TikTok Carousel vs Video in 2026: Reach and the Right Mix

You moved to TikTok photo carousels because they take ten minutes to make. A few product shots, some text on each slide, a trending sound, done. Then you noticed that your carousels sit at a fraction of the views your videos get, and the posts that used to work stopped working.

You are not imagining it. The data on carousels flipped between 2025 and 2026. This post explains what changed, why carousels lose on reach, and how to split your posting between the two formats.

What the studies said in 2025

In 2025, carousels looked like the smart bet. Fanpage Karma analyzed close to 700,000 TikTok and Instagram posts from January to May 2025. On TikTok, carousel reach was only about 3% higher than video reach, so roughly level. But the carousel engagement rate was more than 81% higher, and carousels got almost 82% more likes. Shares were about a third lower than for videos.

That engagement number is what spread. "Carousels outperform video" became the advice in a lot of founder threads, and many small brands shifted most of their output to photo posts.

What the studies say in 2026

Metricool's 2026 TikTok Study looked at 2,314,756 posts from more than 92,000 accounts, comparing January to February 2025 with the same months of 2026. Their findings point the other way:

• Videos got 5.6 times more views and 7.8 times more interactions than image or carousel posts.
• Image and carousel posting grew about 140% year over year, but views per post on those formats dropped 23% and interactions fell 62%.
• Across the platform, average views fell about 31% and reach about 29%, while the number of posts published rose sharply.

So everyone moved to carousels at once, the supply exploded, and each carousel got a smaller slice of attention.

Why carousels reach fewer people

The short answer is watch time. A video gives TikTok a clean signal: how many seconds people watched, whether they finished, whether they watched again. Those signals decide whether a post moves from a small test audience to a bigger one.

A carousel does not carry that signal in the same way. A swipe through five slides is a much thinner reading than "watched 14 of 15 seconds". With less to go on, and far more photo posts competing for the same feeds in 2026, carousels get tested on smaller audiences and stall earlier.

That also explains why the 2025 and 2026 numbers disagree. Engagement rate is interactions divided by views. A carousel shown to fewer people, mostly followers who already like you, can post a high engagement rate on a small base. A video shown to strangers gets more views and a lower rate. If your goal is reaching people who have never heard of your app, views and reach are the numbers that matter, and that is where video wins.

Where carousels still earn their spot

Carousels are not useless. They are good at a different job: being saved. A post someone wants to come back to (a setup guide, a checklist, a before and after, "5 settings to change on day one") works well as slides because people can read at their own pace.

Use carousels for:

• Step by step how-tos for your app ("How to set up your first budget in 3 screens")
• Comparisons and lists your audience would screenshot anyway
• Feature recaps for people who already follow you
• Product detail shots for a physical product, where each slide answers one buyer question

Use video for anything meant to reach new people: the hook-led demo, the problem and fix, the "I built this because" story.

A simple mix rule

For most app makers and small product brands, make most of your posts videos and keep carousels to a small share. A workable starting point:

• 7 posts a week: 6 videos, 1 carousel
• 14 posts a week: 11 or 12 videos, 2 or 3 carousels

That puts carousels at roughly 15 to 20% of your output. Pick the carousel topics on purpose: things worth saving, not just "the video idea, but faster to make".

If you post a lot, keep in mind that TikTok caps how many posts one account can publish per day, around 9 to 10. Spreading your mix across the week works better than stacking it into one day.

Judge each format over a week, not one post

One carousel at 300 views tells you almost nothing. A new account's first posts can sit at very low views whatever the format, because TikTok is still working out who to show the account to. That is about the account, not one hook or one format.

Instead, compare formats over a full week (two if you post less than daily). For each post, copy the numbers from the post's analytics in TikTok Studio into a sheet:

date,format,views,saves,shares
2026-10-13,video,1840,12,9
2026-10-13,carousel,410,15,2
2026-10-14,video,960,4,3
2026-10-15,video,2210,18,14
2026-10-16,video,730,3,1
2026-10-17,carousel,520,21,4
2026-10-18,video,1350,9,6

Then compare median views per format (the median keeps one lucky post from skewing the result) and saves per 1,000 views:

import csv
from collections import defaultdict
from statistics import median

by_format = defaultdict(list)
with open('tiktok_week.csv', newline='') as f:
    for row in csv.DictReader(f):
        by_format[row['format']].append(row)

for fmt, posts in by_format.items():
    views = [int(p['views']) for p in posts]
    saves = sum(int(p['saves']) for p in posts)
    per_1k = 1000 * saves / max(sum(views), 1)
    print(f'{fmt}: {len(posts)} posts, median views {median(views):.0f}, '
          f'saves per 1k views {per_1k:.1f}')

With the sample data, videos come out at a median of 1,350 views and carousels at 465, while carousels earn far more saves per 1,000 views. That is the typical pattern: video for reach, carousel for saves. Read it like this:

• Carousel median views far below video, saves per 1k views higher: carousels are doing their job. Keep them at a small share.
• Carousel views low and saves low too: the topics are not worth saving. Change what you put in carousels before dropping the format.
• Carousel views close to or above video: your niche may be one where photo posts still travel (some beauty, fitness and education accounts see this). Raise the share a little and measure again.

Change one thing per week. If you change the mix, the hooks and the posting times at once, you will not know which one moved the numbers.

Make the video side cheap too

The real reason founders drift to carousels is that video feels expensive to make. The fix is a video format that is nearly as fast: a photo video. Take the same product shots you would use for a carousel, give each scene motion and on-screen text, then add music and a voiceover. You keep the low effort of a carousel and get a post TikTok can measure by watch time.

Social Studio is built around this. It plans a calendar of post ideas for your product, where each idea has its own cadence (daily, twice a week and so on), and makes each post as a photo video, a carousel or a single image from your real product photos, with original music, a voiceover from a catalog of 200+ voices, and a caption and hashtags. So you can set most ideas to photo videos and keep one save-worthy idea as a weekly carousel. Every post waits for your review and only goes to your own TikTok account when you press Post, with an AI-generated-content disclosure attached. You can see real examples on the samples page.

The takeaway

Carousels were a shortcut that worked in 2025 and mostly stopped working once everyone took it. Post mostly videos to reach new people, keep carousels to a small share for posts worth saving, and judge both on a week of medians rather than one post.