This vs That Reels: How to Make Comparison Videos That Settle a Debate

What a this vs that reel is

A this vs that reel compares two options a viewer could choose between, on the same few points, and ends by saying which one wins or who should pick which. The options can be products, apps, foods, habits or two ways of doing the same job. Some add a third, such as three vacuums on the same carpet.

Three nearby formats look similar. A ranking orders many things, covered in the guides to ranking Shorts and tier list reels. A comedy “X vs Y” skit, such as Gen Z vs millennials on a dance floor, contrasts two groups nobody is choosing between. A “this or that” question video asks viewers to pick (“Call or text?”) without comparing anything. Our comparison of AI video editors is the written, long-form version of the same idea.

What 53 breakout comparison reels did in October 2026

What we looked at. On 25 October 2026 we ran two searches in vidIQ’s Instagram and TikTok outlier search. It returns posts from roughly the last 30 days that beat their creator’s usual views by a wide margin. Both searches matched how a post is made, not its subject.

The first described a creator comparing two options side by side or point by point, such as cheap vs expensive, ending with a verdict. The second described a creator on camera explaining the difference between two choices point by point, with labels or a table, then saying which to pick. It left out the 40 creators the first search found and five from our October report.

Each search asked for 20 posts per app with at least 100,000 views and English captions, one per creator. Together they returned 80 posts dated 25 September to 23 October 2026. For each one vidIQ reports the views, the creator’s median views and the multiple between them, plus notes on the opening, layout, text and sound.

What we kept. We kept a post when vidIQ’s notes show it comparing two or more options, versions or methods on at least one stated point. Reading all 80 records by hand, we kept 53: 28 from Instagram and 25 from TikTok. Of the 27 we dropped, nine were comedy “X vs Y” skits where nobody is choosing, such as Gen Z vs millennials dancing. Three were expectation vs reality clips, one was a this-or-that question with no comparison, and 14 were not comparisons at all, such as a card trick filmed from two angles.

Our October report, re-read. Our October 2026 outlier report has no comparison category, so we re-read its 234 posts with the same rule. Four qualified: @hungryoc’s cheap vs expensive taste test, @mariahandbill’s “Toxic versus Normal” sort, @sahilraw22 on a good man vs a bad man and @annie.johnson09’s pimple patch test. None of the report’s 22 split screens compared options a viewer could choose between. Three set an original clip beside an AI-altered copy; most paired the creator with B-roll, a screen recording or a clip they react to.

How to read these numbers. This is a picked sample of winners. vidIQ looks for posts that beat their creator’s usual views, so every row is a success. The data shows what winning comparison reels had in common. It cannot show that comparisons beat other formats, or how many flopped. Views were read on 25 October and may have grown since.

Small accounts, giant multiples. For the 23 posts from creators who usually get under 10,000 views, the median multiple was 563.8x. For the 12 from creators who usually get 100,000 or more, it was about 4x. The largest multiple, 5,508.7x, went to a post on strength training from an account whose posts usually get 135 views. The most-viewed post, a pineapple juice test at 12.3 million views, came from an account with 20 million followers, at 7.3x.

By app, Instagram posts had more median views (1.25 million against 951,500) and TikTok posts a far higher median multiple (301.6x against 48.3x). The TikTok creators usually got about 5,600 views a post, against about 32,400 for the Instagram creators.

Twelve examples and how each was built

Twelve of the 53, six from each app, picked to cover both searches and the four main layouts. “Opening text” is what vidIQ recorded on screen in the first three seconds.

Source: vidIQ outlier search, 25 October 2026. Views and multiples change as posts keep getting views. Descriptions come from vidIQ’s notes and the visible start of each caption; we did not watch every video. vidIQ gave a length for each TikTok here and for none of these Instagram reels.

Patterns: what was compared, layouts, openings, endings, length

These counts describe the 53 kept posts, using vidIQ’s notes on each one. They show what winners had in common; none of them guarantees views.

