Resources / Facebook carousel ads

Facebook Carousel Ads: 5 Complete Examples and a Storyboard

By Maestro · Updated September 29, 2026

A Facebook carousel ad puts two to ten image or video cards inside one ad. Each card can carry its own headline, description, link and call to action. The format earns its extra production work when the cards do different jobs: explain a process, compare products, answer objections or move from a problem to proof.

Use a single image when you have one point and one destination. Use video when movement, a demonstration or a speaker carries the message. Use a carousel when the order or the choice between cards helps the buyer understand what to do next.

The planner on this page turns that decision into a card-by-card handoff. It checks whether the message still makes sense when someone stops after card one, card two or the card before the close. It also keeps the platform CTA, destination, arrival headline and arrival action together. A clear ad can still fail if its click leads somewhere else.

Choose carousel, a simpler format or hold

Start with the number of distinct jobs the message needs to do. Do not start with the number of cards you would like to fill.

A carousel fits when card one gives a complete reason to care and every later card adds something different. If the entire argument depends on reaching card eight, the opening is too weak. If card two merely restates card one, the sequence has not earned a swipe.

Keep four decisions separate:

1. Format fit: Would several cards help this message?
2. Storyboard completeness: Does every card have a role, copy, visual and destination?
3. Local production review: Are the assets, claims, rights, links and tracking ready for handoff?
4. Native readiness: Has the team checked the current account, objective, placements, preview and approval state in Meta?

A missing UTM should hold native readiness. It does not suddenly make a useful step-by-step message unsuitable for a carousel.

Current Facebook carousel ad specs to check

Meta changes fields and placement behavior, so keep the current Ads Guide open while you build. The values below were checked on September 29, 2026 for an awareness carousel in Facebook Feed.

Meta's broader carousel design specifications cover other aspect ratios and placement families. Do not assume one square export will work everywhere. Check the objective and placements selected in your own account, then inspect the actual preview.

The platform may also offer Automatically show the best performing cards first. That can suit a product range where order does not matter. It conflicts with a linear story. If card three depends on card two, verify that the native setup preserves your order.

These are platform requirements and recommendations. They do not prove the ad will be approved, delivered or effective.

Five complete Facebook carousel ad examples

The five records below were captured from Meta's public Ad Library archive on September 13, 2026. Every available card is shown in the archived order. The destination pages were inspected on September 29 where a usable URL existed.

The examples show production choices. They do not reveal spend, targeting, delivery, conversion rate or business results. A weak destination trace is included because it teaches a useful review step.

1. BELFOR Franchise Group: one family, many service options

BELFOR's eight-card Ad Library record opens with: “Make a difference in your community. Contact our team today!” The cards then name eight franchise categories:

1. Contents restoration
2. Portable storage
3. Property restoration
4. HVAC
5. Junk removal
6. Kitchen makeover
7. Plumbing
8. Commercial cleaning

The sequence works as a range. A reader can scan several business models without following a strict story. That makes automatic card reordering less dangerous than it would be for a numbered process.

Every card uses See Details and the same BELFOR Franchise Group destination. The inspected page headline is “BELFOR Franchise Group – Join Our Family of Leading Service Franchises,” followed by “Find out which brand is the best fit for you.” The page continues the broad franchise-family promise and lists the represented brands. It does not open on the named service category from the clicked card.

That is partial continuity. One umbrella page can work if it helps people identify the exact option they clicked. A category anchor, selected state or dedicated route would shorten the distance.

2. IT&3PL Fulfillment: three offers, two destination outcomes

IT&3PL Fulfillment's three-card record combines three offers:

1. Professional business website, starting at $200
2. Warehouse management system, starting at $1,000
3. Dropshipping fulfillment, starting at $1

The first two cards share /it. At inspection, that route resolved to a page headed Warehouse Technology with a Get a Quote action. That continues the warehouse-management card. It changes the offer for the website-building card, so that path receives a weak continuity judgment.

The third card resolves to 3PL Fulfillment Services with Get a Quote. The promise and arrival action match.

The lesson is simple: a working URL can still be the wrong destination. Review the arrival headline and action against the exact card, not the campaign in general.

3. The Battery Post: three cards, three destinations

The Battery Post's three-card carousel begins with “Learn more about our inventory!” and offers three different actions:

1. Batteries for Everything goes to the Batteries category.
2. EZGO TXT goes to a page for a 2016 EZGO TXT.
3. Serving WNY & Beyond goes to the contact page.

