YouTube Benchmarks 2026: Views, Engagement, Growth & Subscriber Data From 6 Studies

YouTube Benchmarks 2026: Views, Engagement, Growth & Subscriber Data From 6 Studies

What is a good YouTube view count?

What engagement rate should you aim for?

How many views should a video have after seven days?

What percentage of your subscribers should watch?

How fast should a healthy channel grow?

Search for any of those questions and you will find a benchmark.

The problem is that most YouTube benchmarks are treated as universal when they are actually measuring different:

• channel sizes
• video formats
• video ages
• formulas
• audiences
• research samples

A 5% views-to-subscriber ratio can look weak for one channel and completely normal for another.

A 5% engagement rate can be strong.

A 5% monthly subscriber growth rate can be modest for a tiny breakout channel and enormous for an established million-subscriber creator.

And a video with:

100,000 views

can be either:

• a major breakout
• normal
• or a severe underperformer

depending on what that channel normally achieves.

So instead of publishing another universal "good YouTube metrics" chart, OverseerOS combined findings from six of our public-data studies into one benchmark framework.

The underlying research includes:

• 2,580 mature long-form videos for average-vs-typical view performance
• 2,043 recent videos for views-to-subscriber ratios
• 2,251 recent videos for engagement benchmarks
• 436 long-form uploads measured around day 7
• 1,051 long-form uploads measured around day 30
• 1,035 long-form uploads measured around day 90 in the latest cohort
• 89 repeatedly observed channels for subscriber-growth analysis

These samples overlap in places and should not be added together as one giant unique-video dataset.

They answer different questions.

Taken together, however, they reveal one consistent rule:

The most useful YouTube benchmark is not one platform-wide number. It is a benchmark matched to the video's age, format, channel size, and the channel's own recent baseline.

This report gives you those reference points.

Key YouTube Benchmarks for 2026

Here are the headline findings.

Do not read the 7-day, 30-day, and 90-day medians as one video's growth curve.

They come from different cohorts.

A video cannot lose views from day 7 to day 30, but a separate day-30 sample can have a lower median because it contains different channels and videos.

That distinction is critical.

The Direct Answer: What Are Good YouTube Benchmarks?

The strongest benchmarks our research supports are:

Typical video performance

Use the median of your own comparable videos, not the arithmetic average.

Strong long-form engagement

Around:

5% likes + comments per view

was approximately top-quartile performance in our recent sample.

Exceptional long-form engagement

Around:

7.3%

placed a channel near the top decile.

Recent long-form views relative to subscribers

Our channel-weighted median was:

11.9%.

But channel size changed this dramatically.

Seven-day views

Use a channel-size benchmark for external context, then compare against your own same-age baseline.

30-day views

Same rule.

90-day views

Same rule.

Subscriber growth

There is no universal good monthly percentage because growth rates naturally change as channels get larger.

The benchmark hierarchy should be:

your channel → same format → same video age → similar channel size → broad external benchmark

Not the reverse.

Benchmark 1: Use Median Views, Not Average Views

This is the most important baseline rule.

OverseerOS analyzed:

2,580 mature long-form videos across 67 channels.

We compared each channel's:

• arithmetic mean views
• median views

The median channel had a mean that was:

1.97x its median.

In other words:

The statistic most creators call "average views" was almost twice what the typical video actually achieved.

Why?

Because YouTube performance was heavily concentrated in a small number of winners.

On the median channel:

• top single video generated 19.6% of qualifying views
• top 10% generated 48.7%
• top 20% generated 64.6%

That makes the arithmetic mean extremely sensitive to viral videos.

Example

Suppose your videos get:

• 8K
• 9K
• 10K
• 11K
• 112K

Mean:

30K

Median:

10K

If you tell yourself:

My average is 30K, so my next video should get 30K

you have created an unrealistic baseline.

Most of the channel's uploads do not get 30K.

The winner inflated the number.

Mature Long-Form Performance Multiples

In that 2,580-video study, the median channel had these thresholds:

These are descriptive reference points, not YouTube algorithm thresholds.

Still, they provide a much stronger vocabulary than:

good views

or:

bad views.

If your channel median is:

20,000

then:

92,400 views

would equal:

4.62x baseline.

