Loading article
Loading page...
Your creative got rejected for being a few pixels off, the team rebuilt it twice, and the campaign still didn't launch cleanly. That's the pain behind Instagram ad specs, not the table itself. The problem is usually not that you don't know the dimensions, it's that one master asset has to survive Feed, Stories, Reels, and Explore without getting cropped, masked, or buried under UI.
Instagram's specs have also stopped being a simple production checklist. Current placement guidance pushes 1:1, 4:5, and 9:16 into different roles, with Feed image ads at 1:1, Feed video ads at 4:5, and Stories at 9:16. That shift matters because Meta's inventory is built around mobile-first vertical attention, not static square creative, and the same logic shows up across marketplace work too, where the dominant placement decides the asset shape.
For brand owners running Meta alongside Amazon and Walmart, this is the same problem in a different wrapper. You don't win by “having the specs.” You win by building creative that survives the placement it's delivered into.
The same thing happens launch after launch. A creative clears one placement, then gets rejected in another, or it publishes and the CTA drops behind the caption UI. Teams usually blame Meta, but the issue is a spec mismatch between the asset and the placement.
Most guides on Instagram ad specs stop at a table. That helps until you have to decide whether 1:1 or 4:5 will hold the product, whether a vertical asset can survive Feed delivery, or whether a Reels edit needs a safer crop for UI masking. Meta's own guidance shows how placement changes the job, since Instagram Feed image ads allow a range from 400:500 to 191:100 with an aspect-ratio tolerance of 1%, while Stories and Reels are built for 9:16 delivery and a full-screen mobile frame. The creative has to fit the surface, not just the file spec Instagram Feed image ad guide, Instagram ad specs reference.
Specs are a system for creative adaptability, not a checklist to clear before upload. That matters when one master asset has to survive Feed, Stories, Reels, and Explore, because the wrong crop can wipe out a headline, a product shot, or the offer.
Practical rule: If the primary CTA, claim, or face cannot survive a crop, the asset is not placement-safe yet.
The same logic shows up in Amazon and Walmart PPC. The dominant placement, whether it is a marketplace tile, a Sponsored Brands unit, or a Meta feed slot, determines the creative shape. Treat specs as a distribution problem, and the team stops rebuilding assets every time Meta shifts the inventory mix.
What Clickstera does differently is simple. We audit creative format fit the same way we audit bids, by asking whether the asset matches the placement economics. That is a different question from whether the file is technically valid. For teams that need a faster way to reshape source assets, social media image formatting guidance can help standardize the starting point.
A strong master file also makes repurposing easier across motion and stills. For teams building that workflow, smart video repurposing for Instagram is part of the same problem set, because the edit has to survive more than one placement without losing the hook.
Feed still carries a large share of direct-response spend, so crop mistakes show up fast. The common mistake is designing for the canvas you like instead of the one Feed rewards. For Instagram Feed image ads, Meta allows a minimum width of 500 px, an aspect-ratio range from 400:500 to 191:100, and a 30 MB file limit. In practice, 1:1 and 4:5 fit native inventory best.
Square still works, but 4:5 earns more screen space in the feed, which is why a lot of DR teams use it as the default. The practical advantage is simple. A taller creative holds more of the viewport before the thumb keeps moving.
For Instagram Feed video ads, the practical production target is about 1440 × 1880 px in 4:5, with a 4 GB file cap. A horizontal edit may still upload, but it does not fill the feed the way a vertical or near-vertical frame does. That weakens the first-second read on product, offer, and motion.
A cleaner workflow is to export Feed-first assets in 4:5, then derive square only when a tighter retargeting version is needed. If your editor is already cutting product demos or founder-led hooks for short-form, a workflow like smart video repurposing for Instagram can cut the time from master edit to placement-ready variants.
Practical rule: If your video starts in 16:9, assume it needs reframing before it is Feed-safe.
For teams running Advantage+ Shopping campaigns, the same asset discipline matters even more. Meta still decides where the creative lands, but automation cannot rescue a weak crop. The asset has to stay readable when the delivery system moves it around.
For a broader production library, we keep a separate asset system for social media marketing images, because crop logic is different from marketplace creative and the export standards are not interchangeable.