The category and contact paths continue their cards. The product path also reaches the named vehicle, but the inspected page now marks it Sold. That is a useful archived-ad check: the creative can stay unchanged while the destination's inventory state changes.

For a product carousel, name the owner who pauses or replaces a card when the item, price or availability changes. A launch-day link check does not cover the next month.

4. Zhong Tuo Machine: a ten-card industrial catalog

Zhong Tuo Machine's ten-card record uses the full card allowance to show different systems and equipment:

1. Color coated steel system
2. Solar mounting system
3. Storage rack system
4. Gutter and downspout system
5. Door roll forming machine system
6. Cable tray roll forming machine
7. High-speed guardrail roll forming machine
8. EPS/EPU sandwich panel production line
9. General equipment and consumables
10. Other industrial equipment

This is a broad product-range structure. Each card gives a buyer a separate path into the catalog, so the headline must do more work than it would in a linear story.

The archive retained only a generic http://fb.me/ URL. There is no defensible arrival headline or action to report. The page must say Destination not assessable. Reconstructing a likely company page would make the example look more complete while weakening its evidence.

5. Eezi HR: three feature angles with no captured link

Eezi HR's three-card record uses three connected messages:

1. Payroll and attendance in one app
2. Simplify payroll for your business
3. From time tracking to payroll

The creative moves from the broad capability to a buyer outcome and then to the workflow behind it. That is a useful feature-sequence pattern.

The archived cards do not contain a destination URL. The sequence can be clear while destination readiness remains unknown. Keep those states separate in your own handoff.

Pick a sequence pattern before writing cards

A useful pattern gives every card a different job. Choose the pattern that matches the buyer's task.

Problem to proof

1. State the costly or frustrating problem.
2. Explain why the usual approach breaks.
3. Show the mechanism or change.
4. Present approved proof.
5. Offer the next step.

Use this for a focused service or software promise. Proof means a sourced fact, product screen, approved quote or other supplied evidence. Do not invent a result to make the fourth card stronger.

Step by step

1. Name the finished task.
2. Show the starting input.
3. Show each meaningful step.
4. Show the finished state.
5. Offer the template, trial or deeper guide.

This works when the buyer can follow the process. If the action requires movement or detailed screen work, video may be clearer.

Product or service range

1. Explain the common buying situation.
2. Give each product or service one card.
3. Route the card to the matching destination.
4. Use the close only if it helps compare or contact.

BELFOR and Zhong Tuo use this structure. Order can be flexible, so native optimization may be reasonable if every card stands alone.

Feature or mechanism

1. Lead with the buyer outcome.
2. Show the input or first capability.
3. Show how the parts connect.
4. Show the output.
5. Ask for the next action.

Eezi HR's three feature angles suggest this route. Keep the outcome visible; a list of product modules without a buyer problem can feel like a navigation menu.

Objection led

1. State the concern in the buyer's words.
2. Answer it with an approved fact.
3. Handle the next concern.
4. Show what the buyer can verify.
5. Offer a low-risk next step.

Use this only when the objections are real and the answers are approved. Do not manufacture resistance so the brand can win an imaginary argument.

Build the storyboard card by card

Use one record for copy, design, destination and paid-media review. A card should not live in one spreadsheet while its URL and claim source live in another.

For every card, record:

• card number and role;
• supplied fact used;
• exact on-card copy;
• headline and description;
• visual to create or show;
• image or video asset reference;
• platform CTA;
• destination URL and UTM values;
• captured arrival headline and action;
• continuity judgment;
• claim or permission source;
• copy, design and destination owner;
• open question.

Then run three stop-point checks.

Stop after card one

Does the intended buyer know what the sequence is about? Can they act without guessing? Card one does not need to explain everything, but it must make a complete promise.

Stop after card two

Has the reader gained a different piece of information? If card two repeats the first card in new words, cut it or give it a new job.

Stop before the final card

Would the reader miss an essential condition, limitation or proof source? Do not hide the qualification on the last card. The close should continue the task, not repair the story.

Decide whether cards share one destination

Use one shared destination when every card supports the same offer and the page continues the entire sequence. A problem-to-proof story often follows this route.

Use separate destinations when cards represent different products, services or buyer paths. The Battery Post is a clear example: category, product and contact are different actions.

For every route, compare four things:

1. Ad promise: what the opening copy and card say.
2. Platform CTA: what the button asks someone to do.
3. Arrival headline: what the page says first.
4. Arrival action: what the page asks for next.

Call continuity strong when the destination continues the same task. Call it partial when the broad promise remains but the clicked choice is not carried through. Call it weak when the offer changes. Use not assessable when the page was not captured.

