If you already run Meta spend, "what pixel size?" is the wrong first question. The real one is which single export covers the most placements without getting cropped, letterboxed, or down-ranked. Get that right and your creative reaches the cheapest inventory; get it wrong and you pay more for every order without ever seeing why.
This guide gives you the exact numbers by placement, then does the part every spec post skips: the profit math.
Facebook ad video pixel sizes by placement (2026)
Here are the recommended export dimensions for the placements that matter to a store running conversion campaigns.
| Placement | Aspect ratio | Recommended pixels |
|---|---|---|
| Feed — vertical | 4:5 | 1440 × 1800 |
| Feed — square | 1:1 | 1440 × 1440 |
| Stories | 9:16 | 1440 × 2560 |
| Reels | 9:16 | 1440 × 2560 |
Dimensions above are from Shopify's 2026 Facebook ad-size guide and Buffer's 2026 Facebook ad specs, which list the same 1440 × 1440 (1:1) and 1440 × 1800 (4:5) feed figures.
Feed video: go vertical, not square
The 4:5 vertical frame is the one to lead with. Shopify notes the 1440 × 1800 pixel 4:5 ratio "takes up more vertical space on mobile screens than square or landscape formats," which means more of the scroll belongs to you, per Shopify's guide.
Square (1:1) still works and is a safe fallback, but it surrenders screen height on exactly the devices where most of your buyers are scrolling.
Stories and Reels: the full-screen 9:16 master
Stories and Reels both want the tall 9:16 frame at 1440 × 2560 pixels, according to Shopify's guide. This is the single highest-value export you can make, because a clean 9:16 file is also what feeds vertical feed placements with minimal cropping.
Keep your logo, text, and price out of the top and bottom edges. The UI overlays (profile, caption, call-to-action button) eat into those zones, and a headline that sits under a button is a headline nobody reads.
In-stream and landscape
For in-stream video that plays inside longer content, a 16:9 landscape or 1:1 square frame fits best. If you only build one horizontal cut, 1:1 is the more flexible choice because it also survives in the feed.
Resolution, file size, and format rules that apply everywhere
A few technical limits hold across every placement. These are the guardrails that decide whether Meta accepts and serves your file cleanly.
- Format: MP4, MOV, or GIF, per Buffer's specs.
- Max file size: 4 GB, per Shopify's guide.
- Minimum dimensions: 120 pixels wide and 120 pixels tall is the technical floor, per Buffer — far below what you should ever ship.
- Aspect ratio tolerance: 3%, meaning a frame slightly off from the target ratio still serves, per Buffer.
- Encoding: H.264 compression, square pixels, fixed frame rate, progressive scan, and 128 kbps+ stereo AAC audio, per Buffer.
The "minimum" numbers are a trap. Meta will accept a tiny, low-resolution file, but it will look soft on a modern phone and read as low-effort. Export at 1440 pixels on the short edge or higher and let Meta downscale.
On copy, keep it tight: Buffer recommends roughly 50 to 150 characters of primary text and a headline near 27 characters, per Buffer's specs. Anything longer gets truncated behind a "See more."
Why the wrong pixel size quietly costs you money
Pixel size is not a design detail — it's a distribution lever. When you upload a square or landscape video, Meta can still place it in vertical Reels and Stories, but it letterboxes your frame with bars or crops it awkwardly.
That matters because of where the cheap inventory lives. Vertical, full-screen placements are often the lowest-cost impressions in the auction, and a video that looks broken there earns a lower click-through rate, which the auction reads as a weaker ad and prices accordingly.
So the wrong export doesn't just look off. It can push you out of your cheapest placements and raise your effective cost per order — a cost that never shows up as a line item, only as a worse blended number at the end of the month.
A worked example: what a cropped placement does to per-order profit
Say you sell a $31 tee. Your Printify cost is $12.40, and Shopify plus payment processing takes about $1.20 on that order. That leaves roughly $17.40 before ad spend.
Now say you run $2,800 in Meta spend this month and it drives 340 orders. That's $2,800 ÷ 340 = $8.24 of ad cost per order, leaving about $17.40 − $8.24 = $9.16 in profit per order.
Here's the hidden tax. If a broken aspect ratio quietly knocks you out of your cheapest Reels and Stories inventory, your blended cost per order drifts from $8.24 toward, say, $10.50. That's $10.50 − $8.24 = $2.26 lost on every one of 340 orders, or about $768 of profit gone in a month — and nothing in Ads Manager labels it "wrong pixel size." It just looks like Meta "got more expensive."
That is the gap between what a spec guide tells you and what actually hits your bank account. The dimensions are easy to look up; connecting them to per-order profit is the part that takes real data from your store, your ad account, and your supplier all in one place.
How Victor keeps your Meta video ads tuned
PodVector AI is built for exactly that gap. Victor is an AI employee who connects to your Shopify store, Meta Ads, and your Printify, Printful, or Gelato supplier, so he sees every order, every ad dollar, and every product cost together.
Victor computes the true per-order profit on each product and campaign, then acts on it inside Meta Ads. He pauses the ad set bleeding money on expensive placements, shifts budget toward the creative and audiences that clear your margin, and edits the ads and audiences so your spend concentrates where orders are actually profitable.
Because Victor knows your Printify, Printful, and Gelato cost before the first sale, he isn't guessing at margin from revenue alone. When your 4:5 vertical cut outperforms the square one on real profit — not just reported ROAS — he's the one who catches it and moves the money. If your attribution numbers disagree across tools, start with our attribution and tracking guide, and tighten the UTM layer with our walkthrough on Klaviyo UTM tracking.
You can start with PodVector AI here.
FAQs
What is the best Facebook ad video pixel size in 2026?
For most stores, export a single vertical master at 1440 × 2560 pixels (9:16) for Stories and Reels, and a 1440 × 1800 pixel (4:5) cut for the feed, per Shopify's guide. The 4:5 vertical frame claims the most mobile screen space, which is why it tends to outperform square in the feed, per Shopify.
What aspect ratio should a Facebook video ad use?
It depends on placement: 4:5 or 1:1 for the feed, and 9:16 for Stories and Reels, per Buffer's 2026 specs and Shopify. If you build only one file, a tall 9:16 export reaches the most placements with the least cropping.
Can I use one video for feed, Stories, and Reels?
Mostly, yes. A clean 9:16 file serves across placements, but keep your text and key visuals in the center "safe zone" so UI overlays on Stories and Reels don't cover them. The downside of relying on one file is cropping — a dedicated 4:5 feed cut usually looks tighter where buyers spend most of their scroll.
What is the maximum file size for a Facebook video ad?
4 GB, per Shopify's guide. You will almost never hit that ceiling with a short, high-quality MP4 — if your file is that large, re-export with H.264 compression at a sane bitrate rather than uploading raw footage.
Does a higher-resolution video actually get better results?
Resolution alone doesn't win the auction, but a soft, low-resolution file reads as low-effort and tends to earn a weaker click-through rate, which raises your cost. Export at 1440 pixels on the short edge or higher and let Meta downscale, rather than shipping near the 120-pixel technical floor noted by Buffer.
How do I know if the wrong size is costing me money?
Watch your blended cost per order, not just the per-placement numbers in Ads Manager. If cost per order drifts up while your creative and audiences haven't changed, a cropped or letterboxed file losing cheap placements is a common culprit — and it's the kind of profit leak Victor is built to catch by reading your ad spend and true product costs together.