After Effects Export Guide: ProRes, H.264, and Clean Deliverables

Why the export decides the quality of everything you made

An After Effects composition is not a video. It is a set of instructions that only becomes a video at the moment you render. Everything you saw in the preview window was an approximation: draft resolution, skipped frames, cached proxies, and a color pipeline tuned for your monitor rather than for your audience. The export is the first time the project is judged by the same rules your viewers will use.

That is why delivery problems so often feel sudden. A gradient that looked smooth on a calibrated display turns into visible banding after a platform re-encodes it. Audio that sounded perfectly in sync in the timeline drifts a few frames on a phone. A file that plays flawlessly on your machine gets rejected by an upload form that only accepts specific containers.

A useful way to think about export is to treat it as three separate decisions:

1. Mastering - creating a high-quality, self-contained file that represents the project as accurately as possible.
2. Delivery encoding - compressing that master to the size and format a destination expects.
3. Verification - checking the result on the device and platform where it will actually be seen.

Most people collapse all three into one step, exporting a compressed file directly from the render queue and hoping for the best. That works for casual projects, but it costs quality, time, and flexibility the moment a client, an editor, or a platform asks for a variant. Splitting the process costs a few extra minutes and pays back almost immediately.

Before touching any settings, answer three questions: Who plays this file? On what hardware? What happens to it next - will someone re-edit it, archive it, or will a platform transcode it again? The answers determine resolution, frame rate, codec, bitrate, and whether you need an alpha channel or separate audio stems. Export settings are not a matter of taste. They are a translation of those answers into technical parameters.

Composition settings that survive every delivery

Export quality is capped by what you build in the composition. No encoder setting can recover detail the composition never had, and no compression trick can fix a mismatched frame rate. Spending two minutes on composition setup prevents hours of troubleshooting later.

Resolution, frame rate, and safe areas

Work at the target resolution or higher, and let the encoder downscale rather than upscale. Upscaling a 1080p composition to a 4K delivery adds interpolation artifacts that no bitrate can hide.

Match the composition frame rate to the dominant source footage, and avoid mixing 23.976, 24, 25, 29.97, and 30 fps clips in a single timeline unless the project genuinely calls for it. Mixed rates force After Effects to retime footage, which produces stuttering on pans and motion graphics that were never designed for it. When you must mix, decide the delivery rate first and convert everything to it before you start animating.

Keep pixel aspect ratio square for anything destined for the web, and design with safe margins. Vertical crops, platform interface overlays, and television overscan all eat into the edges of the frame. If text or logos sit within five percent of the border, assume they will eventually be covered or cut.

Color management and bit depth

Set the project working space deliberately instead of accepting a default. For web and broadcast delivery, a Rec.709 pipeline is usually the right starting point. For material that will be graded or composited later, a scene-referred workflow such as ACES gives you far more headroom. What matters is consistency: a project that mixes untagged footage with a managed working space will shift colors unpredictably at export.

Bit depth matters as soon as you add gradients, glows, or subtle blurs. Working in 16-bit per channel and rendering to a 10-bit intermediate dramatically reduces banding. That extra depth is invisible in the preview and very visible after an aggressive platform encode.

Audio preparation

Standardize on 48 kHz and 24-bit for anything professional, even if the original music is 44.1 kHz. Mixed sample rates are one of the most common causes of small sync drifts that accumulate over a long timeline. Keep peaks below roughly -1 dBTP and target loudness appropriate to the platform. Streaming services typically normalize around -14 LUFS, while broadcast has its own stricter specification.

If anyone else will mix, finish, or localize the piece, render audio as separate stems rather than embedding a single stereo track. It is the difference between a quick revision and a full re-render.

Master files first: the two-stage render that saves hours

The most reliable export workflow has two stages: render a master, then encode deliverables from that master.

A master is a lightly compressed or lossless file that contains the finished picture and audio at full resolution. Once it exists, every other version - a 1080p H.264 for review, a vertical crop for social, a lightly compressed file for a client editor - is a fast transcode rather than a full render. When the client inevitably asks for the same thing but with a different end card, you only re-render the changed section, or re-render once and transcode again from the master.