Do not turn a missing route into a failed route. Missing evidence creates a request: who owns the URL, what page should be inspected, and what must the reviewer capture?

For tracking, keep one campaign convention and vary the card value intentionally. For example:

utm_source=facebook
utm_medium=paid-social
utm_campaign=q4-demand-review
utm_content=card-03-proof

Do not place credentials, personal data or hidden access tokens in a destination URL.

A completed five-card example

The fictional example below uses only the supplied facts in its record. It does not claim a customer result.

Buyer: B2B demand-generation lead preparing a quarterly pipeline review Offer: A worksheet that connects paid-media spend, accepted leads and qualified opportunities Approved facts: The worksheet has spend, accepted-lead and qualified-opportunity fields; it exports a review brief Destination: https://example.com/pipeline-review Arrival headline: “Build your paid-media pipeline review” Arrival action: “Download the worksheet”

After card one, the buyer knows the task. After card two, the sequence has added a specific break. Before the final card, the method and output are already clear. The destination headline and action continue the promise.

This is a planning fixture, not a Maestro result or customer case study.

FACEBOOK CAROUSEL BUILD PROMPT — standalone v1.0

You are preparing one B2B Facebook carousel storyboard and production handoff. Use only the supplied facts and evidence pasted after this prompt. Do not assume access to an article, ad account, CRM, design file or live destination.

Current Meta sources checked 2026-09-29: carousel cards may each have their own headline, description, link and call to action; Facebook Feed guidance lists 2–10 cards, recommends 1:1 and at least 1080 × 1080, and requires a landing-page URL. Meta may offer automatic best-performing-card order. Sources: https://www.facebook.com/business/help/1114358518575630/ ; https://www.facebook.com/business/ads-guide/update/carousel ; https://www.facebook.com/business/help/927639917355563/ . Recheck the selected objective, placements and current native UI before launch.

Public examples are structure references only. They do not establish advertiser spend, targeting, delivery or results. Never invent an owner opinion, result, URL, customer proof, permission, account access or native receipt.

CANONICAL RECORD CONTRACT — maestro.facebook-carousel-record.v1
Root keys, in order: schema_version, record_id, example_status, judgments, campaign, cards, stop_points, owners, checks, source_refs, missing_inputs, next_action.
Judgment enums: format_fit = carousel_fits | simpler_format | hold_missing_evidence; storyboard_completeness/local_preflight/native_readiness = ready | hold; measured_outcome = measured | unknown.
Campaign enums: pattern = problem_to_proof | step_by_step | product_range | feature_mechanism | objection_led | unknown; destination_mode = shared | per_card | unknown; card_order_mode = fixed | eligible_for_optimization | unknown.
Check and stop-point states: passed | failed | unknown | not_applicable. Continuity: strong | partial | weak | not_assessable. Permission: approved | denied | pending | unknown | not_applicable. Arrival state: captured | supplied_fixture | missing | unknown.
Each card must keep: card_id, index, role, supplied_fact, on_card_copy, headline, description, visual_brief, asset_ref, platform_cta, destination_url, utm_content, source_ref, permission_state, arrival_capture, continuity, owners, open_question.
Unknown scalar facts are null. Do not delete required keys or turn unknowns into assumptions.
Missing-input IDs are deterministic: campaign gaps use campaign-<exact_field>; card gaps use <card_id>-<exact_field>; unresolved checks use check-<exact_check_key>; stop points use stop-<exact_stop_point>; delivered evidence uses delivered-campaign-receipt. Replace underscores with hyphens and do not invent aliases. Use the exact supplied owner value when present; otherwise use one canonical owner slot: offer_and_claims, copy, design, paid_media, destination_and_tracking or final_reviewer. Keep one request per gap, in canonical field order, with a reason and the exact evidence that closes it.

METHOD
1. Choose carousel_fits only when 2–10 distinct card jobs help the buyer. Choose simpler_format when one image or a video/document route is clearer. Choose hold_missing_evidence when buyer, offer or desired action is missing. Explain the reason without padding a carousel to a target count.
2. When carousel fits, choose one pattern. Card one must make a complete promise. Every later card adds a different step, product, fact, objection or proof point. The close continues the task.
3. Use only supplied facts. Put the supporting fact and source_ref on every card. If proof, permission, art, URL, tracking, owner or native evidence is absent, keep it null/unknown and add one deterministic missing_inputs request using the contract ID rule, owner and closing evidence.
4. Decide shared versus per_card destinations. For each card preserve platform CTA, URL, UTM content, captured arrival headline/action and continuity. A working URL with a different offer is weak; a broad umbrella page can be partial; an unavailable page is not_assessable.
5. Run the three stop points after card one, after card two and before the final card. Failed or unknown stop points hold storyboard_completeness until resolved.
6. Keep format fit, storyboard completeness, local preflight, native readiness and measured outcome independent. A draft can be ready while launch remains hold. Measured outcome stays unknown without actual delivered campaign evidence.