Products led, and many were selling one side

We sorted each post by what it compared. Products and brands made up 30 of the 53: vacuums, phones, chargers, lipstick, sunscreen, supplements. Choices and habits came next, then “do this, not that” posts that show a wrong way and a right way.

Read the chart with the searches in mind. Our first search named products, and 19 of its 25 kept posts compared them, against 11 of 28 in the second. So the chart shows that each kind appears among winners, not how common each one is.

Product comparisons had the biggest multiples and the smallest accounts. Our guess is that account size explains most of that gap; this data cannot test it.

Of the 30 product posts, 17 were selling one side. Their captions offered a product link, a discount, a shop tag, a partner tag or a price pitch. Otherwise vidIQ’s notes said the post promotes one brand, or the account was the brand or its seller. vidIQ’s caption text stops after about 100 characters and its notes don’t record the apps’ paid partnership labels. So this data cannot show who disclosed. Two caption excerpts carried a partner tag.

Every layout appeared among winners

We sorted each post by how it showed the two sides, counting the main layout when a post mixed them. Showing one side and then the other was the most common, then split screens, including the creator cloned into each half.

Eleven gave each side its own speaker: a couple each holding one vacuum, or the creator cloned so each copy argues for one product. All 11 came from the first search. Ten compared products, and nine of those were selling one side.

Openings named both sides, or asked which

vidIQ recorded text on screen in the first three seconds on 51 of the 53, and the median opening line ran 5 words. Twenty-five put both sides, or all three, in that line, by name or as “this, not that”: “Bissell vs Dyson”, “Original iPhone vs iPhone 18 PM”, “pack like this, not that”. Eleven had a question mark, such as “Which is better for pet owners?”

In 37 of the 53, vidIQ’s notes have a voice starting in the first three seconds. Most of the rest opened on music or on the sound of the test itself: salt on a table, fizzing soda, an RC car’s motor, a crowbar hitting glass. Instagram’s creator FAQ advises making the first 3 seconds of a reel engaging so people don’t move on.

Most named a winner

By our reading of vidIQ’s notes, 44 of the 53 named a winner or told viewers what to do. Five showed a difference and stopped there, such as what goes into Sprite in the UK and the US. Two left the call to viewers, and for two the notes don’t say how they ended.

Of the 53, 39 had a voice, 10 used music only and 4 kept only the sound of the footage. vidIQ’s text notes list static text on 21, guiding text that labels each side or step on 19 and word-by-word captions on 7. That is fewer word-by-word captions than the third of posts in our October report. vidIQ tagged 40 of the 53 as educational.

TikToks ran under a minute

The 25 TikToks had a median of 42 seconds, shorter than the 60-second median across all 113 TikToks with a length in our October report. Sixteen ran 60 seconds or less and 23 ran 90 seconds or less; the longest ran 2 minutes 10 seconds. vidIQ gave lengths for only three Instagram reels, from 12 to 58 seconds.

Pick a real choice your viewers face

Compare two options your viewers are choosing between right now, for a reason they would name themselves.

Several posts in our set named the reason in the first line: “the better vacuum for pet allergies”, “Which is better for pet owners?”, “Preferred Over Google Maps for Driving”, “When To Upgrade Your Samsung?”. The question was narrow enough that one answer could fit it.

1. Someone is choosing this week. A purchase, an app, a habit, a way of doing a job. If nobody is deciding, it is a debate for fun, which makes a different reel.
2. Both options are real contenders. A strong product against one nobody would buy settles nothing.
3. You can name who it is for. “For pet hair”, “for driving”, “for a small kitchen”. The answer often changes with the person.
4. You have used both. The FTC’s guide for influencers puts it plainly: “You can’t talk about your experience with a product you haven’t tried.”
5. You can show the difference. The same spill, the same plate, the same skyline at night. If the difference can’t be filmed, plan a table and source each line.

These tests are our editorial judgment, not a platform rule. Comments are a good place to find choices people are stuck on.

Choose the three points that decide it

List everything that could decide the choice, then keep the three that matter most to the person choosing. Compare both sides on the same three, in the same order.