That is much more strategically meaningful than saying:

It got 92K views.

For the full analysis, see What Is a Good Average View Count on YouTube?.

Benchmark 2: YouTube Views-to-Subscriber Ratio

One of the most common creator benchmarks is:

What percentage of my subscribers should watch each video?

The problem is that video views are not restricted to subscribers.

A video can reach:

• subscribers
• returning non-subscribers
• Search viewers
• Browse viewers
• Suggested viewers
• entirely new audiences

So the public views-to-subscriber ratio is better understood as:

video reach relative to channel subscriber size

not:

percentage of subscribers who watched.

Overall Recent Benchmark

OverseerOS analyzed:

2,043 videos across 273 channels

measured when the videos were:

7 to 30 days old.

Channel-weighted median:

Long-form

11.9%

Short-form, 3 minutes or less

21.7%

But those pooled numbers hide the most important result.

Channel size changed the ratio dramatically.

Long-Form Views-to-Subscriber Benchmarks by Channel Size

The under-10K figure was extremely volatile and should not become a target.

The middle 50% in that group ranged from:

52.4% to 400.9%.

That instability is exactly why small channels should not obsess over one views-to-subscriber percentage.

Why a 5% Ratio Can Be Fine

Imagine:

Channel A

30K subscribers
1,500 recent views

Ratio:

5%

Channel B

3M subscribers
150K recent views

Ratio:

5%

Same arithmetic.

Different competitive context.

In our sample, a 5% long-form ratio was far below the median for 10K to 99K channels but close to the:

5.2%

median among 1M+ channels.

A universal benchmark would diagnose both channels the same way.

The data says that is too simplistic.

Read the complete study at YouTube Views-to-Subscriber Ratio.

Benchmark 3: YouTube Engagement Rate

For public-data research, OverseerOS defines visible engagement rate as:

(likes + comments) ÷ views × 100

We analyzed:

2,251 recent videos across 309 channels

measured between:

7 and 30 days old.

Long-Form Engagement Benchmarks

Short-Form Engagement Benchmarks

The useful shorthand from this sample is:

Around 3% to 4% was typical, around 5% was strong, and around 7% sat near the top decile.

But this metric needs context too.

Engagement Does Not Automatically Fall With Channel Size

Our long-form median engagement rates were:

• Under 10K: 3.41%
• 10K to 99K: 3.44%
• 100K to 999K: 3.66%
• 1M+: 3.33%

That pattern was surprisingly stable.

Subscriber size barely explained long-form engagement rate in this particular sample.

That makes engagement different from views-to-subscriber ratio, where channel size mattered enormously.

Different metrics behave differently.

That is why one generic "channel health score" can be misleading.

For full methodology, see What Is a Good YouTube Engagement Rate?.

Benchmark 4: Like Rate

Engagement becomes easier to diagnose when separated into components.

For recent long-form videos:

Median public like rate:

3.11%.

That equals roughly:

31 likes per 1,000 views.

For short-form videos of 3 minutes or less:

Median:

2.60%.

Approximately:

26 likes per 1,000 views.

Do not interpret that as a recommendation threshold.

There is no evidence that crossing:

3% likes

automatically unlocks distribution.

Use it as a descriptive audience-response benchmark.

Benchmark 5: Comment Rate

Comments require more effort than likes.

So the rate was dramatically smaller.

Long-form median

0.238%

That equals approximately:

2.38 comments per 1,000 views.

Short-form median

0.032%

Approximately:

0.32 comments per 1,000 views.

That is why comparing comment rate with like rate directly is rarely useful.

They measure different levels of effort.

Benchmark 6: Views After 7 Days

OverseerOS tracked:

436 recent long-form uploads across 232 channels

as close as possible to day 7.

Median measurement age:

6.96 days.

Overall pooled median:

28,604 views.

But the channel-size breakdown is much more useful.

The ranges are huge.

For 1K to 9.9K channels:

P25:

604

P75:

28,795

That is nearly a:

48x spread.

So even "same subscriber tier" is not enough to create a precise expectation.

Seven-Day Views Relative to Subscribers

The small-channel numbers are particularly unstable because the denominator is tiny.

That does not make them useless.

It means you should treat them as:

context

not:

targets.

Read the full 7-Day YouTube View Benchmark Study.