MISSING CORE INPUT OVERRIDE
If buyer, offer or desired action is absent, return the complete canonical record with format_fit hold_missing_evidence, all later states held/unknown, no invented card copy, and one focused request for each missing core field, ordered buyer, offer, desired_action. Preserve required keys and the supplied facts.

OUTPUT
Return: A) a short format decision; B) one valid JSON fence containing the canonical record itself, with every required key; C) the same cards as a readable Markdown table; D) missing inputs and one next action. No wrapper around the JSON record.

BLANK CANONICAL RECORD
```json
{
  "schema_version": "maestro.facebook-carousel-record.v1",
  "record_id": "facebook-carousel-blank",
  "example_status": "blank_template",
  "judgments": {
    "format_fit": {
      "state": "hold_missing_evidence",
      "reason": "Buyer, offer and card jobs are not supplied."
    },
    "storyboard_completeness": {
      "state": "hold",
      "reason": "Required card fields are incomplete."
    },
    "local_preflight": {
      "state": "hold",
      "reason": "Claims, assets, destinations and tracking are not reviewed."
    },
    "native_readiness": {
      "state": "hold",
      "reason": "Account access and native preview are not evidenced."
    },
    "measured_outcome": {
      "state": "unknown",
      "reason": "No delivered campaign outcome is supplied."
    }
  },
  "campaign": {
    "company": null,
    "buyer": null,
    "situation": null,
    "offer": null,
    "desired_action": null,
    "primary_text": null,
    "pattern": "unknown",
    "destination_mode": "unknown",
    "card_order_mode": "unknown"
  },
  "cards": [
    {
      "card_id": "card-01",
      "index": 1,
      "role": null,
      "supplied_fact": null,
      "on_card_copy": null,
      "headline": null,
      "description": null,
      "visual_brief": null,
      "asset_ref": null,
      "platform_cta": null,
      "destination_url": null,
      "utm_content": null,
      "source_ref": null,
      "permission_state": "unknown",
      "arrival_capture": {
        "state": "unknown",
        "headline": null,
        "action": null,
        "observed_at": null,
        "evidence_ref": null
      },
      "continuity": {
        "judgment": "not_assessable",
        "reason": "Destination has not been inspected."
      },
      "owners": {
        "copy": null,
        "design": null,
        "destination": null
      },
      "open_question": null
    },
    {
      "card_id": "card-02",
      "index": 2,
      "role": null,
      "supplied_fact": null,
      "on_card_copy": null,
      "headline": null,
      "description": null,
      "visual_brief": null,
      "asset_ref": null,
      "platform_cta": null,
      "destination_url": null,
      "utm_content": null,
      "source_ref": null,
      "permission_state": "unknown",
      "arrival_capture": {
        "state": "unknown",
        "headline": null,
        "action": null,
        "observed_at": null,
        "evidence_ref": null
      },
      "continuity": {
        "judgment": "not_assessable",
        "reason": "Destination has not been inspected."
      },
      "owners": {
        "copy": null,
        "design": null,
        "destination": null
      },
      "open_question": null
    }
  ],
  "stop_points": {
    "after_card_1": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": "Does card one give the buyer a complete reason to care?"
    },
    "after_card_2": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": "Does card two add a distinct piece of information?"
    },
    "before_final_card": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": "Is any essential condition or proof hidden on the close?"
    }
  },
  "owners": {
    "offer_and_claims": null,
    "copy": null,
    "design": null,
    "paid_media": null,
    "destination_and_tracking": null,
    "final_reviewer": null
  },
  "checks": {
    "claim_review": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    },
    "asset_rights": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    },
    "destination_inspection": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    },
    "tracking_review": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    },
    "native_preview": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    },
    "ad_account_access": {
      "state": "unknown",
      "evidence_ref": null,
      "owner": null,
      "note": null
    }
  },
  "source_refs": [],
  "missing_inputs": [
    {
      "request_id": "campaign-buyer",
      "owner": "offer_and_claims",
      "reason": "The intended buyer is missing.",
      "evidence_request": "Supply the intended buyer."
    },
    {
      "request_id": "campaign-offer",
      "owner": "offer_and_claims",
      "reason": "The exact offer is missing.",
      "evidence_request": "Supply the exact offer."
    },
    {
      "request_id": "campaign-desired-action",
      "owner": "offer_and_claims",
      "reason": "The desired action is missing.",
      "evidence_request": "Supply one desired action."
    }
  ],
  "next_action": "Supply the intended buyer, exact offer and one desired action."
}
```

