Facebook Ads Examples: 15 Real Ads From Meta's Case Studies (2026)

Most posts about facebook ad examples show you a screenshot and a guess about why it worked. This one does not. Every example below comes from a case study Meta itself published on business.meta.com or facebook.com/business/success, so the format, the offer, and the results are Meta's own words, not ours.

We grouped 15 of them by ad format: photo, video, Reels and Stories, carousel and collection, lead ads, and messaging ads on WhatsApp, Messenger and Instagram Direct. For each one you get who ran it, what the ad actually did, the Meta-documented reason it worked, and one thing to copy for your own account. Every result links straight to its source.

What Makes a Facebook or Instagram Ad Work in 2026

The 15 case studies below do not share a budget, industry, or country. They share five habits, each traceable to Meta's own product data or creative guidance.

• More creative variety beats one hero ad. On its own Advantage+ creative page, Meta reports a 2 to 3% lift in conversions from background generation on catalog ads, a 2% lift for Facebook Reels from video expansion, and 13% more conversions from related media (Meta Advantage+ creative).
• Vertical, sound-on, made-for-the-feed beats a resized video. The Nespresso Türkiye example below spells out what Meta calls its Reels creative best practice: entertaining and visually varied, understandable in the first seconds with text overlays, full-screen 9:16, audio on.
• Creator and partnership ads reach past your own followers. A partnership ad publishes from both the brand's handle and the creator's. Condé Nast's own meta-analysis of its partnership ad campaigns found a 1.8x median incremental return on ad spend in the second half of 2024 (Condé Nast case study).
• Messaging destinations convert what a landing page loses. Five of the examples below route the click into WhatsApp, Messenger or Instagram Direct instead of a website, because a question answered in a conversation closes a sale a form never would.
• Every case study below ran a test. An A/B test, a conversion lift study, or a brand lift study, not a launch-and-hope. That is the most copyable habit on this list.

Feed and Photo Ads

A well-made photo ad still owns most Feed and Marketplace placements. The gains Meta reports now come from testing more variations per campaign, not from one perfect shot.

Living Proof: Advantage+ Shopping with AI-generated backgrounds

The haircare brand under Unilever Prestige fed three product photos into Meta's generative AI background generation feature, got 20 new image variations back, and ran them inside an Advantage+ Shopping campaign against its usual creative and setup, shown to US adults 18 and older, from May 3 to 19, 2024 (Meta case study).

Why it worked: the AI backgrounds cost almost no production time yet won on every metric Meta measured: 18% more purchases, 16% lower cost per purchase, and 43% lower cost per purchase among 18 to 24 year olds. Copy this: before shooting a new batch of product photos, generate background variations of the ones you already have and let Advantage+ placements pick the winner.

PRESS Healthfoods: stacking every Advantage+ creative feature at once

The UK plant-based food retailer ran an A/B test from April 26 to May 13, 2024, comparing its usual photo ads to an Advantage+ Shopping campaign using Advantage+ creative's full AI stack (standard enhancements, image generation, and text generation), shown to UK adults 18 and older (Meta case study).

Why it worked: stacking multiple AI creative features, instead of testing them one at a time, drove 40% more product sales and 33% lower cost per purchase. Copy this: turn on Advantage+ creative's image and text generation together on your next Shopping campaign, this case study measured the combination, not either feature alone.

Video Ads

Meta's video guidance below still comes down to a hook in the first seconds and sound on, and it applies to plain Feed video just as much as Reels.

Underneat: partnership video ads, checked against search demand

The Indian shapewear brand ran video ads with established creators as partnership ads, published from both the brand's and the creators' handles, driving viewers to shop now on its website, then measured impact with Meta's search lift methodology (Meta case study).

Why it worked: a partnership ad borrows the creator's audience and credibility, which is why the lift showed up off-platform too: a 99% lift in organic search traffic and a 75% lift in branded search. Copy this: turn creator content you already have into a partnership ad, then check search demand, not just Meta's own dashboard, for the real effect.

Vanaura Organics: video partnership ads inside an Advantage+ Sales campaign

The Kerala-based organic skincare brand put 30 to 40% of its budget behind video partnership ads from local creators who already used the product, run alongside its usual product ads inside one Advantage+ Sales campaign, from June 18 to July 2, 2026 (Meta case study).