Benchmark 7: Views After 30 Days

A separate OverseerOS cohort analyzed:

1,051 long-form uploads across 486 channels

measured between:

28 and 32 days after publication.

Median measurement age:

29.92 days.

Overall pooled median:

27,831 views.

Again, do not compare this pooled number directly with the separate 7-day sample as though they are the same videos.

30-Day View Benchmarks by Channel Size

One thing should be obvious from every row:

variance remains massive.

A million-subscriber channel at the 90th percentile had:

3.71 million views

versus:

63,847

at P25.

Same broad subscriber tier.

Almost:

58x difference.

Read the complete 30-Day YouTube View Benchmark Study.

Benchmark 8: Views After 90 Days

Our latest approximately 90-day long-form cohort contained:

1,035 videos across 404 channels.

Median observation age:

89.82 days.

Overall distribution:

90-Day Benchmarks by Channel Size

These numbers reinforce the same lesson.

There is no defensible platform-wide statement like:

Every video should have 100K views after three months.

For a small channel, 100K might be an extraordinary breakout.

For a huge channel, it may be a major miss.

Read the 90-Day YouTube View Benchmark Study.

Do Not Turn 7-Day, 30-Day, and 90-Day Tables Into a Growth Curve

This mistake is easy to make.

You might notice:

• 7-day median: 28,604
• 30-day median: 27,831
• 90-day median: 45,608

Then conclude:

Videos somehow declined between day 7 and day 30.

They did not.

These are:

different research cohorts.

The correct interpretation is:

The typical observed video in the independent day-7 sample had 28,604 views, while the typical observed video in the independent day-30 sample had 27,831.

You can only measure actual view accumulation by following:

the same video

through multiple checkpoints.

That is an important general rule for benchmark research.

Benchmark 9: Monthly Subscriber Growth

What is a good YouTube subscriber growth rate?

Again:

there is no universal percentage.

OverseerOS analyzed a strict cohort of:

89 repeatedly observed channels

with at least seven days between comparable snapshots.

Median observation window:

15.7 days.

We normalized observed growth to a monthly-equivalent pace.

Overall distribution:

Those numbers are unusually high compared with broad all-channel studies because this is a selected OverseerOS research cohort containing channels discovered through research and breakout workflows.

Do not use:

3.10%

as a universal "normal YouTube" target.

The structural findings are more useful.

Channel Size and Growth Rate

Starting channel size had a:

-0.683 Spearman relationship

with percentage subscriber growth.

Larger channels tended to grow more slowly in percentage terms.

That makes mathematical sense.

Going from:

1,000 to 1,500

is:

50% growth.

Going from:

1 million to 1.05 million

is:

5%.

The second channel added:

50,000 subscribers

while posting a much lower percentage.

Percentage growth and absolute growth answer different questions.

Growth and Views Moved Together

Subscriber growth and public view growth had a:

0.826 Spearman relationship

in the strict cohort.

This does not prove:

view growth caused subscriber growth.

They can share underlying causes such as:

• stronger topics
• breakouts
• increased publishing
• audience expansion

But it supports an important diagnostic rule:

Judge subscriber growth alongside view growth, not alone.

Read What Is a Good YouTube Channel Growth Rate?.

Benchmark 10: What Counts as a YouTube Outlier?

A video can have:

1 million views

and still be ordinary for one channel.

Another can have:

50,000

and represent a huge breakout.

That is why outliers should be defined relative to channel baseline.

The simplest formula is:

Outlier multiple = video views ÷ comparable channel median

If the normal channel median is:

20K

then:

20K

1x

Typical.

40K

2x

Strong.

100K

5x

Major outlier.

200K

10x

Extreme channel-relative winner.

Our mature long-form study found the median channel's:

• top-quartile threshold around 2.14x
• top-decile threshold around 4.62x
• top-5% threshold around 7.33x

Those are much more useful for competitor research than one raw view threshold.

The Most Useful YouTube Benchmark Table

Use the metric that matches the question.

This framework avoids most benchmark mistakes immediately.

Benchmark Hierarchy: What Should You Compare Against First?

Not all benchmarks deserve equal weight.

1. Your Own Historical Baseline

Strongest.

