happens after acquisition:

• Do those subscribers open the next three comparable issues?
• Do they click into the subjects central to the publication?
• Do they remain subscribed after the initial interest fades?
• If paid subscriptions are offered, which sources or topics are associated with upgrades?

Substack’s subscriber dashboard includes source, recent activity, subscription type, revenue, and email-open fields that can be filtered or exported. This lets you examine a cohort without treating every subscriber as identical. The subscriber dashboard guide explains those current fields and segments.

Retention reveals whether acquisition created a durable relationship

Growth is the difference between people arriving and people leaving. A publication can add paid subscribers every week while barely changing its total if cancellations and expirations rise at the same time.

For publications with payments enabled, Substack’s Retention tab includes paid cohort analysis, paid net growth, paid churn, and free-subscriber retention. Cohorts show whether people who joined in the same period remain subscribed after later intervals. Substack’s retention documentation also distinguishes a cancellation initiated from a subscription that has actually expired.

Do not wait for a dramatic churn event before looking at retention. Compare cohorts associated with major changes in topic, acquisition source, publishing cadence, or paid offer. The comparison cannot prove that one change caused the difference, but it can expose a question worth investigating.

For example, if a large influx from a broadly popular Note remains subscribed but rarely opens later issues, the acquisition message may have promised something wider than the newsletter delivers. If a smaller search-driven cohort keeps opening related articles, that topic may express the publication’s continuing value more accurately.

Use conversion rates to compare unlike traffic volumes carefully

Raw subscription totals favor the post that received the most traffic. A simple conversion rate can reveal a quieter post that attracted a more aligned audience.

Consider a constructed example using the same 30-day observation window:

• Article A: 4,000 views and 20 free subscriptions.
• Article B: 900 views and 18 free subscriptions.

Article A produced more total subscribers. Its view-to-subscription rate was 20 ÷ 4,000 = 0.5%. Article B’s rate was 18 ÷ 900 = 2%. On this limited comparison, Article B converted recorded views at four times Article A’s rate.

That does not prove Article B is four times better. Repeat views can affect the denominator, source mixes may differ, and Substack may not identify every touchpoint. The calculation answers a narrower editorial question: which topic appears to give arriving readers a stronger reason to subscribe?

The practical response could be to keep Article A as a discovery asset while developing more work around Article B’s problem. Analytics should help different content perform different jobs; it should not force every article to chase the same ratio.

Five patterns and what to inspect next

Views rise while subscriptions stay flat

Find the posts and sources responsible for the additional traffic. Check whether their subject matches the publication promise, whether the page provides a clear path to subscribe, and whether the next issues continue the same reader interest. High traffic from an unrelated or unusually broad subject may be real without becoming a durable audience.

Subscribers grow while open rate falls

Compare absolute opens, recipient growth, and the sources of the new subscribers. Then inspect their activity across later issues. A changing denominator or colder acquisition source may explain the percentage before subject-line fatigue does.

Notes earn likes but few article clicks

Read the Note as a standalone reader. It may resolve the entire thought, attract agreement unrelated to the newsletter, or fail to explain what the article adds. Compare it with Notes created specifically to introduce a useful question, not with every Note in the feed.

A post has fewer views but more subscriptions

Preserve the topic and acquisition path. The post may be reaching a smaller group with stronger intent. Look for related questions that deepen the same need instead of rewriting it as a broad viral subject.

Paid subscriptions rise while net paid growth stalls

Open retention and churn reporting. New paid conversions can be offset by cancellations or expirations. Examine joining cohorts, renewal timing, offer changes, and the work published for paying readers before trying to solve the problem with acquisition alone.

Substack and Google Analytics will not produce identical totals

Substack now supports connecting outside analytics tools, including Google Analytics 4, through the Analytics section of publication settings. GA4 can add deeper web traffic, source, and conversion reporting. Substack’s current external analytics guide explains the connection.

The two systems still have different scopes. Substack can see native email, app, subscriber, recommendation, billing, and retention events that an outside web analytics tool may not observe in the same way. GA4 applies its own identity, session, consent, attribution, and event rules to the publication pages where its tag runs.

Use each tool for the question it can answer. Substack is authoritative for native subscriptions, app and email reporting, paid status, and platform-attributed outcomes. GA4 can deepen analysis of web journeys and external acquisition. Compare named events, date ranges, and directional patterns instead of forcing the top-line totals to agree.

Build a weekly review that ends with one decision

An analytics routine should change the next publishing cycle. A useful 20-minute review can follow this sequence:

1. Choose the intended outcome for the content you are reviewing: discovery, engagement, free acquisition, paid conversion, or retention.
2. Compare the same metric, content type, and observation window. Keep Notes with Notes and seven-day results with seven-day results.
3. Identify one meaningful pattern across several pieces, not merely the largest or smallest number.
4. Inspect the source, topic, format, audience, and subscription path that could explain the pattern.
5. Write one hypothesis and one change for the next article, Note, or offer.
6. Decide in advance when and how you will review the result.

In Narrareach’s Substack analytics workspace, available article, Note, traffic, engagement, and subscriber signals can sit beside the next draft and schedule. A practical workflow is to select a comparable group, preserve the metric and date context, choose the strongest supported topic or format, prepare the follow-up, review it, publish it, and return to the same measurement window.

Narrareach does not replace Substack billing or subscriber administration, reveal Substack’s private recommendation logic, or turn nearby activity into proven attribution. Its role is to keep the evidence close enough to the editorial workflow that the next action remains connected to what you actually observed.

Open your last eight articles and assign each one a primary job: discovery, engagement, acquisition, paid conversion, or retention. Compare only the metric that matches that job over the same period. Then choose one pattern strong enough to influence the next piece. The goal is not to find the most flattering number in the dashboard. It is to make one better decision than the numbers allowed you to make last week.