> ## Content Index
> Fetch the complete content index at: https://narrareach-blog.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# What Is a Good Medium Read Ratio or Feed Clickthrough Rate?
- URL: https://narrareach-blog.ghost.io/good-medium-read-ratio-feed-clickthrough-rate/
- Published: 2026-09-19T12:56:53.000Z
- Updated: 2026-09-19T12:56:53.000Z
- Author: Ian Kiprono
- Tags: Medium, Analytics

There is no universal good Medium read ratio or feed clickthrough rate. Medium explicitly recommends comparing stories within your own catalogue because topic, audience, voice, distribution and traffic mix change what the percentages mean. Feed clickthrough measures how often a feed presentation leads to an opening. Read ratio measures how often a view lasts at least 30 seconds. A useful benchmark therefore compares stories with similar purposes, subjects, ages and sources—and keeps the raw counts beside each percentage.

My view is that writers should stop asking whether one percentage is “good” and ask whether it is strong enough for the job assigned to that story. A low clickthrough points toward packaging or audience fit. A low read ratio points toward the promise-to-opening transition. Neither number, by itself, grades the whole article. Narrareach can keep the available story counts beside the next draft, while Medium remains the source of truth for its native rates and distribution details.

## Read the two percentages as separate decisions

| Metric                 | Calculation                                            | Best question                                                     | Common mistake                           |
| ---------------------- | ------------------------------------------------------ | ----------------------------------------------------------------- | ---------------------------------------- |
| Feed clickthrough rate | Feed-attributed openings divided by feed presentations | Does the story earn an opening when Medium shows it in feeds?     | Dividing all views by all presentations  |
| Total read ratio       | Reads divided by views                                 | How often does a landing reach Medium's 30-second read threshold? | Treating a read as completion            |
| Member read ratio      | Member reads divided by member views                   | How do eligible member visits progress into member reads?         | Substituting it for the total read ratio |

Medium says feed clickthrough appears only after enough data accumulates. An absent rate is therefore unavailable evidence, not a zero. The platform also says there is no universal good or bad value for either feed clickthrough or read ratio. Its advice is to compare your own stories over longer periods.

That boundary matters. A percentage becomes useful only when its numerator, denominator and context match the decision you are making.

## What Medium's feed clickthrough rate actually measures

Medium's [current Stats FAQ](https://help.medium.com/hc/en-us/articles/360024113314-Frequently-asked-questions-about-Medium-Stats?ref=narrareach-blog.ghost.io) defines feed clickthrough as the percentage of people who opened a story after seeing it in a feed on the app or website. If a story receives 1,000 feed presentations and 50 feed-attributed openings, its feed clickthrough rate is 5%.

The denominator does not include every presentation surface. Medium presentations can also occur in search results, topic pages, publications, profiles, recommended reads, subscriber notifications and lists. External exposure on Google or social platforms is outside the presentation count. Total views can therefore exceed presentations, and total views divided by all presentations will not reproduce the feed rate.

Use feed clickthrough to examine the package a reader sees before opening: subject, title, subtitle, image, publication context and the audience Medium selected for that presentation. It does not tell you whether the article satisfied the promise after the click.

A lower rate is not automatically a failure. Medium notes that feed clickthrough can fall as presentations expand to a broader audience. A story shown to 300 highly aligned followers may earn a larger percentage than the same story shown to 30,000 people with mixed interests. The second result can still produce far more readers.

Consider two constructed results:

| Story | Feed presentations | Feed openings | Feed clickthrough | What the count adds                         |
| ----- | ------------------ | ------------- | ----------------- | ------------------------------------------- |
| A     | 500                | 50            | 10%               | Strong rate, 50 feed openings               |
| B     | 10,000             | 500           | 5%                | Lower rate, ten times as many feed openings |

Calling A the winner ignores scale. Calling B the winner ignores audience fit. The right interpretation depends on whether you are diagnosing packaging efficiency, total reading reach or a specific audience segment.

## What Medium's read ratio tells you

On the [detailed story stats page](https://help.medium.com/hc/en-us/articles/34831991136151-Story-s-detailed-stats-page?ref=narrareach-blog.ghost.io), Medium counts a read when someone stays with a story for at least 30 seconds. Total read ratio is reads divided by views. A story with 200 views and 80 reads has a 40% read ratio.

That 40% does not mean 40% reached the final paragraph. It means 40% crossed the platform's reading threshold. The metric is most useful for finding a weak transition between arrival and sustained attention.

Traffic mix changes that transition. A search visitor may land on one paragraph for a narrow answer. A follower may arrive because they already understand the writer's subject. A social click can be more casual. A paywall or friend link changes what different visitors can access. These visits can enter the same total ratio while representing different intentions.

Story length also complicates the interpretation. The threshold remains 30 seconds whether the displayed reading time is three minutes or twelve. A read on the longer article still does not establish completion. Comparing a concise definition with a long reported essay as though the percentage measured the same experience creates false precision.

## Current Narrareach data shows why a single benchmark misleads

A September 19, 2026 snapshot of Medium data available in Narrareach contained 321 distinct stories published from January 1, 2025 onward across 15 connected accounts whose caches had refreshed within the previous 48 hours. Eighty-seven stories had at least 20 lifetime views and valid read counts. Within that selected group, the 25th percentile read ratio was 21.3%, the median was 28.6%, and the 75th percentile was 37.5%.

Those figures describe this product snapshot, not Medium as a whole. The accounts are self-selected, the sample is small, story ages differ, and lifetime totals combine different sources and publishing circumstances. They are unsuitable as a target for another writer. Their useful lesson is the spread: even after removing the smallest denominators, one number does not describe every story's context.

