Back to Blog
The Half-Life of a YouTuber
Analysis8 min read

The Half-Life of a YouTuber

We analyzed ~100K YouTube channels. Median lifespan: 8.8 years. Annual mortality has nearly tripled since 2022. Six charts reveal what's actually driving it.

We analyzed 105,891 independent YouTube channels. One question: how long do they actually last?

The answer is more precise, and more alarming, than most creators or brands expect.

How long do YouTube channels actually last?

The age-at-death distribution has a clear peak. Of all the channels in our dataset that have gone inactive, the single most common age at last upload is 8 to 9 years. The median sits at 8.8 years, with the middle 50% of channels falling between roughly 6 and 12 years.

Age at going dark: % of 40,379 dormant channels

Peak at 8–9 years (10.1% of dormant channels). Median: 8.8 years. Source: CreatorMap dataset, May 2026.

This is not a flat distribution. YouTube channels don't stop posting randomly across their lifespan. There's a distinct mortality cliff that builds from year 6 and peaks sharply at years 8 to 9. Channels that reach 12 or more years demonstrate exceptional resilience, and relatively few make it there. Only 3.8% of dormant channels went dark within their first 2 years; the majority post for years before eventually going quiet.

Monthly mortality rate: no plateau in sight

The percentage of alive channels going dark every month has risen without interruption from fractions of a percent in 2015 to over 2% per month by mid-2025.

Monthly mortality rate: % of alive channels going dark per month

Continuous acceleration 2015–2025, no plateau. Sep 2025: 2.36%/month. Key events marked. Source: CreatorMap dataset, May 2026.

The trend shows continuous acceleration with no plateau through late 2025. Two events mark visible inflection points on the chart: the mainstreaming of YouTube Shorts in mid-2021, and the beginning of the AI content wave in early 2023. Neither caused the acceleration on its own, but both appear to have steepened it.

Do newer cohorts burn out faster?

Tracking each creation year as its own cohort tells a nuanced story. The classes of 2012 through 2017 all show more than half their channels still posting as of mid-2026. YouTube creators are stickier than the discourse usually suggests.

Survival curves by creation cohort: % still posting at each age

2018 cohort (★ orange) reached half-life at age 7, the earliest of any tracked cohort. Source: CreatorMap dataset, May 2026.

The exception is the 2018 cohort, which reached its half-life at age 7, a full 1.8 years earlier than the overall median. Channels created in 2018 are dying at above-average rates relative to their age. The 2019, 2020, and 2021 cohorts show even steeper early attrition, though they are still young enough that drawing firm conclusions is premature. The trajectory is worth watching.

Your content genre is worth 4 years of lifespan

Genre turns out to be one of the strongest predictors of channel longevity in our data. Across 15 content categories, the spread between shortest and longest median lifespan is 4 full years.

Genre lifespan: years above or below the 8.8y overall median

Deviation from overall median (8.8y). Spread: People & Blogs (7.45y) to Nonprofits & Activism (11.49y). Source: CreatorMap dataset, May 2026.

Personal vlogging and general entertainment (the highest-volume, lowest-barrier categories) burn out fastest. Channels built around specific passions, skills, or communities outlast the average by years. Music channels last 10.4 years. Travel and activism channels push past 10.5 and 11.5 years respectively.

The data suggests specificity is a survival strategy. For brands: the genre of a creator's content is a meaningful signal for partnership longevity, not just current reach.

The 2023–2025 spike: aging demographics or platform shock?

The most common pushback on rising "going dark" numbers is demographic: a large cohort of channels was created in 2015–2017, so a pile-up of deaths around 2023–2026 is exactly what you'd expect from natural aging. This is partially true, and partially wrong.

