10,001
Videos with caption + view data
0
Hashtags = highest median views
2.6x
View gap between 0 hashtags and 4-6 (Instagram)
Open any "TikTok growth" or "Instagram strategy" article and you will get the same advice: stack your captions with 20 to 30 hashtags. It has been the default playbook for years.
We counted hashtags in 10,001 analyzed TikTok and Instagram videos and compared median view counts by hashtag bucket. The result: the advice is wrong, and on Instagram it is actively hurting reach.
The headline finding
@michaeldean2.0
Reset your shoulders and upper back 😙 Push slowly with your knees & elbows for 30 seconds. Your spine will thank you.
no hashtags
24,528
Median views
0 hashtags
2,300 IG videos
Pure prose. The video does the discovery work, not the caption.
@itstheyear2026
TATUM SHOULDNT HAVE COME BACK…
#jaysontatum #celtics #nba
17,478
Median views
3 hashtags
699 IG videos
Niche-specific tags, conversational caption. Still beats the standard 4-6 pattern by 1.9x.
@imma_eat_this
So I went to this spot @g_and_r_deli_ in the Bronx!
#gandrdeli #thebronx #newforkcity #eeeeeats #foodphotography
9,420
Median views
5 hashtags
1,378 IG videos
Highest save rate in our data (0.616%). Reference content (food, recipes, location reviews) is the niche where 3-5 niche-specific tags pull weight.
@golftipsdaily
Follow for more golf tips 🏌️
#golf #golfcourse #golfcoach #golfpro #golfclub #golftiktok #tiktokgolf #golfswing #golftips #golfer #golftime #sportsclub #golfgram #golfdaily #golfworld #golfaddict
4,836
Median views
16 hashtags
46 IG videos
5x reach penalty vs zero hashtags. The 'discoverability boost' from stacking tags is folklore.
Zero hashtags wins at the median. 17,400 views across 4,499 videos. Every additional bucket trends down: 1 to 3 hashtags drops to 12,400. 4 to 6 hashtags drops to 9,002. By the time you are stacking 21 or more, the median sits at 4,068, a 76 percent collapse from the no-hashtag baseline.
This is not a small effect. The drop from 0 to 4-6 hashtags (the "sweet spot" most guides recommend) is 48 percent on the median. The drop from 0 to 21+ is 4.3x.
The average looks different on purpose. The 7 to 10 hashtag bucket has the highest average views in our data (569,556) because it contains a few mega-viral outliers. The median for that same bucket is 8,717. When a stat looks too good, check whether you are reading an average inflated by a handful of monster videos.
Instagram and TikTok behave differently
Splitting the data by platform shows the real story. Instagram shows a clean monotonic decline. TikTok is closer to flat with a slight zero advantage.
Median views by hashtag count, split by platform
Instagram (4,888 videos): Every step up in hashtag count drops the median. Zero hashtags = 24,528 median. The standard 4 to 6 hashtag pattern most agencies recommend = 9,420 (a 2.6x gap). 16 or more hashtags = 4,836 (a 5x gap). On Instagram, hashtags are an unforced reach penalty.
TikTok (4,448 videos): Less dramatic, but zero still leads (11,600 median) over 1 to 3 hashtags (8,064). The 11 to 15 bucket has a higher median (22,100) but only 30 videos, so treat that as noise rather than a signal. The 16+ bucket has 3 videos total. TikTok is closer to "hashtags don't matter much" than "hashtags help."
The pattern across both platforms: any hashtags is worse than none, and going past 6 hashtags is consistently worse than going under.
What the engagement numbers say
Hashtag count also affects engagement rate, just in a less linear way:
| Hashtags | Videos | Engagement % | Save % | Share % |
|---|---|---|---|---|
| 0 | 4,499 | 14.05 | 0.265 | 0.176 |
| 1-3 | 1,673 | 6.07 | 0.368 | 0.198 |
| 4-6 | 3,082 | 7.50 | 0.616 | 0.252 |
| 7-10 | 567 | 5.71 | 0.377 | 0.150 |
| 11+ | 180 | 4.38 | 0.171 | 0.066 |
Two things stand out.
Zero hashtags has 2x the overall engagement rate (14.05 percent) of any other bucket. Likes drive most of that gap, and likes correlate with raw reach (more views = more like opportunities). Consistent with the median view story.
The 4 to 6 hashtag bucket has the highest save rate (0.616 percent), more than double the zero bucket. This is the one place hashtags pull weight, and it tracks with content type. Recipe posts, "spots in [city]" food content, and how-to threads frequently use 4 to 6 niche-specific tags AND get saved for later reference. Location tags on a Bronx deli post are useful. Generic "viral" or "fyp" tags on a hot take are not.
If your content type is reference (food, fitness, finance how-tos), 3 to 5 niche-specific hashtags can earn you saves. For everything else, fewer is better.
The "but X creator hits millions with hashtags" objection
Yes, plenty of viral videos use 8 to 10 hashtags. A few real ones from our data:
How to clean your fruits and vegetables...