It controls for:

• audience
• niche
• creator
• style
• production
• channel history

2. Same Format

Do not compare:

• Shorts
• long-form
• livestreams

as if they behave identically.

3. Same Video Age

A seven-day video should be compared with:

seven-day videos.

Not lifetime results.

4. Similar Channel Size

Subscriber count changes several public ratios.

5. Similar Niche

Audience behavior can differ by topic.

6. Broad YouTube Benchmark

Useful as final context.

Not your primary target.

Why External Benchmarks Disagree

Two credible benchmark studies can report different numbers without either one being fraudulent.

Differences can come from:

• sample selection
• channel-size distribution
• country
• niche
• format
• measurement age
• formula
• channel weighting
• view-count methodology
• date of collection

For example:

engagement rate by views

is not the same as:

engagement per subscriber.

A study of:

recent active channels

is not the same as:

all channels with one subscriber.

A study of:

successful competitor channels

is not representative of every channel uploaded to YouTube.

Methodology is part of the benchmark.

Why You Should Always Ask for the Denominator

A percentage without a denominator is dangerous.

5% engagement

Five percent of what?

• views?
• subscribers?
• impressions?

20% views-to-subscriber ratio

At what video age?

• 24 hours?
• 30 days?
• lifetime?

4% growth

Over what period?

• seven days?
• 30 days?
• a normalized short window?

Before using a benchmark, define:

numerator + denominator + time window + comparison group.

The Five Benchmark Mistakes That Ruin YouTube Analysis

Mistake 1: Comparing Raw Views Across Channel Sizes

100K views means different things at:

• 5K subscribers
• 500K
• 5M

Normalize first.

Mistake 2: Using Average Views When the Distribution Is Skewed

Our mature long-form study found the arithmetic mean was:

1.97x

the median on the median channel.

Use the median for typical performance.

Mistake 3: Mixing Video Ages

A day-2 video and a day-90 video are not comparable.

Views accumulate.

Choose one checkpoint.

Mistake 4: Mixing Formats

Short-form and long-form showed different:

• engagement
• reach
• view behavior

Separate them.

Mistake 5: Treating External Percentiles as Algorithm Thresholds

P75 means:

upper quartile in the observed sample.

It does not mean:

YouTube rewards this exact number.

Benchmarks are descriptive.

They are not platform rules.

The Better Way to Audit Your Channel

Here is a practical benchmark workflow.

Step 1: Choose 20 Comparable Videos

If possible, use:

20

rather than only 5 or 10.

Our mature-video research found latest-20 medians were materially closer to fuller channel baselines than latest-10 medians.

Step 2: Separate Formats

Build different baselines for:

• long-form
• Shorts
• live

Step 3: Choose a Fixed Age

Examples:

• day 7
• day 30
• day 90

Step 4: Calculate Median Views

This is your typical baseline.

Step 5: Calculate P25, P75, and P90

Now you have a performance distribution.

Step 6: Calculate Views-to-Subscriber Ratio

Use it as size context.

Step 7: Calculate Engagement

Use:

(likes + comments) ÷ views × 100

when comparing public videos consistently.

Step 8: Track Growth Separately

Subscriber growth is a channel-level outcome.

Do not mix it into the video benchmark without context.

Step 9: Investigate Outliers

Find videos at:

• 2x
• 5x
• 10x

your normal baseline.

Then inspect:

• topic
• title
• thumbnail
• structure
• timing

Step 10: Repeat Monthly

A benchmark becomes stale when the channel changes.

Your Channel Can Outgrow Its Own Benchmark

Suppose your 90-day median used to be:

20K.

Then recent comparable videos begin reaching:

• 31K
• 35K
• 39K
• 42K
• 48K

Your old 20K benchmark is no longer the right baseline.

That is good.

Benchmarks should update with the channel.

They are measurement systems.

Not permanent standards.

Why Recent Benchmarks Can Be Better Than Lifetime Benchmarks

A channel can:

• change niche
• improve thumbnails
• shift formats
• post less often
• accelerate
• decline

Lifetime performance averages all those eras together.

That can hide the strategy currently working.

For current decisions, recent comparable videos usually deserve more weight.

For historical context, lifetime data remains useful.