Three is our rule of thumb for time. At our script timer’s natural pace of 150 words a minute, a 45-second reel holds about 110 words: an opening line, about 25 words per point and a verdict.

• Make each point a test with a result. “Cleans the same spill”, “calories on the same plate”, “a photo of the same skyline at night”.
• Run both sides under the same conditions. Same light, same distance, same settings, same spill.
• Put the point that splits them most last, just before the verdict. That is our habit, not something the data shows.

Sources: YouTube’s pages on three-minute Shorts and Shorts monetization; Instagram’s creator FAQ and Earn money page, all read 25 October 2026. The third point comes from your own numbers, so the verdict changes by creator. The Shorts monetization guide covers the 2027 rule, and the worksheet below has this example with blank rows for yours.

Be fair, and say who pays you

Give each side its best case, and if either side gave you money, a free product or a discount, say so in the video. In our sample, 17 of the 30 product comparisons were selling one side.

• Test each option at the job it was built for. A cordless stick vacuum on a quick spill, not only on a deep carpet.
• Say what the losing side does well, in one line, before the verdict.
• Use the current version of both, and put a date on anything that changes, such as a price.
• Show the test, not only the result, so viewers can judge it for themselves.

Those four are our judgment. The disclosure rules below come from the FTC and the platforms.

What the FTC says

In the United States, the FTC’s Endorsement Guides cover this. Its Disclosures 101 guide says to disclose any “financial, employment, personal, or family relationship with a brand”. It adds that “Financial relationships aren’t limited to money”, naming free or discounted products, and that you should disclose “even if you think your evaluations are unbiased.”

The FTC’s answers to common questions cover the comparison case directly. An influencer with a brand deal asks whether posting negative comments about a competing brand needs a disclosure. The answer: “Yes. If you criticize a competitor of a brand that you are paid to endorse, you should disclose your paid relationship.”

Put it in the video. The same page says a disclosure in a YouTube description is “not enough on its own” and “has the best chance of being clear and conspicuous if it’s included in the video itself.” Disclosures 101 says viewers are “more likely to notice disclosures made in both audio and video”, so say it and show it.

And keep it honest. Disclosures 101 also says: “If you’re paid to talk about a product and thought it was terrible, you can’t say it’s terrific.”

Instagram’s and YouTube’s rules

Instagram’s Branded Content Policies define branded content as “a creator or publisher’s content that features or is influenced by a business partner for an exchange of value”. Its help page on the paid partnership label says that for organic branded content, “our branded content policies require that you use the paid partnership label to tag the brand.” Both pages build their text with JavaScript; we read these lines in their HTML source.

YouTube’s help page says that if you feature “branded content, sponsorships, endorsements, or other commercial relationships”, you have to select the paid promotion button in your video details. That adds a disclosure label at the start of the video. It also says creators and partners are responsible for “all applicable legal requirements”, and points to the FTC in the United States.

The FTC warns, “Don’t assume that a platform’s disclosure tool is good enough.” Use the label and say it in the video. TikTok’s pages don’t load from our network, so check its branded content rules in the app.

Layouts: split screen, alternating cuts, table graphic

Choose the layout by what the viewer needs to see: both sides at the same moment, each side full-frame in turn, or the points written out.

Split screen: both at once

Use it when the difference shows in the same moment: two vacuums on one spill, two phones pointed at one skyline, two maps apps on the same route, as in @giaphuongle0’s post. Fifteen posts in our set used a split screen, or the creator cloned into each half. Keep each side in the same half for the whole reel.

Alternating cuts: one, then the other

Each side gets the full frame in turn. It was the most common layout in our set, on 18 posts, and it needs no split-screen edit. Frame both sides the same way, label each one as it appears and cut on the word that names it. @tianagervasoni did it in a single shot, switching between the two phones.

Table graphic: the points written out

Use a table or grid when the points are words or numbers, like @aru_13275’s phrase table or @iamzacsmith’s two-box grid. Nine posts used one. Instagram says it aims to make “reels that are majority text” less visible, so keep the graphic to part of the frame with you in view.