Output module settings for a master

Choose ProRes 422 HQ for most HD and UHD work, ProRes 4444 when you need an alpha channel or intensive grading headroom, or DNxHR HQX if you are delivering into an Avid-centered pipeline. Set the output module to a QuickTime container, keep the audio embedded as uncompressed PCM at 48 kHz, and double-check that the color space tag on output matches your working space.

Naming and versioning

Consistent naming prevents the single most embarrassing delivery error: sending the wrong version. A simple pattern such as project_asset_v04_master.mov for the master and project_asset_v04_web_1080p.mp4 for the delivery copy removes ambiguity. Avoid spaces in filenames destined for web uploads, and never overwrite a previous master. Storage is cheaper than a re-render.

Codec decisions: ProRes, H.264, H.265, and image sequences

The ProRes family at a glance

Variant	Typical use	Notes
ProRes 422 Proxy	Offline editing, quick reviews	Smallest files, lowest quality
ProRes 422 LT	Fast turnaround delivery	Good balance for internal work
ProRes 422	General post-production	Common interchange default
ProRes 422 HQ	Mastering HD and UHD	Slightly larger, near-visually lossless
ProRes 4444	Alpha channels, heavy grading	Supports transparency
ProRes 4444 XQ	High-dynamic-range finishing	Largest files, highest headroom

H.264 and H.265 for delivery

H.264 remains the safest delivery codec because virtually every platform accepts it. Use two-pass variable bitrate when size matters less than quality, constant bitrate when uploading to a live streaming ingest, and a keyframe interval of about two seconds for platform compatibility. Keep dimensions divisible by two - ideally by sixteen - and check the level setting if older hardware must play the file.

H.265, also called HEVC, buys roughly the same quality at a lower bitrate, but acceptance is patchier, encoding is slower, and some older devices struggle to decode it. Use it when the platform explicitly supports it, or when you control playback on modern hardware.

Lossless and mezzanine alternatives

For visual effects handoff, image sequences such as PNG, TIFF, or EXR preserve maximum fidelity and allow frame-level recovery. Remember they carry no audio and produce enormous file counts. For very large projects, a lossless or near-lossless mezzanine codec keeps the workflow manageable while retaining quality, and it is far friendlier to shared storage than tens of thousands of individual frames.

Platform-specific delivery targets

Streaming platforms

Upload the highest-quality master you can afford to send. The platform re-encodes anyway, and a cleaner source survives that process better. Large-frame uploads at high bitrates frequently look better after encoding than smaller uploads, even when the underlying content was captured at a lower resolution, because the encoder has more data to work with. Match frame rate exactly to avoid duplicated or dropped frames.

Vertical social formats

Short-form vertical video rewards clarity over subtlety. Export at 1080x1920, keep text well inside the safe area, and raise the bitrate above what you would use for a landscape file of the same resolution, because fast cuts and motion graphics are harder to compress. Native uploads almost always outperform videos with visible letterboxing, so reframe the composition rather than cropping the finished master.

Broadcast and client review

Broadcast deliverables come with strict specifications: specific codecs, specific loudness targets, specific slate and countdown requirements. Client reviews, in contrast, benefit from small, fast-loading files with a burned-in timecode and version number. Use a low-bitrate H.264 for review and keep the master untouched.

Archive

Store three things: the project file with collected assets, the master render, and a delivery copy. Years later, the delivery copy shows what actually shipped, and the master gives you something to re-encode when formats change. Keep the archive on at least two drives, and verify the checksums of the most important master files once a year.

A repeatable export pipeline with Adobe Media Encoder

Media Encoder turns a one-off export into a queue. The practical workflow looks like this:

1. Render the master from After Effects using an output module, not a compressed delivery preset.
2. Add the master to the Media Encoder queue once, then duplicate the entry for each delivery format.
3. Apply saved presets: one for web 1080p, one for social vertical, one for review copies, one for audio-only.
4. Set the output destination using a consistent folder structure, ideally one folder per deliverable type.
5. Enable or disable parallel encoding based on stability. Heavy third-party plugins often prefer serial rendering.
6. Verify every file after encoding with a media inspection tool before sending anything.

Batching a series of deliverables

If you produce a recurring series - a weekly explainer, a set of product cutdowns, a seasonal campaign - save the entire set of encodes as a preset group or a watch folder. Drop the master into the watch folder and let the queue produce every version automatically while you move on to the next edit.

