Do TikTok slideshows perform better? Official statements, vendor data and opinion

What is the short answer?

No source in our set shows that slideshows perform better. The evidence comes in three layers, and each says something different:

1. Official statements describe which signals the platforms use. They name no ranking of formats.
2. Vendor datasets report real numbers from large post samples. Their metrics, periods and samples differ, and on TikTok they point in opposite directions.
3. Opinion covers hooks, slide counts, timing and most advice you will read. It may be right. It is untested on your account.

Mixing the layers is how weak claims get repeated. The practical answer is to treat every tip as a hypothesis and test one variable at a time, as set out near the end of this page.

What do TikTok and Instagram say officially?

TikTok

• A June 2020 Newsroom post (opens in a new tab) says recommendations draw on user interactions (likes, shares, follows, comments, content you create), video information (captions, sounds, hashtags) and device and account settings, the last with lower weight. A strong signal, such as finishing a longer video, counts for more than a weak one.
• The same post says neither follower count nor previous high-performing videos are direct factors. Larger accounts still tend to collect more views, so do not read this as "account size is irrelevant".
• TikTok's Help Center (opens in a new tab) gives the current wording: interactions (likes, shares, comments, watching to the end or skipping, accounts followed), content information (sounds, hashtags, view counts, country) and user information (device settings, language, location, time zone, device type). For most users, interactions, including time spent watching, are generally weighted more than other factors. We read this page in German, translated.

Instagram

• In "Instagram Ranking Explained" (opens in a new tab) (31 May 2023), Instagram wrote that it has no single algorithm; Feed, Stories, Explore, Reels and Search each rank content their own way.
• For Feed, the post lists the main signals roughly in this order: your activity (liked, shared, saved, commented), information about the post (including how quickly people react to it), information about the poster, and your history with that person. For Explore, the key predicted actions are likes, saves and shares.
• The post gives no ranking of carousels against photos or Reels. It says only that the feed shows a mix.

What the official sources leave out

The TikTok Help Center says nothing about weighting a photo post differently from a video. Instagram's ranking post names no preferred format. The 2020 Newsroom post does not either, and Instagram's ranking post dates from 2023. Both platforms may have changed their systems since.

What do vendor datasets report, and how were they built?

None of these is peer reviewed. All are produced by companies that sell social-media tools, and each uses its own definition of "engagement". Always quote the metric.

Where do the datasets conflict?

• TikTok photo versus video has no consensus. Fanpage Karma (opens in a new tab) found carousels ahead on engagement rate. Buffer (opens in a new tab) found video ahead of images (3.39% versus 1.92%). Periods, samples and definitions differ, so a claim that photo posts beat video on TikTok is not supported.
• Instagram looks more consistent, with a catch. Buffer, Socialinsider and Fanpage Karma all place carousels at or above Reels on engagement rate, but the reach figures conflict: Buffer's sub-study (opens in a new tab) finds Reels with 1.36 times the reach of carousels, while Fanpage Karma (opens in a new tab) reports Instagram carousels with 13% higher reach than Reels. Buffer notes that Reels are built for views. One metric divides by reach, one by followers, one is undefined.
• Headlines and detail can disagree. Buffer's highlights say carousels earn 12% more engagement than Reels, while its detailed section implies about 109% (Buffer report (opens in a new tab)). Read the method before you quote a number.

What is opinion rather than evidence?

These are common claims. Each might help you; none has a source on this site that supports it as a rule.

• "Saves matter most." Instagram does list saves among the key Explore actions. For TikTok, saves are not named in the sources above.
• "A strong first slide drives everything." It follows from the stated importance of watching or skipping, and we find it plausible. No source measures the share of impact that comes from slide one.
• "Instagram shows a carousel again from slide two." One blog relays this as a comment by Instagram's head (Planoly (opens in a new tab)), but no primary source was found. Do not plan around it.
• Best times, ideal slide counts, hashtag counts. No official source in our set gives a number for a photo post. Treat any number you see as a guess to test.

For specific claims checked one by one, see slideshow myths.

How do you test what works on your own account?

1. Write one question. Example: "Does a question as slide one earn more saves than a statement?"
2. Change only that. Keep topic, design, posting window and slide count as close as you can.
3. Post several of each variant. In our view five per variant is a floor for noticing a pattern, not proof of one.
4. Log the same fields every time, including a context note.
5. Compare medians, not your best post. Do not compare your numbers with a vendor's median, because the denominators differ.
6. Decide beforehand what result would change what you do.
7. Repeat after a month. The platforms change.

Here is a sample log. These are fictional numbers, there only to show the layout.

Fyslide lets you link a published post and type in its numbers by hand, which suits a log like this (analytics guide). Automatic analytics are marked "Coming soon".

What if the test shows no difference?

That is a result. It means the variable you changed did not matter much at your account's size, so you can stop spending effort on it. Write it in the log and pick the next variable. Candidates for the next test include the first-slide wording, the cover image, the number of slides and the caption opening. A flat result is more useful than a noisy win, because it removes one thing from your list.

Do not rerun a test until it gives the answer you wanted. Decide in advance how many posts you will compare, then stop.

What would count as strong evidence?

• A platform statement that ranks formats, with a date.
• A dataset that names its metric, its sample, its period and whether it reports means or medians.
• Two independent datasets that agree on direction. For TikTok photo posts, we do not have that yet.

If you are weighing formats rather than tactics, slideshow versus carousel versus photo mode explains the naming, and how many photos covers the slide-count question. For Instagram specifics, start with the Instagram carousel guide. If posts are already underperforming, work through the diagnostic for slideshows not getting views.