PASTE SUPPLIED FACTS AND EVIDENCE BELOW:

FACEBOOK CAROUSEL REVIEW PROMPT — standalone v1.0

Review one proposed B2B Facebook carousel record. Use only the supplied record, assets and receipts pasted after this prompt. Do not assume article context, account access, destination state or performance.

Current Meta sources checked 2026-09-29: carousel cards may each have their own headline, description, link and call to action; Facebook Feed guidance lists 2–10 cards, recommends 1:1 and at least 1080 × 1080, and requires a landing-page URL. Meta may offer automatic best-performing-card order. Sources: https://www.facebook.com/business/help/1114358518575630/ ; https://www.facebook.com/business/ads-guide/update/carousel ; https://www.facebook.com/business/help/927639917355563/ . Recheck the selected objective, placements and current native UI before launch.

Public examples are structure references only. They do not establish advertiser spend, targeting, delivery or results. Never invent an owner opinion, result, URL, customer proof, permission, account access or native receipt.

CANONICAL RECORD CONTRACT — maestro.facebook-carousel-record.v1
Root keys, in order: schema_version, record_id, example_status, judgments, campaign, cards, stop_points, owners, checks, source_refs, missing_inputs, next_action.
Judgment enums: format_fit = carousel_fits | simpler_format | hold_missing_evidence; storyboard_completeness/local_preflight/native_readiness = ready | hold; measured_outcome = measured | unknown.
Campaign enums: pattern = problem_to_proof | step_by_step | product_range | feature_mechanism | objection_led | unknown; destination_mode = shared | per_card | unknown; card_order_mode = fixed | eligible_for_optimization | unknown.
Check and stop-point states: passed | failed | unknown | not_applicable. Continuity: strong | partial | weak | not_assessable. Permission: approved | denied | pending | unknown | not_applicable. Arrival state: captured | supplied_fixture | missing | unknown.
Each card must keep: card_id, index, role, supplied_fact, on_card_copy, headline, description, visual_brief, asset_ref, platform_cta, destination_url, utm_content, source_ref, permission_state, arrival_capture, continuity, owners, open_question.
Unknown scalar facts are null. Do not delete required keys or turn unknowns into assumptions.
Missing-input IDs are deterministic: campaign gaps use campaign-<exact_field>; card gaps use <card_id>-<exact_field>; unresolved checks use check-<exact_check_key>; stop points use stop-<exact_stop_point>; delivered evidence uses delivered-campaign-receipt. Replace underscores with hyphens and do not invent aliases. Use the exact supplied owner value when present; otherwise use one canonical owner slot: offer_and_claims, copy, design, paid_media, destination_and_tracking or final_reviewer. Keep one request per gap, in canonical field order, with a reason and the exact evidence that closes it.

REVIEW TASK
1. Validate exact root keys, enums, required card keys, sequential unique card indexes and 2–10 cards when format_fit is carousel_fits. A simpler-format or missing-core-input record may have zero cards.
2. Trace every piece of card copy to supplied_fact and source_ref. Reject invented capabilities, proof, results, URLs, permissions or approvals.
3. Check that card one works alone, card two adds information and no essential condition is hidden on the final card. Recompute the three stop-point states from the supplied record.
4. Inspect each platform CTA → destination → arrival headline/action → continuity judgment. Missing or inaccessible arrivals stay not_assessable. A mismatched offer is weak.
5. Recompute judgments independently. Format fit cannot be rewritten by a later missing UTM. Storyboard readiness requires complete card roles/copy/visual/destination intent and passed stop points. Local preflight requires approved claims/assets/rights plus destination and tracking evidence. Native readiness requires real account access and preview evidence. Measured outcome requires delivered campaign evidence.
6. Recompute missing_inputs. Preserve useful supplied owner and evidence request wording. Remove resolved requests; add stable requests for real gaps.