Carousel is still underused because it's often treated like a product gallery instead of a pacing tool. The format allows 2 to 10 cards, and Meta's current guidance supports 1:1 or 4:5 cards in Feed, with 1080 × 1080 px and 1080 × 1350 px as practical production targets Instagram carousel ad guidance. The same file-size rules still apply per card, so the creative budget has to survive every slide, not just the first one.
For direct response, 3 to 5 cards is usually enough to build a hook, proof, and offer sequence. For catalog storytelling, 7 to 10 cards gives you room to show variants, bundles, or problem-solution progression without cramming too much into one frame. A carousel only works if people keep swiping, so the structure has to earn the next card.
That's why the first card matters so much. If card one looks like a standard static post, swipe behavior usually suffers. If card one creates a clear reason to continue, the format starts doing real work.
The more recent mixed-aspect-ratio carousel behavior means you can't assume every card is treated identically across placements. That's useful for flexibility, but it also raises the odds that one card looks great while another gets masked or visually compressed. Build to a shared safe-zone system, then test the cards as a sequence.
Practical rule: Keep the same headline logic across the deck, but don't force the same composition onto every card.
For brands also selling through Walmart, the same creative principle applies even when the platform looks different. Product tiles and Sponsored placements reward clarity, not clutter, so the carousel habit of sequencing offers still helps when you adapt visuals for Walmart keyword variations and catalog-based media.
Stories and Reels are where Instagram has put the most pressure on vertical creative. Reels now takes a bigger share of ad impressions than it did before, which is why more accounts are forced to build around 9:16 instead of treating vertical video as a backup format Instagram ads statistics. The issue is not whether the asset is vertical. It is whether one master file can survive placement masking, safe-zone cropping, and automated delivery without breaking the message.
The standard production target for both placements is 1080 × 1920 px or equivalent high-resolution vertical assets, with image ads capped at 30 MB and video ads at 4 GB Instagram Stories and Reels specs. That spec is only the starting point. Full-screen mobile placements reward restraint, because the job is to protect the parts that convert, not to cram every message into the frame.
The practical safe-zone habit is simple. Keep key copy, logos, claims, and disclaimers out of the top and bottom UI areas, then place the message in the middle band where it is less likely to be covered by profile elements or the CTA. If you have ever watched a coupon code disappear under app chrome, you already know why the middle band matters.
The same rule applies when a creative is being adapted across vertical placements. Build the composition so the offer still reads when the app overlays, captions, and buttons land on top of it. If you are also adapting short-form assets for other platforms, the production logic is similar to YouTube Shorts ad creative, but Instagram usually gives you less room for sloppy bottom placement.
Reels behaves more like an attention environment than a standard ad slot. The opening frame has to earn the scroll, and the pacing has to feel native enough that it does not get dismissed on sight. Stories can be more direct and utility-heavy. Reels usually needs a stronger motion-first hook.
The upload standard from the vertical video guides is still 9:16 at about 1080 × 1920 px, and you can use the same master file for both placements if the safe zones are respected 9:16 video specs for Instagram. That is the part many teams get wrong. They make one edit for Stories, one for Reels, then wonder why the creative stack becomes slow and inconsistent. A better workflow is one vertical template with a centered offer block, a top buffer for account UI, and a bottom buffer for captions or the CTA button.
Reels also punishes weak pacing faster than Stories does. If the opening beat does not establish the offer, the proof, or the product use case immediately, the rest of the cut rarely gets a fair shot. Stories gives you a little more room to speak plainly. Reels wants the first visual to do real work.
Practical rule: Build the 9:16 master once, then version it for Stories and Reels rather than treating them like separate design jobs.
Teams skip these placements because the spec tables live deep in Meta's product docs. That leaves reach-efficient inventory on the table, especially for brands that need incremental volume without rebuilding the whole creative package. Explore, Collection, and Shopping all behave differently, and the wrong setup can make a strong asset look broken.
Explore ads show up in a discovery context, so the first seconds have to do more work than they would in Feed. The same asset logic still applies, with image inventory centered on 9:16 and video inventory using 4:5 according to current guides Instagram Explore specs. That makes Explore less about a unique format and more about placement context. If the opening frame depends on context the viewer does not have yet, the ad gets ignored.
In-stream behavior is similar in one important way. The viewer did not come to buy from an ad. They came to consume content, so the hook has to earn attention fast. If you reuse a Feed creative here, make sure the opening frame does not rely on the user already being primed.
