Which wins, carousel or single image?
It depends on what you sell and what you want the click to do. Every public dataset that favors carousels measures clicks, and clicks are what the format is built for: Meta’s help page lists “help raise your clickthrough rate” as its first use.[1] The datasets that measure purchases are closer. MHI’s vendor analysis has single images converting at 2.1% against 1.8% for carousels on products under $50, level between $50 and $100, and carousels ahead at 1.7% against 1.4% above $100.[10] Lebesgue’s 2024 medians give image, video and carousel ads the same 1.5% conversion rate.[11]
The pattern is simple. A carousel earns more clicks because it gives people more to do. Whether those clicks become sales depends on whether the product needed more than one frame. A $20 item with one obvious benefit does not. A $300 item with four features, or a service with a four-step process, often does.
What does Meta say a carousel ad is?
Meta’s definition: “The carousel format lets you show 2 or more images and videos in a single ad, each with its own headline, description, link and call to action.” It runs across Facebook, Messenger, Instagram and Meta Audience Network.[1] The best-practices page caps it at 10 cards and asks for images “from the same shoot or that feel similar to each other” and a different link on each card when you promote different products.[2] Two things happen on their own. A carousel “may appear as a slideshow video automatically when this format is likely to improve your ad performance”, and on Instagram the ad may open on a different card for different people, with the order unchanged.[1]
The third is card ordering. Meta’s help page says: “We automatically order the carousel cards based on how they perform among people who view your ad.” The toggle is called Highlight carousel card, it is on by default, and it applies only in Facebook Feed, Instagram Feed and Instagram Stories; elsewhere, cards show in the order you set.[3] For a story that develops card by card, Meta says to turn it off.[4] Specs are per placement: in Facebook Feed a carousel is 1:1 at 1080 by 1080 or more, with 80 characters of primary text and a 20-character headline per card;[5] in Instagram Feed an image-only carousel can be 4:5, but any video makes it 1:1.[6] A single image in Facebook Feed is 4:5 at 1440 by 1800.[7]
What do the CTR and conversion figures actually measure?
Five sources account for nearly every number quoted in carousel-versus-image posts. Only one is Meta’s, and it is a sentence, not a figure. The verdict column is ours.
| Source | Sample and period | Measured | Finding | Verdict |
|---|---|---|---|---|
| MHI Growth Engine[10] | 1,247 ecommerce campaigns, Q4 2025 to Q1 2026; its own clients | CTR, conversion rate by price band, ROAS | Carousel CTR +42%; single image converts better under $50, carousel above $100; ROAS 3.1x vs 2.7x | Vendor data, no method; the same lift appears as 42%, 44% and 48%. |
| Lebesgue, July 2024[11] | Its clients; $18.6 million of spend cited for one chart | Median CTR, CPM, conversion rate | CTR 0.9% carousel, 0.91% image; CR 1.5% for both; carousel CPM lowest at about $5.13 | Vendor medians; level. |
| Socialinsider, 2026[12] | 35 million organic Instagram posts | Engagement rate of organic posts | Q2 2026: carousels 0.50%, images 0.33%, Reels 0.48% | Organic posts, not ads. |
| Kinetic Social via Marketing Dive, 2015[13] | Kinetic Social clients, 2015 | Clickthrough to advertiser sites | 1% for carousels against 0.1% for other formats | The origin of “10x”. Eleven years old. |
| Meta, conversion testing help[9] | Direct response research; count not given | Conversion lift | “Upgrading from still images to the carousel format can also be effective.” | A sentence, not a number. |
So the honest range is “level” to “about 40% more clicks”, and the one figure that sounds decisive, 10x, is from 2015.[13] Socialinsider’s organic Instagram figures, where carousels lead images by 0.17 points in Q2 2026, are often pasted into ad posts; they measure likes and comments on unpaid posts, not ad clicks.[12] Static vs video Facebook ads: what the data shows reads the image-versus-video studies the same way.
When does each format win? A decision table by goal and budget
This table is our judgment, built on Meta’s stated uses for each format and the vendor price-band data above. Meta draws one line itself: “Use product imagery to reach people who have visited product pages on your website. Use lifestyle imagery to reach prospective customers.”[2]
| Goal | Under $50 a day | $50 to $150 a day | Why |
|---|---|---|---|
| Cold reach, one product or service | Single image | Single image, plus one carousel to test | One message lands in one glance, and a small budget cannot read two formats at once. |
| Catalog or several products | Carousel | Carousel, or Advantage+ catalog ads | Meta’s first listed use: “Feature multiple products that link to different landing pages.”[1] |
| One offer or promotion | Single image | Single image | The offer is the headline. A second card dilutes it. |
| A process or a before-and-after | Carousel, ordering off | Carousel, ordering off | Meta’s “Tell a story” and “Explain a process” uses; ordering off so the steps stay in order.[1][4] |
| Retargeting site visitors | Carousel of the products they viewed | Carousel or catalog ads | Meta: product imagery for people who visited product pages.[2] |
| Local service, first ad | Single image | Single image, then a carousel of reviews | A proven layout is the fastest way to learn whether the message works. |
Price matters as much as goal. In MHI’s data the conversion gap flips with price: single images win by 17% under $50 and carousels by 21% over $100, by that vendor’s own arithmetic.[10] People buy a cheap thing on one glance and an expensive thing after a look around.
Does budget change the answer?