% of creation cohort going dark each calendar year

2015201620172018201920202021202220232024202520102010 cohort → went dark 2015: 0.1%2010 cohort → went dark 2016: 0.3%2010 cohort → went dark 2017: 0.3%2010 cohort → went dark 2018: 1.6%1.62010 cohort → went dark 2019: 2.6%2.62010 cohort → went dark 2020: 3.3%3.32010 cohort → went dark 2021: 3.9%3.92010 cohort → went dark 2022: 4.6%4.62010 cohort → went dark 2023: 5.4%5.42010 cohort → went dark 2024: 6.7%6.72010 cohort → went dark 2025: 10.1%10.120112011 cohort → went dark 2015: 0.1%2011 cohort → went dark 2016: 0.1%2011 cohort → went dark 2017: 0.6%0.62011 cohort → went dark 2018: 1.8%1.82011 cohort → went dark 2019: 3.2%3.22011 cohort → went dark 2020: 3.9%3.92011 cohort → went dark 2021: 4.3%4.32011 cohort → went dark 2022: 4.5%4.52011 cohort → went dark 2023: 5.7%5.72011 cohort → went dark 2024: 7.2%7.22011 cohort → went dark 2025: 10.6%10.620122012 cohort → went dark 2015: 0.1%2012 cohort → went dark 2016: 0.3%2012 cohort → went dark 2017: 0.7%0.72012 cohort → went dark 2018: 2.0%2.02012 cohort → went dark 2019: 3.5%3.52012 cohort → went dark 2020: 3.6%3.62012 cohort → went dark 2021: 4.1%4.12012 cohort → went dark 2022: 4.3%4.32012 cohort → went dark 2023: 5.9%5.92012 cohort → went dark 2024: 7.6%7.62012 cohort → went dark 2025: 10.9%10.920132013 cohort → went dark 2015: 0.1%2013 cohort → went dark 2016: 0.3%2013 cohort → went dark 2017: 0.7%0.72013 cohort → went dark 2018: 1.9%1.92013 cohort → went dark 2019: 3.6%3.62013 cohort → went dark 2020: 4.5%4.52013 cohort → went dark 2021: 4.6%4.62013 cohort → went dark 2022: 4.7%4.72013 cohort → went dark 2023: 6.2%6.22013 cohort → went dark 2024: 7.9%7.92013 cohort → went dark 2025: 10.9%10.920142014 cohort → went dark 2015: 0.1%2014 cohort → went dark 2016: 0.3%2014 cohort → went dark 2017: 0.8%0.82014 cohort → went dark 2018: 2.1%2.12014 cohort → went dark 2019: 4.2%4.22014 cohort → went dark 2020: 4.8%4.82014 cohort → went dark 2021: 5.3%5.32014 cohort → went dark 2022: 4.8%4.82014 cohort → went dark 2023: 6.7%6.72014 cohort → went dark 2024: 7.9%7.92014 cohort → went dark 2025: 10.7%10.720152015 cohort → went dark 2015: 0.0%2015 cohort → went dark 2016: 0.4%2015 cohort → went dark 2017: 0.8%0.82015 cohort → went dark 2018: 2.1%2.12015 cohort → went dark 2019: 4.2%4.22015 cohort → went dark 2020: 5.5%5.52015 cohort → went dark 2021: 5.8%5.82015 cohort → went dark 2022: 5.7%5.72015 cohort → went dark 2023: 6.7%6.72015 cohort → went dark 2024: 8.0%8.02015 cohort → went dark 2025: 10.6%10.620162016 cohort → went dark 2015: 0.0%2016 cohort → went dark 2016: 0.3%2016 cohort → went dark 2017: 0.8%0.82016 cohort → went dark 2018: 2.1%2.12016 cohort → went dark 2019: 5.1%5.12016 cohort → went dark 2020: 5.9%5.92016 cohort → went dark 2021: 6.3%6.32016 cohort → went dark 2022: 5.6%5.62016 cohort → went dark 2023: 6.7%6.72016 cohort → went dark 2024: 8.3%8.32016 cohort → went dark 2025: 10.9%10.920172017 cohort → went dark 2015: 0.0%2017 cohort → went dark 2016: 0.0%2017 cohort → went dark 2017: 0.6%0.62017 cohort → went dark 2018: 2.7%2.72017 cohort → went dark 2019: 5.8%5.82017 cohort → went dark 2020: 6.3%6.32017 cohort → went dark 2021: 5.8%5.82017 cohort → went dark 2022: 5.7%5.72017 cohort → went dark 2023: 6.5%6.52017 cohort → went dark 2024: 8.4%8.42017 cohort → went dark 2025: 10.2%10.220182018 cohort → went dark 2015: 0.0%2018 cohort → went dark 2016: 0.0%2018 cohort → went dark 2017: 0.0%2018 cohort → went dark 2018: 2.0%2.02018 cohort → went dark 2019: 7.7%7.72018 cohort → went dark 2020: 9.2%9.22018 cohort → went dark 2021: 7.4%7.42018 cohort → went dark 2022: 6.1%6.12018 cohort → went dark 2023: 6.6%6.62018 cohort → went dark 2024: 7.1%7.12018 cohort → went dark 2025: 9.7%9.720192019 cohort → went dark 2015: 0.0%2019 cohort → went dark 2016: 0.0%2019 cohort → went dark 2017: 0.0%2019 cohort → went dark 2018: 0.0%2019 cohort → went dark 2019: 8.0%8.02019 cohort → went dark 2020: 8.9%8.92019 cohort → went dark 2021: 7.2%7.22019 cohort → went dark 2022: 5.5%5.52019 cohort → went dark 2023: 7.2%7.22019 cohort → went dark 2024: 7.6%7.62019 cohort → went dark 2025: 10.0%10.020202020 cohort → went dark 2015: 0.0%2020 cohort → went dark 2016: 0.0%2020 cohort → went dark 2017: 0.0%2020 cohort → went dark 2018: 0.0%2020 cohort → went dark 2019: 0.0%2020 cohort → went dark 2020: 3.5%3.52020 cohort → went dark 2021: 2.3%2.32020 cohort → went dark 2022: 2.3%2.32020 cohort → went dark 2023: 2.3%2.32020 cohort → went dark 2024: 2.9%2.92020 cohort → went dark 2025: 7.6%7.6Year channel went dark →Creation year →
Diagonal bands = age-driven mortality. Vertical orange bands in 2023–2025 = platform-level pressure across all cohorts. Source: CreatorMap dataset, May 2026.

