TikTok Slideshow Hooks: Cover Slide Ideas From Our Data

The cover slide decides whether anyone sees the rest of your slideshow. This is the cover-slide chapter of our complete guide to TikTok slideshows, built on what we found in the slideshows we analysed.

What makes a good TikTok slideshow hook?

A TikTok slideshow hook is the cover slide: one image and, usually, a short line of text. It works when it makes a specific promise that only swiping will settle. In the slideshows we analysed, the median cover carried about 9 words, and curiosity gaps and bold claims were the most common hook types.

If you want hooks for talking-to-camera videos, that's a different job, covered in the best TikTok hooks. A slideshow has nothing to say out loud and nothing moving, and the viewer reads the cover at their own pace. That changes what works.

The data behind this post

FYPNow analyses TikTok slideshows slide by slide. For each one, an AI model reads the cover and labels its hook type, transcribes the on-image text, assigns every slide a role (hook, setup, value, proof, payoff, CTA) and rates the cover, the swipe-through and the save-worthiness from 0 to 100.

The figures here come from FYPNow's slideshow analyses as of October 2026, and they come with three caveats you should keep in mind for every number below:

• It's not a random sample of TikTok. The slideshows come from accounts that FYPNow customers chose to track or inspect, and a few creators account for a large share of them. Where the most prolific creator skews a figure, we show the figure without them.
• The niche mix is narrow. The sample leans toward business, finance, motivation and education. Food, beauty or travel covers might follow different habits.
• Labels and scores are our model's ratings. "Bold claim" is the model's reading of a cover, and a cover score of 75 is the model's opinion. Neither is a measured outcome like views or swipe-through.

We don't quote views by hook type anywhere in this post. Those medians rest on a handful of creators per type, and they would tell you more about those accounts than about the hooks.

The cover hook types we see most

Here's how the model labelled the covers. Because the largest creator leans heavily on one or two types, the second column is the fairer picture.

Shock covers were too rare to say anything about.

Look at what the largest creator does to the first row: they contribute nearly half of the curiosity-gap covers. Without them, curiosity gaps and bold claims are tied, and aesthetic covers, the ones with little or no promise in the text, move up to third.

One more thing the table can't tell you: which type is best. The model gave a median cover score of 75 to every common hook type. Either it rates most covers alike or the hook type matters less than the execution. Either way, the data can't rank them, so treat the types below as a menu rather than a leaderboard.

Curiosity gap

The cover opens a loop and refuses to close it. Examples in the style of covers in our sample: "Growing up is noticing..", "The boring side of running a bakery :". Note the trailing dots and colons, which were common. The sentence literally isn't finished, and slide 2 finishes it.

Formulas: "The [unglamorous] side of [topic]:" · "[Thing everyone does] is realising.." · "What [insider] doesn't tell you about [topic]"

Use it when slide 2 really completes the thought. A loop that closes on something obvious costs you the next swipe.

Bold claim

A flat statement the viewer either agrees with hard or wants to argue with. Examples in the style of covers in our sample: "Being shy isn't your weakness.", "No gym, No excuse.", and "Your routine decides your results Not your talent."

Formulas: "[Common belief] is wrong." · "You don't need [X]. You need [Y]." · "[Outcome] comes from [unexpected cause], not [obvious cause]."

The slides after a bold claim have to back it: reasons, proof, an example. Otherwise it's just a quote card, which is a different (and much rarer) format.

Aesthetic

The picture does the work and the text, if any, is a label or a mood: a row of arrows, a one-word label like "recap", or a screenshot with no caption (examples in the style of covers in our sample). These were spread across more creators than any other type, which fits: lifestyle, fitness and photo-dump accounts post them.

Formulas: a striking first photo plus "[place / month / moment]" · arrows ("→" or ">>>>") that just say there's more

This only works if the image genuinely stops someone. If your first photo is ordinary, add words.

Story opener

The first line of a story, cut off. Examples in the style of covers in our sample: "Behind every finished marathon...", "Our weekends now..", "Attempt one."

Formulas: "How I [result] after [setback]" · "[Time] ago I [low point]." · "Attempt one." (and slide 2 is attempt two)

Story-shaped slideshows ran longer than most in our sample: the median one had 5 slides. Give the story room, and put the outcome last.

List promise

The cover promises a number of things, or a set of signs. Examples in the style of covers in our sample: "SIGNS YOUR HOUSEPLANT IS THIRSTY", "4 common mistakes first-time renters make >>>>", "6 signs you're a morning person".

Formulas: "[N] [mistakes / signs / rules] [audience] [makes / shows / breaks]" · "Signs you're [state]" · "Our go-to [places / tools] for [goal]"

Half the list-promise covers in the raw sample come from the one large creator, so the type is less common elsewhere than the raw count suggests. The number on the cover also fixes your slide count: promise 5, show 5. Our post on how many slides a TikTok slideshow should have works through this.