When Lifetime Data Is Useful

Use lifetime statistics for questions like:

• How much total reach has this channel accumulated?
• How large is the catalog?
• Which historical videos dominated?
• How concentrated are lifetime views?

Do not use lifetime averages alone to answer:

What should the next upload get?

Those are different questions.

Public Benchmarks vs Your Private YouTube Analytics

Public research can benchmark:

• subscribers
• video views
• likes
• comments
• upload dates
• public video count

Your own YouTube Analytics can go much deeper.

For your channel, you should also inspect:

• impressions
• CTR
• watch time
• average view duration
• retention
• returning viewers
• unique viewers
• traffic sources

That private data should outrank a public benchmark when diagnosing your own videos.

Public benchmarks are especially valuable for:

competitor research.

How OverseerOS Helps Benchmark a YouTube Channel

The free YouTube Channel Analyzer helps establish public channel context including:

• channel statistics
• recent uploads
• top-performing public videos
• publishing patterns

The useful research process is:

establish baseline → find outliers → inspect the pattern → compare similar channels → create something original

Viral Channel Finder can help discover breakout channels worth benchmarking.

Channel Blueprint Cloner can then help turn those public patterns into an original channel strategy rather than copying individual videos.

A Simple YouTube Benchmark Scorecard

Do not combine these into one mysterious score.

Keep them visible.

Typical performance

Median same-age views:

Strong performance

P75:

Exceptional performance

P90:

Views-to-subscriber ratio

____%

Engagement rate

____%

Recent subscriber growth

____%

View growth

____%

Strongest recent outlier

____x baseline

Now you can understand why a channel looks strong.

How to Know Whether a Video Is Actually Good

A useful decision hierarchy:

Around 1x your median

Normal.

1.25x to 2x

Promising.

Around 2x+

Strong enough to inspect.

5x+

Serious breakout candidate.

10x+

Extreme relative winner.

Do not copy the video.

Research:

What changed?

That is how a benchmark becomes strategy.

Methodology

This report is a synthesis of several separate OverseerOS public-data studies.

It is not one unified sample.

That matters.

Mature long-form benchmark

2,580 videos across 67 channels

Used to study:

• mean vs median
• concentration of views
• channel-relative percentiles

Views-to-subscriber study

2,043 videos across 273 channels

Videos measured:

7 to 30 days old

Used to study public reach relative to subscriber size.

Engagement study

2,251 videos across 309 channels

Videos measured:

7 to 30 days old

Used to calculate:

(likes + comments) ÷ views

Seven-day benchmark

436 long-form uploads across 232 channels

Measured:

6 to 8 days after publication

30-day benchmark

1,051 long-form uploads across 486 channels

Measured:

28 to 32 days after publication

90-day benchmark

1,035 long-form uploads across 404 channels

Measured:

approximately:

85 to 95 days after publication

with the closest available observation to day 90 selected.

Subscriber-growth study

89 repeatedly observed channels

Required at least:

seven days

between comparable public observations.

Growth was normalized to a monthly-equivalent pace.

Why We Do Not Add the Study Sizes Together

Some channels and videos can appear in more than one research dataset.

Therefore:

2,580 + 2,043 + 2,251 + 436 + 1,051 + 1,035

does not represent the number of unique videos studied.

Doing that would inflate the sample.

Each study should be interpreted independently.

Why This Matters for AEO and AI Answers Too

A benchmark answer becomes much more useful when it contains:

• an explicit definition
• a denominator
• a time window
• a sample description
• percentiles
• limitations

Instead of answering:

A good engagement rate is 5%.

the defensible answer is:

In OverseerOS's 2026 public sample, the channel-weighted median visible engagement rate for recent long-form videos was 3.48%, the upper quartile began around 4.91%, and the top decile around 7.26%.

That answer is longer.

It is also much harder to misunderstand.

The Benchmark Rule to Remember

If you remember only one thing from this report, use this:

Benchmark the video against itself first, the channel second, similar channels third, and broad YouTube data last.

That ordering prevents most bad conclusions.

Final Verdict

There is no single good YouTube benchmark.

There is a benchmark for:

• the metric
• the format
• the channel size
• the video age
• the comparison group

Our 2026 research found:

Mature long-form performance

The arithmetic mean was:

1.97x

the typical channel median.