@doctortim.md · 69.8M views
So I went to this spot @g_and_r_deli_ in the Bronx!
@imma_eat_this · 17.1M views
Follow @almostaveragegolf if you want more awesome golf videos like this one
@almostaveragegolf · 44.5M views
These are real outliers. They exist. The bucket they live in still has a 8,717 median. So when you ask "should I copy doctortim's 10-hashtag pattern," the honest answer is "for every doctortim there are dozens of accounts with the same hashtag stack getting 8K views."
For comparison, the highest-performing zero-hashtag posts in our data:
Reset your shoulders and upper back 😙
@michaeldean2.0 · 36.3M views
The internet is OUTRAGED after this Phillies Karen stole a ball from a kid
@ev_handd · 7.9M views
recently my dad has been apologizing to me a lot. he feels guilty...
@jun_yuh · 8.7M views
The zero-hashtag mega-virals tend to look like one of two things: tight emotional storytelling (jun_yuh, ev_handd) or no-context skill content (michaeldean2.0). They depend on the content doing the discovery work, not on the algorithm's keyword index.
Stop padding captions with hashtags. See what your account actually needs.
TCL audits your hashtag patterns plus your top competitors and writes a 30-day script calendar with caption strategies tuned to what's actually moving views in your niche, not generic SEO advice.
47,598 creators·No credit card required·60 seconds
What to actually do
A quick visual reference for what each hashtag count actually looks like, color-coded to median performance:
If you're on Instagram, default to zero or 1-3 hashtags. The 24,528 median for zero-hashtag IG posts vs 9,420 for the standard 4-6 pattern is the cleanest finding in this study. Whatever you're doing now with 10+ hashtags is leaving views on the table.
If you're on TikTok, you have more latitude, but zero or 1-3 still leads. Stop using 'fyp', 'viral', 'foryou', and 'trending' as filler. They were never an algorithmic signal worth what creators believe they are.
If your content is reference-style (recipes, location reviews, how-tos), 3 to 5 niche-specific hashtags can help saves. The 0.616 percent save rate in the 4-6 bucket is real. But pick tags that describe the actual content (e.g. #norwalkct, #ricecooker, #dividendinvesting), not generic discovery hashtags.
Stop reading 'hashtag strategy' guides written before 2024. The algorithms on both platforms shifted toward content-based discovery. Hashtags are either a small save signal (in specific niches) or a small reach penalty (everywhere else). They are not the engine they used to be.
If you're auditing your captions, count hashtags first. If your average is over 6, cut to under 4. If your average is over 10, cut to zero. The median view lift you get from removing tags is larger than almost any caption optimization you can make.
The bottom line
Across 10,001 videos with caption and view data, more hashtags consistently correlated with lower median views. The advice everyone repeats (stack your caption with 20 to 30 hashtags for discoverability) does not match what's actually winning in our data.
The honest answer to "how many hashtags should I use" is closer to zero than thirty. The honest answer to "do hashtags help reach in 2026" is no on Instagram and barely on TikTok. The exception is reference-style content where a few niche-specific tags help saves.
For everyone else, the cheapest single edit you can make to your caption strategy is deleting hashtags.
See your real caption + hashtag patterns. Get a strategy built from your data.
Connect TikTok and Instagram. TCL audits the hashtag, caption, and hook patterns across your videos plus your top competitors, then writes a 30-day script calendar with the caption strategy actually winning in your niche.
47,598 creators·No credit card required·60 seconds
Methodology
Dataset: 10,001 distinct TikTok and Instagram videos with both caption and view data populated. Pulled from scrapes and content_audits tables on 2026-04-24.
Hashtag counting: Counted via Postgres regexp_matches(caption, '#[A-Za-z][A-Za-z0-9_]*', 'g') to match every # followed by a letter (avoids false positives like "#1" rankings). Captions were treated as the source of truth because the dedicated hashtags column had inconsistent storage formats (JSON arrays, JSON objects, comma-strings, and a few malformed entries) across the dataset.
Sample distribution by bucket: 0 hashtags = 4,499 videos. 1-3 = 1,673. 4-6 = 3,082. 7-10 = 567. 11-15 = 131. 16-20 = 32. 21+ = 17.
Why median, not average: A small number of mega-viral outliers in the 7-10 hashtag bucket pull that bucket's average view count to 569,556, even though the median sits at 8,717. Averages mislead in a power-law distribution like short-form video. Median is the honest summary for most creators.
Known limits:
- The 16-20 (n=32) and 21+ (n=17) buckets in the overall view are small. Treat the magnitudes as directional and the direction as confident.
- The TikTok 11-15 bucket (n=30) has a high median (22,100) likely driven by a few outliers; the broader TikTok pattern still favors zero or low-hashtag captions.
- The 4-6 hashtag save rate spike is partly driven by food and how-to content composition. The pattern is real but niche-dependent.
- Captions were inferred from the source pipeline. Cross-posted videos count once per platform.