Why it worked: authentic creator video plus automated placement testing compounded into a 25% lift in purchases and a 4.06x return on ad spend on add-to-cart. Copy this: run creator content and Advantage+ automation inside the same campaign, Vanaura's win came from combining both, not from creator content alone.

Reels and Stories Ads

Nespresso Türkiye: Reels trending ads built to Meta's own reels brief

To introduce its Vertuo brewing system to younger audiences, Nespresso Türkiye built video specifically for Reels trending ads, following what the case study describes as Meta's reels creative best practice: entertaining and visually varied, understandable in the first seconds with text overlays, full-screen vertical 9:16, audio on, featuring its celebrity ambassador using the product on camera. The test ran from April 17 to May 31, 2026, against the brand's usual ads (Meta case study).

Why it worked: creative built for the format, not cropped into it, lifted ad recall by 3.8 points and favorability by 4.3 points. Copy this: before boosting a Reels budget, check that the hook lands in the first two to three seconds in native vertical video, not a cropped landscape clip.

Carousel and Collection Ads

When there is more than one product or angle to show, collection and carousel formats show up in every case study below that also ran a real test against a single image.

ALDI Nord: one creator concept, four ad formats at once

To introduce its private-label dairy brand MILSANI to 18 to 34 year olds in Germany, ALDI Nord built one creator-led concept, using nostalgia, playful visuals and casual language, then shipped it as carousel, collection, photo and video ads plus creator partnership ads, from September 22 to October 5, 2025 (Meta case study).

Why it worked: a Meta Brand Lift Study credited running the same idea natively across every format with an 8-point lift in purchase intent and a 7.8-point lift in brand awareness. Copy this: rebuild a working concept natively in each format instead of resizing the same crop.

American Eagle Mexico: carousel and collection for an omnichannel test

The apparel retailer built the same discount offer as a shoppable collection ad and a swipeable carousel ad, then split traffic between a web-only optimized version and an omnichannel version, optimized for both online and its 100+ Mexican stores through Meta's Conversions API, from April 16 to May 16, 2026 (Meta case study).

Why it worked: the format stayed constant, the optimization goal was the variable, and omnichannel won: 20.4% more incremental purchases at 46% lower cost per incremental purchase. Copy this: if you have physical stores and offline event tracking connected, do not optimize a collection or carousel ad for web sales alone.

Nespresso in the UAE: one campaign, four formats, three objectives

For a limited-edition product line, Nespresso in the UAE ran an Advantage+ Sales campaign alongside carousel, collection, photo and video ads, set to awareness, consideration and sales objectives at once instead of one funnel stage at a time, from September 1 to October 12, 2025 (Meta case study).

Why it worked: measured with both a Conversion Lift test and a Brand Lift test, the combined approach produced a 9.2-point ad recall lift, an 11.2-point purchase intent lift, and a 2.5x return on ad spend. Copy this: for a limited-time launch, brief every format for one campaign and one window instead of staggering awareness and sales separately.

Lead Ads

OREO: lead ads, video, Stories and Reels for a sweepstakes

To promote a sweepstakes, OREO ran ads on the exterior of the Sphere in Las Vegas alongside two Meta lead ad campaigns, one offering cash and one offering cookies for life, placed across Feed, Stories and Reels, with creative asking "How many OREO cookies do you see?" and a pre-filled instant form behind the click (Meta case study).

Why it worked: running awareness and lead objectives at once, with two different offers, let OREO see which prize actually moved people: 28 million people reached and 982,000 leads, with only 3.4% audience overlap between the two campaigns. Copy this: test two offers in parallel lead campaigns instead of two headlines, the offer moved OREO's numbers more than any copy tweak would have.

Peugeot Türkiye: one automated setting on a lead campaign

Instead of manually sending every lead ad click to one destination, Peugeot Türkiye let Meta's Advantage+ destination choose, person by person, between a website lead form and an in-app instant form, tested against its usual manual setup from March 18 to April 6, 2026 (Meta case study).