Automation options

Scripting can add items to the encoding queue without manual steps, and command-line tools such as ffmpeg are excellent for post-processing checks: verifying duration, confirming stream counts, or extracting a frame for a thumbnail. These small automations remove the most common human error in delivery, which is forgetting one variant.

Troubleshooting encoding errors, corrupted renders, and quality loss

Corrupted or incomplete renders

If a render stops partway through, the cause is usually resource-related: insufficient disk space, a full cache, a memory-hungry effect, or a corrupted source file. Empty the disk cache, render to a different drive, and try splitting the timeline into sections to isolate the failing frame. Rendering a short section as an image sequence is a useful diagnostic, because it tells you whether the problem is the content or the codec.

Codec-specific failures

Odd dimensions break H.264 encoding. Alpha channels fail silently in some players when exported as ProRes 4444. Unusual frame rates create playback issues on older hardware. When an encode fails, change one variable at a time - container, codec, dimensions, audio - so you learn what actually caused it instead of guessing.

Audio sync drift

Sync problems usually trace back to variable frame rate footage, mixed sample rates, or a mismatch between the audio device clock and the project sample rate. Conform audio to 48 kHz, force constant frame rate on any captured footage, and keep audio and video durations identical. A tiny offset at the start of a long timeline can grow into a visible problem by the end.

Color shifts and banding

If the exported file looks washed out, compare it in a player that respects color tags, not just the preview window inside your editing application. Tagging the output color space correctly, rendering to 10-bit, and avoiding two rounds of compression solve most cases. Adding a touch of dithering to smooth gradients also helps when a platform will compress the file aggressively.

File size and delivery friction

When a file is too large to send, never re-encode the master to shrink it. Create a small review copy instead and keep the master intact. Also check whether your transfer method is the real bottleneck: a fast upload link often solves a problem that people try to solve with destructive compression.

Pre-flight checklist and common mistakes

Before you render, confirm:

• Composition resolution and frame rate match the delivery target.
• Working color space and output color space are explicitly set.
• Audio is 48 kHz, correctly leveled, and in sync.
• Every source file is online and at full resolution, with proxies switched off.
• Output module settings have been reviewed rather than accepted by default.
• There is enough free disk space for the master plus all deliverables.

Common mistakes that cause real damage:

• Exporting from the preview at reduced resolution and forgetting to switch it back.
• Encoding H.264 twice, once from After Effects and again from Media Encoder, which compounds artifacts.
• Deleting the master after a successful upload.
• Assuming the platform accepts everything the local player does.
• Skipping verification, then discovering the problem after delivery.

FAQ

Should I export directly from After Effects or through Media Encoder?
Render a master from After Effects, then encode all delivery versions in Media Encoder. The master protects your quality and makes future variants nearly free to produce.

ProRes or H.264 for a streaming platform upload?
Upload the highest-quality version the platform accepts, often ProRes or a high-bitrate H.264, and let it handle the final encode. Low-bitrate sources degrade noticeably after re-encoding.

Why is my exported file so much larger than expected?
ProRes files are large by design because they preserve quality. If size matters, encode a separate H.264 delivery copy rather than changing the master.

Why does my video look different after uploading?
Either the color space was not tagged correctly, or the platform re-encoded with different gamma handling. Export a short test clip and compare it before uploading a long piece.

Can I export with a transparent background?
Yes, with a codec that supports alpha, such as ProRes 4444 or an image sequence. H.264 does not carry an alpha channel.

Why does rendering fail only near the end?
It is often a disk space or memory issue triggered by the largest, most complex section. Render in sections and check available storage before starting.

How do I export several aspect ratios at once?
Build each version as its own composition, then queue them all in a single batch. Reframing manually beats cropping the master, especially for vertical formats where the subject often sits off center.

What bitrate should I use for vertical video?
Higher than you would use for landscape at the same resolution. Fast cuts and motion graphics need the headroom, and platforms reward cleaner sources with fewer visible artifacts.

Treat delivery as part of the edit

Export is not the paperwork at the end of a project. It is the last creative decision you make. A carefully built composition can still arrive looking flat, soft, or out of sync if the final encode ignores the destination. Build a master, define your delivery targets, keep a small preset library, and verify what you send. Those four habits eliminate almost every delivery problem, and they leave your energy for the part that actually matters: making something worth watching.