This is why I would use external benchmarks only as prompts for investigation. A claim that every story below 40%, 50% or 60% is bad strips away the evidence needed to decide what to change. Medium itself publishes no such threshold.

## Wait until the percentage can support a decision

Small denominators make ratios jump. With five views, one additional read changes the ratio by 20 percentage points. With 50 views, the change is two points. With 500 views, it is 0.2 points.

The same problem applies to feed clickthrough. Two openings from 20 feed presentations produce 10%. Twenty openings from 200 presentations also produce 10%, but the second result is less vulnerable to one person's decision. Neither is a controlled experiment because audience composition can still differ.

Medium does not publish a fixed minimum sample for these comparisons. It says the feed rate appears after enough data and warns against decisions based on only a few presentations. A practical response is to preserve the count, wait several days, and label the result preliminary until the denominator is large enough that one or two events no longer rewrite the conclusion.

Do not switch titles and images repeatedly during that wait. Medium warns that rapid changes make comparisons difficult because each version may reach a different audience. Record one meaningful change, its date and the period you will observe.

## Build a benchmark from comparable stories

Start with five to ten of your own stories that performed the same job. A useful group might contain beginner tutorials intended for Medium discovery, or essays sent to an established subscriber audience. Avoid mixing a search reference, a personal essay and a news reaction simply because all three were published in the same month.

For each story, record:

| Field                                            | Why it belongs                                              |
| ------------------------------------------------ | ----------------------------------------------------------- |
| Publication date and observation date            | Separates a seven-day result from a one-year lifetime total |
| Story purpose                                    | Defines what success should mean                            |
| Topic and publication                            | Reveals major audience differences                          |
| Feed presentations and reported feed rate        | Preserves the correct exposure scope                        |
| Total views, reads and read ratio                | Shows the arrival-to-threshold transition                   |
| Traffic-source mix                               | Flags stories dominated by outside visits                   |
| Followers or subscribers attributed to the story | Adds the next audience action                               |

Compare medians rather than choosing the single best result. One exceptional story can distort an average, especially in a small catalogue. Then look for a pattern that can lead to an editorial decision: perhaps tutorials with precise outcome-led titles earn more feed openings, while their long background sections lose readers before the first example.

The comparison still does not prove causation. Publication placement, recommendation breadth, seasonality and reader familiarity can change at the same time. Treat the pattern as a reason to run a restrained editorial test.

## Diagnose clickthrough before rewriting the article

When feed clickthrough is weak relative to comparable stories, examine what readers see before opening.

First, test whether the title identifies a concrete problem or result. “Some Thoughts on Better Writing” asks the reader to supply the relevance. “How to Compare Two Medium Headlines Without Misreading Clickthrough” names the task. Specificity helps only when the article delivers the promised answer.

Next, check topic and publication fit. A clear title can still perform poorly when presented to people who do not need the subject. Medium's presentation system uses reader interests and history, while publications bring their own audience context. A mismatch is not evidence that the title is universally weak.

Finally, inspect the image and subtitle as part of one package. They should add information rather than repeat the headline or manufacture curiosity. Medium explicitly advises against clickbait because a click followed by disappointment weakens the reading stage and can limit distribution under its policies.

## Diagnose read ratio from the click onward

When read ratio is weak relative to similar work, compare the title with the first 150 words. Does the opening deliver the expected answer, or begin with unrelated autobiography, throat-clearing or a long definition the reader did not request?

Then inspect the route to the first useful example. A technical reader may tolerate necessary setup when the article explains why it matters. They are less likely to stay through repeated promises that the answer is coming. Move the decision framework or central distinction earlier, then use the rest of the article for evidence, exceptions and application.

Do not shorten every story to chase the 30-second threshold. A higher ratio can accompany a less useful article if the writer removes the depth readers needed. Preserve the mechanism, examples and objections. Cut repetition and delayed answers.

Also keep total read ratio separate from member read ratio. Medium defines the latter as member reads divided by member views and uses it within Partner Program earnings adjustments. A story can have different total and member ratios because the two populations and access conditions differ. Neither percentage can be substituted for the other.

## Turn the comparison into one publishing experiment

[Narrareach's Medium publishing workflow](https://www.narrareach.com/integrations/medium?ref=narrareach-blog.ghost.io) can keep the available presentations, views and reads beside the next article you prepare. Use Medium's native detailed report for the definitive feed rate, traffic sources, member ratio and distribution status.

Choose a comparable set, write down the earliest weak stage, and create one hypothesis. If clickthrough is the issue, you might write: “The next tutorial title will name the decision and the constraint without changing the subject or publication.” If read ratio is the issue: “The next guide will answer the query in its first 120 words and move the worked example above the background section.”

Draft the story, review whether the title and opening make the same promise, and check the image, links, formatting, paywall and destination before publishing. Schedule it when the piece is ready. After a comparable period, record counts and percentages beside the earlier cohort. Narrareach can support the measure-to-draft-to-review-to-publish loop, but it cannot prove that one edit caused Medium's distribution response.

For the broader relationship among presentations, views, reads and earnings, use the [complete Medium analytics guide](https://narrareach-blog.ghost.io/medium-analytics-explained/).

Take one story with enough data today. Write its feed presentations, feed clickthrough, views, reads and read ratio on one line. Add its purpose and largest traffic source. Compare it with three similar stories, then choose the first stage where the evidence repeatedly weakens. That is a more useful benchmark than a percentage borrowed from someone else's account.