Collection ads are different because they are not just single creative units. They rely on a cover plus product grid and are built for mobile shopping, with Instant Experience as part of the flow Collection ad guidance. That means the catalog structure matters as much as the visual. If the product tiles are mismatched, stale, or poorly cropped, the whole ad reads as careless.
Shopping ads are even more operational. They inherit catalog specs rather than behaving like a flat uploaded image, so the work sits in the feed structure and product data, not just the creative file. If the catalog is messy, the ad unit will be messy too. That is why catalog QA belongs in the same workflow as creative QA, not after it.
For operators working across channels, the same discipline shows up on Walmart. Product-led placements are unforgiving when the source structure is weak, so catalog health is part of the creative spec, not separate from it. The same principle is why our Amazon attributions work often overlaps with Meta planning, which you can see in how Facebook and Instagram drive Amazon sales.
What Clickstera Does Differently: in marketplace and Meta work, we do not separate creative from inventory health. Our audits look at whether the asset, the feed, and the placement economics all line up before more spend goes live.
Specs are only half the job. A file can hit the right dimensions and still get rejected or softened by bad export settings. The cleanest fix is a single export preset your editor uses every time, so nobody is guessing on codec, audio, or compression from one project to the next.
For Instagram video, the safest baseline is H.264 in MP4 or MOV, with AAC audio, 4:2:0 color, and a minimum canvas of 1080 × 1920 px for vertical or 1440 × 1880 px for 4:5 Instagram ad specs reference. Keep video files under 4 GB and image assets under 30 MB.
Bitrate is where a lot of teams drift. Push it too high and you ship a heavy file without getting better delivery. Push it too low and the ad starts to look soft, especially on text, skin, and product edges. A practical export range is usually enough to keep detail intact without making uploads slow or brittle.
Codec mismatch is a common failure point, especially when motion teams hand off high-end masters. ProRes 422 is usually the wrong upload format because it is heavier than Meta needs and can force extra re-encoding. H.264 with a sane profile is the cleaner route.
Audio causes problems too. Missing tracks, strange sample rates, or unusual container settings can all slow processing or create upload issues. If the creative uses music, export the exact mix you intend to deliver, because last-minute swaps often introduce avoidable mistakes.
Practical rule: For paid social, keep upload stability first, then compression efficiency second.
Here's the shortest version to hand to your editor, cut it once, then reuse it:
A lot of teams default to “support everything” because Meta will accept it. That usually burns time in production and leaves budget spread across formats that do not earn the same quality of attention.
The two formats that deserve the most weight in Instagram ad specs for 2026 are 4:5 Feed and 9:16 Reels. 4:5 takes up more vertical real estate than a square post, which helps on crowded feeds. Reels is where a larger share of impression volume is showing up, and the vertical frame matches how people already consume content on the app Instagram format guidance. Square still has a place, but it is rarely the first asset I would build when budget is constrained.
For prospecting, 4:5 Feed should usually be the starting point. It gives you a strong default for direct response without forcing a full redesign for every placement, and it keeps the creative legible in both feed and mixed delivery setups. Then add 9:16 Reels when you need scale and a format that supports motion-led hooks.
Square fits narrower jobs. Retargeting and catalog-style creative can still work well in it, especially when the product needs to stay clear and the message is already warmed up. Horizontal should be treated as a fallback, because it rarely owns the viewport the way taller formats do.
The decision is not which size Meta allows. It is which size earns the most useful attention with the least waste. That is the same practical logic behind placement-aware creative planning across feed, stories, reels, and explore.
Practical rule: Ship 4:5 first, add 9:16 for scale, keep square for tight retargeting or catalog reuse.