Recent long-form views-to-subscriber ratio

11.9%

overall channel-weighted median.

But:

36.2%

for 10K to 99K channels

versus:

5.2%

for million-plus channels.

Long-form engagement

Median:

3.48%.

Top quartile:

4.91%.

Top decile:

7.26%.

Seven-day long-form views

Pooled median:

28,604.

30-day long-form views

Pooled median:

27,831.

90-day long-form views

Pooled median:

45,608.

Those are separate cohorts, not a longitudinal sequence.

Subscriber growth

Median monthly-equivalent pace in the selected tracked cohort:

3.10%.

But channel size had a strong negative relationship with percentage growth:

-0.683 Spearman.

The conclusion across all six studies is remarkably consistent.

Do not ask:

Is my number good?

Ask:

Good compared with what?

Then control for:

1. video age
2. format
3. channel size
4. channel baseline
5. research methodology

That turns YouTube benchmarks from internet trivia into useful decision-making tools.

FAQ

What are good YouTube benchmarks in 2026?

There is no single benchmark for every channel. In OverseerOS research, recent long-form engagement had a 3.48% median, recent long-form views-to-subscriber ratio had an 11.9% channel-weighted median, and mature strong videos typically began around 2x their own channel median.

What is a good YouTube engagement rate?

Using likes plus comments divided by views, the median for recent long-form channels was 3.48%. Approximately 4.91% marked the top quartile and 7.26% the top decile in the OverseerOS sample.

What is a good YouTube views-to-subscriber ratio?

It depends strongly on channel size. For recent long-form videos, the sample median was 36.2% at 10K to 99K subscribers, 11.3% at 100K to 999K, and 5.2% at 1M+.

How many views should a YouTube video get after seven days?

There is no universal target. In the 436-video long-form study, the pooled median was 28,604 views, but subscriber-band medians ranged from 40 for sub-1K channels to 254,270 for million-plus channels.

How many views should a YouTube video get after 30 days?

The pooled median in the separate 1,051-video study was 27,831. Channel-size medians ranged from 45 for sub-1K channels to 292,667 for million-plus channels.

How many views should a YouTube video get after 90 days?

The pooled median in the latest 1,035-video cohort was 45,608. The median was 12,314 for 10K to 99K channels, 74,089 for 100K to 999K channels, and 364,233 for 1M+ channels.

Why is the 30-day pooled median lower than the 7-day median?

They are different samples. The numbers should not be interpreted as one video's growth curve. Cross-sectional benchmark cohorts can have different medians.

What is a good monthly YouTube subscriber growth rate?

There is no universal percentage. In a selected 89-channel OverseerOS cohort, median monthly-equivalent growth was 3.10%, but percentage growth fell strongly as starting channel size increased.

Is 5% engagement good on YouTube?

In OverseerOS's recent public-video sample, roughly 5% visible engagement was around top-quartile performance for both long-form and short-form groups.

Is a 10% views-to-subscriber ratio good?

It depends on channel size and video age. Ten percent would be below the recent long-form median for many smaller channels while sitting above the 5.2% median observed among million-plus channels.

Should I use average or median YouTube views?

Median is usually the stronger measure of typical video performance. In OverseerOS's mature long-form study, the mean was 1.97 times the median on the median channel because a small number of winners generated a large share of views.

How many recent videos should I use for a YouTube benchmark?

Use around 20 comparable videos when possible. In OverseerOS's mature long-form study, latest-20 medians were materially more stable than latest-10 medians relative to fuller channel histories.

What makes a YouTube video an outlier?

A simple definition compares the video's views with the median of comparable channel videos. Around 2x baseline was strong, while the median channel's top-decile threshold in one mature long-form study was around 4.62x.

Should Shorts and long-form use the same benchmarks?

No. Their public reach and engagement distributions differed substantially in OverseerOS research. Benchmark each format separately.

What is the best YouTube benchmark for my channel?

Your own same-age median across comparable recent videos is usually the best starting benchmark. Use external data afterward to understand where that baseline sits relative to similar channels.

Are YouTube benchmarks algorithm thresholds?

No. These are descriptive observations from public datasets. They do not prove that crossing a specific engagement rate, view count, or ratio causes YouTube to recommend a video more widely.