OUTPUT
Return: A) verdict and reason for each judgment; B) discrepancies by JSON path with consequence and repair; C) stop-point and destination trace; D) one valid JSON fence containing the corrected canonical record itself; E) missing inputs and one next action. Do not wrap the record in a report object. Never lower a gate to produce Ready.

PASTE CANONICAL RECORD, ASSETS AND RECEIPTS BELOW:

Use the storyboard outside this page

Download the blank storyboard when a team wants a readable brief. Use the blank JSON record when the handoff needs to move between systems. The completed Markdown example and completed JSON record show the same fields filled in.

The carousel build prompt turns supplied facts into a provisional sequence. The review prompt checks format fit, card roles, stop points, claims, destinations and readiness. Both prompts must work without this article.

Use these missing-input rules:

• If the buyer, offer or desired action is missing, ask for it before writing cards.
• If proof, permission, art, destination or tracking is missing, keep the relevant field Unknown and name the owner and exact request.
• Never invent a URL, result, customer quote, product fact or approval.
• A storyboard can be complete while local production or native readiness remains on Hold.
• Export the recommendation, reason, evidence gaps and next action with the card data.

Move the reviewed storyboard into Meta

The local planner does not create an ad or prove the native setup. Use it to make the handoff reviewable before someone works inside the account.

1. Confirm access and campaign context. Name the ad account, objective, conversion location and paid-media owner. If account access is unavailable, leave native readiness on Hold.
2. Choose the carousel format. Confirm the selected objective and placements still expose the expected card fields.
3. Add primary text and cards. Use the reviewed order. Upload the exact accepted assets and enter each headline, description, CTA and destination.
4. Check card order behavior. If the story is linear, verify the option to show best-performing cards first will not reorder it. If order is flexible, record why optimization is acceptable.
5. Inspect every placement. Check crop, small text, video behavior, headline truncation, CTA and destination. A clean Feed preview does not prove Stories or another placement is ready.
6. Open every destination. Verify the final URL, arrival headline, arrival action and tracking values. Repeat after redirects or page changes.
7. Save the native receipt. Record who reviewed the preview, when, and which ad or draft reference they checked. Keep approval, delivery and results as later states.

Watch Paid Media Pros from 11:49 to 13:40 for a Creative Hub handoff into Ads Manager. The presenter selects a saved mockup, then checks the URL, primary text, headline and CTA. He also shows that unresolved placement issues remain unresolved after import. The tutorial was visually checked on September 27, 2026. Use it to see the handoff, then follow the current controls in your own account.

Measure the sequence without inventing a winner

Separate what the platform delivered from what the buyer completed.

Delivery and interaction can include impressions, reach, card clicks, outbound clicks and landing-page views. Check the current reporting definitions in the account. A card that earns more clicks may be a useful entry point. It does not prove pipeline impact.

Destination behavior can include page views, form starts, downloads or another action tied to the offer. First confirm that the event belongs to the right card and URL.

Business outcome can include accepted leads, qualified opportunities and pipeline when the CRM can connect them. Keep unmatched or missing records visible.

Test one material question at a time where possible:

• first-card promise;
• controlled order versus eligible optimization;
• number of cards;
• one shared destination versus card-specific pages;
• product-led versus problem-led opening.

Record the version, the question and the evidence that would change the next decision. Do not describe a public example as high-performing without comparable result evidence.

Facebook carousel ads: common questions

How many cards can a Facebook carousel ad have?

Meta's Facebook Feed guide allows two to ten cards. Use fewer when the message has fewer distinct jobs. A ten-card limit is a ceiling, not a target.

Can each card link to a different page?

Yes. Meta describes cards with their own links and CTAs, and its format overview gives product-per-card landing pages as a use case. Use separate routes when cards represent separate products or services. Inspect every destination.

Should I let Meta reorder the cards?

Only when order does not carry the meaning. A product range may tolerate reordering. A numbered process or problem-to-proof story usually cannot. Verify the current native option and record the choice.

Can a carousel contain video?

Meta's current design guidance allows images and videos in carousel cards. Check the objective, placement, file and duration rules in the current Ads Guide, then preview every video card in the account.

What should the first card say?

Give the intended buyer a complete reason to care and a clear sense of what follows. Do not rely on the final card to explain the promise.

Is a clear storyboard ready to launch?

No. The storyboard can be complete while claims, permissions, destinations, tracking, account access, native preview or approval remain unresolved.

Want another set of eyes on the sequence?

Bring the opening copy, every card and the destination plan to Maestro's free Demand Gen Roast. The review can check whether the format fits, where the sequence loses meaning and whether the click continues the promise. Get a free Demand Gen Roast.