A fourth option is to hold both things in one shot, as @cfknutrition did with two plates; ten posts did. If you play both sides yourself, as the cloned creators did, give each side a fixed look and position. The data shows every layout among winners; it cannot say which works best. These suggestions are ours.

Source every fact you put on screen

Every number or claim that appears on screen needs a source you could show if asked: your own test, the maker’s spec page, a receipt or the product label.

• Your own test. Film the measurement itself, such as the juice in a measuring jug or the reading on a scale, the way @jordan_the_stallion8 set both results side by side.
• Specs and prices. Take them from the maker’s own page or your receipt, and note the date. Prices change, so date them on screen or in the caption.
• Rules and features. Take them from the company’s own help pages, as in the worked example above, not from another creator’s video.
• Labels. Our approach: a small “Source: YouTube Help” on screen while the claim is spoken, and the full links in the post caption. The worksheet has a source column for every line.

If a brand pays you, one more line from the FTC is relevant. Its answers say an endorsement “can’t be used to make a claim the marketer of the product couldn’t legally make.” Our reading of that general principle, not an FTC rule about comparisons: hold every claim you make about the rival to the standard the brand itself would have to meet.

End on a verdict: pick A if…, pick B if…

Close with one line that tells each viewer which side fits them: “Pick A if…, pick B if…”. If one side wins outright, say that instead.

Most posts in our set ended on a pick: 44 of 53 named a winner or told viewers what to do. The worked example above ends like this: “Pick Shorts if you want a share of ad revenue and can work toward YouTube’s views threshold. Pick Reels if your audience is already on Instagram and you earn from brands.”

• Build the verdict from your three points. No new reason in the last line.
• Say it and show it. Put the verdict on screen as you say it, in the same labels and colors as the two sides.
• Promise it at the start. An opening such as “Dyson or Shark for pet hair? I tested both” tells viewers an answer is coming. The hook library has opening lines to adapt.
• Ask which they would pick. Two captions in our set that named a winner still asked “Which one would you pick?” or “Which would you choose?”. It gives viewers who disagree something to answer.

These are our editing choices. The data shows that most winners ended on a pick; it can’t show that a pick made them win.

Captions, labels and the safe zone

Caption every word, label each side the same way every time it appears, and keep labels and the verdict inside the safe zone.

• One label per side, fixed. The same name, color and position from the first frame to the verdict. In a split screen, put each label in its own half.
• Short labels. “Shark” and “Dyson”, not the full model names; put those in the caption.
• Captions in two short lines, using the same name for each side as its label.
• No third-party watermark. Instagram’s creator FAQ says one may be hurting a reel’s reach.

Meta’s Reels ad specs suggest keeping the top 14%, the bottom 35% and 6% of each side free of text and key elements. They are written for ads, but they make a useful layout check. On a 1080 × 1920 frame, that leaves a band from about 270 to 1,250 pixels down, about 65 pixels in from each side. In a top-and-bottom split screen, keep both halves’ labels and the key action inside that band. Check a frame in the Reels safe zone checker before you export.

Where EditorOP fits

This vs that is not available in EditorOP. If your comparison is really a ranking of several options, the Tier list format on EditorOP’s start screen covers it: “Rank things S to C. Each one drops into its row as you say it.” The free tier list maker builds the same kind of board as a 1080 × 1920 image you can talk over.

Comparison reel mistakes

Run through this list before you post.

• A choice nobody is making.
• One side tested at its weakest, or under different conditions.
• More points than the time holds, so each gets half a sentence.
• A number on screen with no source, or a price with no date.
• A brand paid you, sent the product or gave a discount, and the video doesn’t say so. The FTC counts all three.
• A disclosure only in the caption or description, which the FTC says is not enough on its own.
• Criticizing the rival of a brand that pays you without disclosing the deal.
• A table that covers most of the frame on Instagram, which aims to make “reels that are majority text” less visible.
• Labels that change side, color or name halfway through.
• No verdict, so the viewer has to guess what you think.
• Labels or the verdict under the app’s buttons and caption.

This vs that questions