Relatable

A line the viewer has heard or said. Often it's a quote with a reply. Examples in the style of covers in our sample: "It's too hard" / That's why it works., or Dad : "did you eat today?"

Formulas: "'[Thing people say]' / [your reply]" · "Only [specific group] will get this..." · "[Person]: '[question]'"

These lean on recognition rather than a promise, so the slides after need to escalate the joke or the feeling, not explain it.

Question

A direct question on the cover. Examples in the style of covers in our sample: "Why can't you focus?", "COFFEE AND ANXIETY ARE THEY LINKED?"

Formulas: "Why does [problem] happen?" · "Are [X] and [Y] related?" · "Can [product / habit] really [claim]?"

Questions were the rarest common type (4%). A question cover is a promise to answer it, so the answer is your payoff slide.

How long should cover text be?

Shorter than you'd think. Across the covers with text, the median carried 49 characters, about 9 words. Three quarters stayed at 18 words or fewer, and 9 in 10 at 22 or fewer. 6% of covers had no text at all.

Nine words is roughly "4 common mistakes first-time renters make" plus an arrow. It's one line that can be read at a glance on a phone held at arm's length.

Practical rules that follow:

• Write the cover line, then cut it to about 9 words. If it won't fit, the extra words probably belong on slide 2.
• Put the promise before the detail. "Signs your houseplant is thirsty" works; "If you've been feeling tired lately it might be because..." buries it.
• Leave the image room. A wall of text over a photo turns the cover into a paragraph.

The text inside the slideshow follows its own rules (more text is normal on value slides), which our guide to text on TikTok slideshows covers.

What the words on covers have in common

We ran simple pattern checks over the English covers with text. Categories overlap, so the shares don't add up to 100%.

Three readings, with the caveats attached:

Numbers are everywhere, but list counts aren't. 44% of covers contained a digit, yet only 10% used a proper "5 tips" style promise. The rest were years, prices, percentages and dates (one story opener was nothing but three consecutive years). A number makes a cover concrete even when it isn't a list. "I" outnumbered "you" two to one. 39% of covers used first person, 18% used second person. That's partly the niche mix: motivation and trading creators write in the first person (in the style of "I missed the deadline"). But it's a useful counterweight to the advice that every hook should say "you". "How to", questions and POV were rare. Between them they're under 15% of covers. We can't tell you whether that's because they work less well on slideshows or because this group of creators doesn't use them. We can tell you that you won't look like everyone else if you use one.

Why people swipe past the cover

For every slideshow, the model also writes down what it thinks will make people swipe. Grouped by theme, the two most common reasons were the cover's promise (41% of posts) and curiosity or an open loop (30%). These are the model's hypotheses, not observed behaviour, but they point at the same thing: the cover's job is to create a debt that the next slides repay.

The structure data agrees. The model labelled slide 1 as the hook in 98% of slideshows (though the analysis expects a hook there), and in slideshows with 4 or more slides it placed the payoff on the last or next-to-last slide in 83% of them. So the cover makes the promise, the middle builds, and the end pays it off. A cover that gives the answer away leaves nothing to swipe for.

Cover mistakes to avoid

Some of these are judgement rather than data, so treat them as ours:

• A promise the slides don't keep. "7 signs" followed by four signs. People notice, and they don't save it.
• The answer on the cover. "Drink water to fix your skin" has nowhere to go. "The boring fix for dull skin" does.
• Unreadable text. Thin fonts on a busy photo. If you can't read it on a phone at a glance, nobody else will either.
• A video-style opener. "Hey guys, so today..." wastes the only line you have. Start with the promise.
• Copying a cover word for word. Borrow the structure ("[N] mistakes [audience] make"), never the wording.

Write cover lines faster

If you're stuck on the line itself, our TikTok hook generator takes a topic and gives you a batch of opening lines to choose from. It's written with video hooks in mind, so treat its output as raw material for a cover: pick the line with the clearest promise, cut it to around 9 words, and check it still makes sense as text on an image with no voice behind it. Picking from a long list of candidates is quicker than staring at a blank cover.

Study covers that already worked

Formulas get you started. The fastest way to find what works in your niche is to look at slideshows that did well there.

On the Niches wall in FYPNow, the Slideshows filter shows only photo slideshows, ranked by how far they reached past their creator's follower count. Open one in Video Analysis and you get the cover's hook type and on-image text, our model's cover, swipe-through and save ratings, and the role of every slide after it, so you can see how the cover's promise was paid off. From an analysed slideshow, the script generator can write a slideshow script on the same structure: numbered slide cards, each with its on-image text and role, ready for your own cover line. The slideshow docs cover what the breakdown includes.

For a cover line to go with a full idea, our list of TikTok slideshow ideas pairs every idea with an example cover and a slide-by-slide structure.