The heatmap plots creation year against year-of-going-dark, normalized by cohort size. Before 2022, the pattern is mostly diagonal: channels dying at expected ages. From 2023 onwards, vertical bands appear. Channels created in 2010, 2015, and 2019 all show elevated death rates in the same calendar years. Age alone doesn't explain that. When channels of every age die at higher rates simultaneously, that is the fingerprint of a platform-level shift.

Annual mortality rate has nearly tripled since 2022

The clearest way to separate cohort aging from platform-level pressure is the annual mortality rate: deaths per 1,000 channels that were actually alive at the start of each year. This controls for both population growth and the 2015–2017 cohort bulge.

Annual mortality rate: deaths per 1,000 alive channels

Rate tripled from 57.4 (2022) to 161.6 (2025). Orange bars = acceleration phase. Source: CreatorMap dataset, May 2026.
YearDeaths per 1,000 alive
20163.1
201941.7
202257.4
202376.5
2024104.1
2025161.6

The rate has nearly tripled since 2022 alone. Channels created in 2012 and channels created in 2019 are both dying at higher per-capita rates in 2024–2025 than in any prior year. When elevated risk appears across all ages simultaneously, that's platform pressure, not demographics.

What's driving the acceleration?

The data doesn't tell us directly. But the timeline aligns with several converging pressures:

  • Short-form video mainstreaming: YouTube Shorts launched mid-2021 and permanently rewired what gets recommended and what audiences expect from the platform
  • The AI content wave: starting late 2022, the cost of generating competing content dropped sharply, raising the relevance floor every independent channel must clear
  • Sustained algorithm changes: repeated rewiring of the recommendation system has repeatedly disrupted channels that were previously stable

Any one of these could compress the viable window for independent creators. Together, they appear to be doing exactly that.

The 8.8-year lifespan isn't disappearing. But the probability of reaching it is declining at the fastest rate in YouTube's history.

Analysis based on 105,891 independent YouTube channels sourced from the CreatorMap dataset. Channels classified as active or low-engagement as of May 2026 are treated as right-censored observations in all survival calculations.

TagsYouTube channel lifespanYouTube creator longevitycreator burnout statisticsYouTube survival rateinfluencer longevitycreator economy dataYouTube analytics

Ready to Find Your Perfect Creators?

Start your free trial and experience AI-powered creator discovery today.