Why it worked and what to copy: removing that one manual choice was worth 4x more leads at 74% lower cost per lead, often the biggest untested lever in a lead campaign.

Messaging and AI Agent Ads

The fastest-growing pattern in this list is not a placement, it is a destination: ads that open a conversation on WhatsApp, Messenger or Instagram Direct instead of a landing page.

Analu Ballestero: click-to-WhatsApp with a bigger budget

A Brazilian custom-accessories maker whose website traffic rarely converted switched to ads that click to WhatsApp so she could answer personalization questions directly, then ran a two-week A/B test from August 13 to 27, 2022, comparing a higher budget on that same campaign against the original (Meta case study).

Why it worked and what to copy: more budget behind the messaging ad bought more conversations, not just more reach: 57% more messaging conversations started and 69% more link clicks. If your product needs explaining before a sale, put the click into a conversation instead of a landing page.

DIVA Beauty Center: video lead ads to Messenger, answered by an AI agent

The Vietnamese skincare clinic ran video lead ads that click to Messenger, then split incoming conversations in an A/B test between its usual human-staffed team and Meta Business Agent, an AI agent trained on the clinic's own pricing, locations and FAQ, over two weeks from April 9 to 23, 2026 (Meta case study).

Why it worked: the AI-handled conversations ran around the clock and converted better than the human-only control: 30% lower cost per conversation, 38% more bookings, 58% more revenue. Copy this: put an AI agent on the Messenger thread you already use before hiring for night and weekend coverage.

ELSA Speak: Messenger ads for markets with no local support team

The AI English-learning app paired its usual ads that click to Messenger with Meta Business Agent to qualify subscribers in markets where it has no on-the-ground support team, testing it against its previous third-party messaging plugin over two weeks from April 12 to 25, 2026 (Meta case study).

Why it worked and what to copy: after one training session on ELSA's own knowledge base, conversations converted at 50%, driving 26% more qualified leads at 18% lower cost per lead, before the company hired any local support.

Sprite Indonesia: a whole campaign built inside WhatsApp

For Ramadan 2026, Sprite Indonesia rebuilt a retailer promotion that used to require a selfie upload and a separate web app into a single WhatsApp flow: store owners registered using only their store name, received a personalized AI-generated video featuring a celebrity, and shared it to earn a cash incentive (Meta case study).

Why it worked: removing every step that happened outside WhatsApp fixed the prior year's sub-50% AI video success rate, store owner registrations grew 232% year over year. Copy this: count how many steps of your campaign mechanic happen outside the messaging app itself, each one is a drop-off point this case study shows you can remove.

GUMAC: live shopping ads that click to Messenger

The Vietnamese fashion brand, which streams live shopping twice daily and gets 60 to 70% of its online orders from Facebook, ran live shopping ads that click to Messenger and switched the campaign's optimization goal from views to purchases, A/B tested over five days from December 25 to 29, 2025 (Meta case study).

Why it worked and what to copy: optimizing for the action GUMAC actually wanted, not audience size, produced a 6.8x return on ad spend and a 5.2x lower cost per purchase. If your live shopping or Messenger ads still optimize for views, that is usually the wrong goal to isolate first.

Social Ads Beyond Meta: Two TikTok Examples

Two more examples, sourced directly from TikTok's own case studies rather than Meta's, for anyone running both platforms. Virgin Voyages mobilized 1,103 creators through TikTok's Creator Marketplace and Creative Exchange, tagged under #SailingWithVirgin and amplified with Branded Buzz, generating 22.7 million video views and a 696% increase in TikTok-referred website traffic, published May 29, 2026 (TikTok case study).

Hormel Foods connected Smart+ Catalog Ads to Instacart for the playoff and Big Game snacking window, running TikTok-first video and carousel creative, and reported a 1.17% click-through rate on Smart+ Sales alongside a 10.9-point lift in ad recall and a 4-point lift in purchase intent from its Brand Lift Study (TikTok case study).