Yes, in two ways. First, a carousel costs more to make. Meta asks for 2 to 10 cards that look like they came from the same shoot,[2] so a three-card carousel is three pictures, three headlines and three links. At a small budget that time is better spent on three different single-image concepts, which gives Meta three separate bets instead of one ad with three parts. How many ad creatives to test at once covers the arithmetic.
Second, a carousel splits attention. Meta’s default ordering moves the card people respond to first,[3] but every card after the first is seen by fewer people. If what sells your product fits on one card, the other cards are cost. The cheap test is the strongest card alone as a single image against the full carousel, in one ad set, for a week. Under $50 a day, start with the single image and add the carousel once the first ad has paid for itself.
How to build a carousel out of single-image ads
The quickest carousel is three or four of your static ads in a row, since each card is a static ad with its own headline and link.[1] Pick the layouts first from the eight static ad layouts big brands use.
- 1
Make card one stand on its own
Meta’s ordering applies in three placements; everywhere else, people see your first card first.[3] Give it the headline and the product. - 2
Keep the cards from one shoot
Meta asks for images “from the same shoot or that feel similar to each other”.[2] Same background, same type, same logo corner. - 3
One link per card
The blouse card goes to the blouses and the shoes card to the shoes, Meta’s own example.[2] A smaller shop sends each card to the category page. - 4
- 5
- 6
Test it against your best single image
Same ad set, same week. Showfolk makes the single-image ads a carousel is built from: four varied versions of every ad in the feed and story shapes, lettered in your brand’s fonts. How the Instagram ad maker does it.
Questions people ask
- Do carousel ads perform better than single-image ads on Facebook?
- On clicks, usually; on sales, it depends on the product. MHI’s vendor analysis of 1,247 ecommerce campaigns found carousels drew 42% more clicks but single images converted better for products under $50. Lebesgue’s 2024 medians put both formats at about 0.9% CTR and 1.5% conversion rate. Meta publishes no head-to-head.
- What is a good CTR for a carousel ad?
- Lebesgue’s 2024 median for carousel ads is 0.9%, level with single images at 0.91%. MHI reports carousels 42% above single images in its own client data. Neither is a Meta figure, so judge your carousel against your own single images, not against these.
- How many cards should a carousel ad have?
- Between 2 and 10, Meta’s technical range. In practice three to five cover most uses: a product set, a process in steps, or a before-and-after. Each card after the first is seen by fewer people, so put the strongest card first.
- Does Meta automatically reorder carousel cards?
- Yes, by default. The Highlight carousel card toggle orders cards by how they perform among people who view the ad, in Facebook Feed, Instagram Feed and Instagram Stories only; elsewhere cards show in the order you set. Turn it off for a carousel that tells a story in sequence.
- Is the claim that carousels get 10x the clickthrough true?
- It was a 2015 figure. Marketing Dive reported Kinetic Social data showing 1% clickthrough for carousels against 0.1% for other formats, months after the format launched. Current vendor benchmarks put the gap between level and about 40%.
Sources
Every figure above points at one of these. Facts were last checked on .
- [1]Meta Business Help Center: About carousel ads. The definition, the six uses (products with different links, features, story, process, larger canvas, benefits), the placements, the automatic slideshow note and custom positioning on Instagram
- [2]Meta Business Help Center: Best practices for carousel ads. Up to 10 cards, images from the same shoot, square video, a link per card, product imagery for site visitors and lifestyle imagery for prospects, the carousel optimizations list
- [3]Meta Business Help Center: Show the best-performing cards first in a carousel ad. Highlight carousel card is on by default, orders cards by performance, and applies only in Facebook Feed, Instagram Feed and Instagram Stories
- [4]Meta for Business: Carousel ad format. Up to 10 images or videos each with its own link; let Facebook optimize the order based on each card’s performance, or opt out for a sequential story
- [5]Meta Ads Guide: Carousel ad specs for Facebook Feed. 2 to 10 cards, 1:1, at least 1080 by 1080, 30MB per image, 80 / 20 / 18 characters recommended
- [6]Meta Ads Guide: Carousel ad specs for Instagram Feed. 4:5 for image-only carousels; 1:1 only when a carousel contains video; 125 characters of primary text
- [7]Meta Ads Guide: Image ad specs for Facebook Feed. 4:5 at 1440 by 1800 pixels for a single image
- [8]Meta Business Help Center: About image ads across Meta technologies. An image ad is a single photo and an optional footer; one of its uses is to deliver a simple message
- [9]Meta Business Help Center: Creative best practices for conversion testing. “Upgrading from still images to the carousel format can also be effective.”
- [10]MHI Growth Engine: Single image ads vs carousel ads for ecommerce, CTR and conversion data comparison. Vendor analysis dated February 17, 2026: 1,247 ecommerce campaigns, 42% higher carousel CTR, conversion rates by price band (2.1% vs 1.8% under $50; 1.4% vs 1.7% over $100), ROAS 2.7x vs 3.1x; the same lift is also given as 44% and 48% on the page
- [11]Lebesgue: Facebook ads creatives, benchmarks and best practices. Posted July 8, 2024: median CTR 0.91% image, 0.9% carousel, 1.0% video; median conversion rate 1.5% for image, video and carousel; carousel CPM about $5.13; $18.60 million of spend cited
- [12]Socialinsider: Instagram benchmarks 2026. Organic Instagram engagement by format from 35 million posts: Q2 2026 carousels 0.50%, images 0.33%, Reels 0.48%
- [13]Marketing Dive, October 20, 2015: Facebook’s carousel ads are already outpacing other ad formats. Kinetic Social’s 1% clickthrough for carousels against 0.1% for other formats, the origin of the 10x claim