This is the table many teams screenshot and drop into the creative brief. It's not meant to replace testing. It's meant to stop avoidable production mistakes before they hit review.
| Format | Aspect Ratio | Recommended Size | Max File Size | Max Video Length | Codec |
|---|---|---|---|---|---|
| Feed Image | 1:1 | 1440 × 1440 px | 30 MB | N/A | JPG, PNG |
| Feed Video | 4:5 | 1440 × 1880 px | 4 GB | 1 second to 60 minutes | MP4, MOV |
| Carousel Feed | 1:1 or 4:5 | 1080 × 1080 px or 1080 × 1350 px | 30 MB image, 4 GB video | 1 second to 2 minutes | JPG, PNG, MP4, MOV |
| Stories Image | 9:16 | 1440 × 2560 px | 30 MB | N/A | JPG, PNG |
| Stories Video | 9:16 | 1080 × 1920 px or equivalent vertical | 4 GB | 1 second to 60 minutes | MP4, MOV |
| Reels Image | 9:16 | 1440 × 2560 px | 30 MB | N/A | JPG, PNG |
| Reels Video | 9:16 | 1440 × 2560 px | 4 GB | 0 to 15 minutes | MP4, MOV |
| Explore Image | 9:16 | 1080 × 1080 px | 30 MB | N/A | JPG, PNG |
| Explore Video | 4:5 | 1080 × 1080 px | 4 GB | 1 second to 60 minutes | MP4, MOV |
| Collection Feed | 1.91:1 to 1:1 | 1080 × 1080 px | 30 MB image, 4 GB video | N/A | JPG, PNG, MP4, MOV |
| Shopping | Catalog-driven | Catalog assets | Varies by catalog asset | Varies by format | Catalog-based |
Meta's commonly cited copy limits are 125 characters for primary text, 40 characters for headline, and 30 characters for description Meta copy limits reference. Those limits don't behave the same way in every placement, so character count should be planned per destination, not per ad object.
Most upload failures are boring, which is exactly why they waste so much time. The fix is usually mechanical, but if you diagnose the wrong problem, you'll keep re-exporting the same bad file.
Missing audio tracks create avoidable upload friction. Unsupported sample rates can do the same. If you're using copyrighted music in Reels, expect auto-mute behavior to change how the ad is delivered, so don't build a concept that depends on the track carrying the message.
The quickest pre-publish check is mechanical. Open the asset at full size, confirm that the CTA is visible, and verify that nothing important sits in the top or bottom overlay zones. Then upload the exact file you intend to serve, not a review copy.
Practical rule: If the asset only works when you zoom out mentally, it's not review-ready.
A spec choice affects auction performance because it changes how much of the frame the ad owns and how quickly the message lands. A 4:5 Feed unit usually helps thumb-stop behavior because it takes up more vertical space, while 9:16 Reels matters more for reach and attention in a full-screen environment. That trade-off is the part that matters if you care about profit, not just delivery.
If the goal is first-contact attention, use the format that claims the most screen real estate without breaking the message. If the goal is efficient scaled delivery, use the placement where people are already consuming vertical content. In practice, that usually means building at least two masters, because one asset rarely carries every placement well.
Marketplace operators should read Meta the same way. Walmart Sponsored Product placements and Amazon Sponsored Brands bring their own layout pressure, and the creative still has to fit the dominant inventory shape. Clickstera's Amazon attribution for Meta ads work reflects that overlap, because the media mix only makes sense when the creative and the marketplace path line up.
We score every asset on three things. Spec compliance, safe-zone integrity, and screen-coverage percentage. That gives a cleaner read on whether a creative should be scaled, iterated, or cut.
What Clickstera Does Differently: we use a 4-stage PPC audit framework across Meta and Walmart, so we look for budget leaks before we ask for more spend. In practice, creative format fit is reviewed alongside placement economics, not after the campaign has already burned budget.
It gets masked or cropped, which is exactly why the safe-zone layout matters. A vertical master can survive cross-placement delivery, but only if the headline, product, and CTA are centered in a way that still reads when the asset is compressed into a smaller frame.
Use one master with protected zones, then build placement-specific variants only where the message would break under cropping. Advantage+ will optimize delivery, but it won't rescue a layout that puts essential copy in the wrong part of the frame.
Not always. Reels ad policies can be stricter than organic posting, so licensed or copyrighted tracks may be muted or restricted. If the ad depends on audio for the hook, export a version that still works when sound is unavailable.
Sometimes, but not without adjustment. The visual logic can carry over, especially for product-led creative, but the dominant placement shape and text density rules are different. A good cross-channel asset is adaptable, not identical.
Want us to audit your Amazon ad account for free? Clickstera offers a no-obligation PPC audit where we identify your top 3 budget leaks within 48 hours. Book yours at Clickstera Solutions LLC.
Talk to Clickstera and get a clear next-step plan to scale your performance marketing.