Swipe File Checklist: What These 15 Ads Have in Common

• Build creative for the format it runs in (vertical, sound-on for Reels; a real second angle for carousel), not a resized version of something else.
• Test more than one creative variation per campaign. Meta's own Advantage+ data shows the gain comes from volume, not one perfect asset.
• Turn existing creator relationships into partnership ads instead of only boosting your own handle.
• Route anything that needs explaining, customization, booking, or pricing questions, into WhatsApp, Messenger or Instagram Direct instead of a landing page.
• Optimize for the action you actually want (purchase, qualified lead), not the easiest one to hit (views, messages).
• Measure with more than platform-reported clicks where you can: a conversion lift test, a brand lift test, or search lift, the way most of the brands above did before calling a result real.

Turn Any Example Into Your Own Campaign

None of the formats above need a special account, they are standard Meta placements and objectives. Start with the example closest to your own offer, then rebuild the same mechanism, not the same creative, for your product. In Adship's Canvas, generate ad copy and image or video variations around that same angle instead of starting from a blank page, matching the hook and format rather than copying its exact words.

Then use Adship's bulk ad creation flow to assign those variations across campaigns and ad sets and publish them with delivery status tracked for every target, not rebuilt by hand in Ads Manager.

Frequently Asked Questions

Where can I find more real Facebook ad examples?

Meta publishes new case studies continuously at facebook.com/business/success, organized by product (lead ads, carousel, Advantage+) and by industry. Everything in this article was sourced from that same collection, so the format, offer and result described for each example match what Meta itself reported.

Are the results in Meta's case studies typical for every advertiser?

No, and Meta says so directly. Several of the case studies above carry Meta's own disclaimer that results are self-reported, not identifiably repeatable, and that individual results will differ. Treat every percentage here as evidence a mechanism can work, not a guarantee for your account, budget or industry.

What is the difference between a lead ad and an ad that clicks to Messenger?

A lead ad opens an instant form inside Facebook or Instagram and submits structured fields (name, email, phone) without leaving the app. An ad that clicks to Messenger or WhatsApp opens a conversation instead, which suits products that need a question answered, pricing, sizing, availability, before someone is ready to hand over their details.

Do I need Advantage+ to get results like these?

No. Several examples above (Living Proof, PRESS Healthfoods, Vanaura Organics, Nespresso in the UAE) specifically tested Advantage+ Shopping or Sales campaigns against a manual setup and won. Others (Analu Ballestero, OREO, GUMAC) used standard lead, video or messaging ad types with no Advantage+ campaign involved. The format mattered more than the automation layer in most of these results.

How big does my ad account need to be to run partnership ads with creators?

None of the case studies above state a minimum spend for partnership ads. What they share is an existing creator relationship: Underneat, Vanaura Organics and ALDI Nord all worked with creators who already used or knew the brand before turning that content into a partnership ad.

How long should I run a test before trusting the result?

Most of the tests above ran two to six weeks: Analu Ballestero's budget test ran two weeks, GUMAC's ran five days, Nespresso Türkiye's ran about six weeks. Match your test length to how long your account normally takes to exit Meta's learning phase, not a fixed calendar rule.

Can I use these same formats on TikTok?

Most of them, yes. TikTok's own case studies, see Virgin Voyages and Hormel Foods above, show the same underlying patterns: creator partnerships extending reach, and catalog or Smart+ automation handling optimization. The platform-specific piece is the creative itself, TikTok's native format expectations sit closer to Reels than to a Feed photo ad.

Sources

Every example above links to its original case study. Full list:

• OREO: Facebook ads case study
• Analu Ballestero: Facebook ads case study
• Living Proof: Facebook ads case study
• PRESS Healthfoods: Meta for Business case study
• Condé Nast: Facebook ads case study
• Peugeot Türkiye: Facebook ads case study
• DIVA Beauty Center: Facebook ads case study
• Nespresso Türkiye: Facebook ads case study
• Sprite Indonesia: Facebook ads case study
• Vanaura Organics: Facebook ads case study
• Underneat: Facebook ads case study
• ALDI Nord: Facebook ads case study
• American Eagle Mexico: Facebook ads case study
• Nespresso in UAE: Facebook ads case study
• ELSA Speak: Facebook ads case study
• GUMAC: Facebook ads case study
• Meta Advantage+ Creative, Meta for Business product page
• Virgin Voyages, TikTok for Business case studies
• Hormel Foods, TikTok for Business case studies