YouTube End Screens and Cards Guide for US Creators (2026)

End screens are the clickable elements that appear in the last 5 to 20 seconds of a video, sending viewers to another video, a playlist, your channel, or a subscribe prompt. Cards are smaller, less intrusive prompts that pop up during the video — usually a small "i" icon in the top right that expands into a thumbnail link. Both are layered on top of your normal frame, so they don't replace your content; they sit over it. The mental model I use: cards are mid-video reminders, end screens are end-of-video CTAs. When I led ad spend at my old company we used to say "every screen needs a next click," and that logic transfers cleanly to YouTube.

If a viewer watches your entire video and the screen just goes dark, you've trained the algorithm that you don't extend session time. Session time — how long a viewer stays on YouTube after watching your video — is one of the inputs the algorithm uses to decide whether to recommend you more aggressively. End screens and cards are how you nudge that number up. Every click into a second video, then a third, builds session length, and the algorithm rewards it more than it rewards per-video retention in 2026. For a deeper dive on the signals the algorithm cares about, see the YouTube algorithm explained, and the official rules live in YouTube's own help docs.

You add end screens in YouTube Studio after upload, in the Editor tab on the left side of the video edit page. You get up to 4 elements per end screen, and each can be a video, a playlist, a subscribe button, or a channel link. Placement is drag-and-drop on a grid overlaid on your last frame. The minimum length is 5 seconds, the maximum is 20 seconds, and you can stagger when each element appears so the screen builds rather than dumping everything at once. End screens also require a video at least 25 seconds long; YouTube will gray out the option in the Editor tab if a clip isn't eligible. They work on standard uploads, livestreams, and premieres, and Made for Kids videos can use them with restrictions on certain element types.

One trick that consistently improves clicks: stagger your elements by 2 to 3 seconds so the eye has time to land on each one. Dropping all 4 at once creates visual chaos that viewers tune out. Leave room in your edit for the end screen too — design the last 15 to 20 seconds with it in mind, not as an afterthought. A clean, slightly darker outro plate with framed slots where elements will land reads as intentional; a panicked end screen jammed onto your closing frame reads as amateur. A slightly darkened or blurred version of your last frame, with clean rectangles or circles where elements will sit, gives the eye somewhere to land. Some creators add a subtle outro animation that builds the layout, which signals "this is the end screen" and lifts click engagement. For thumbnail and end screen design as a system, YouTube thumbnail tips covers visual consistency.

Cards are the quiet workhorse most creators skip. You can add up to 5 cards per video, scheduled at any timestamp, each linking to another video, a playlist, your channel, or an approved external link if you're in the YouTube Partner Program. Think of them as contextual links, not interruptions. The best placement is the moment a viewer might want more on a topic you're glossing over. If you're making "How to Start a YouTube Channel" and you say "you'll also want to think about thumbnails," that's the moment to surface a card to your thumbnail tutorial. The viewer either ignores it — no harm, cards are subtle — or clicks, which YouTube reads as positive engagement. Click-through on well-placed cards typically runs 1 to 5 percent in my experience, which sounds small until you remember it compounds across every video. The biggest mistake is using cards as generic "check out this other video" plugs at random points. Be specific. Treat them like footnotes in an academic paper — they add depth where the reader is already curious.

YouTube allows up to 5 cards per video, but I rarely use more than 2 or 3. Cards work because they're contextual, and most videos don't have 5 natural moments where a card would help. Two well-placed cards typically outperform five randomly placed ones. The exception is long-form content over 20 minutes, where 4 to 5 cards spread across the runtime can each serve a different purpose without crowding. Quality and timing beat quantity. On external links specifically: only Partner Program members can add them, the domain must be verified in YouTube Studio, and even then external link cards click through worse than internal video links because they pull viewers off the platform — which YouTube quietly de-prioritizes. For most beginners, reserve external links for descriptions and pinned comments, not mid-video card slots.

After watching hundreds of channels, the templates that outperform fall into three patterns. Pattern one, "Next Video + Subscribe": two elements only — the most relevant next video on the left, a subscribe button on the right. Clean, focused, one decision; best for tutorial channels where the video is part of a clear sequence. Pattern two, "Best For You + Latest + Subscribe": three elements — a video YouTube auto-recommends based on viewer history, your most recent upload, and a subscribe prompt. That auto-recommend slot is YouTube's algorithmic best guess for that specific viewer and often beats anything you'd hand-pick. Pattern three, "Playlist + Subscribe": one playlist element plus a subscribe button, ideal when you have a binge-watch series. Playlists are underrated because they auto-queue the next video, so one click can lead to 30+ minutes of watch time. Avoid the four-element "throw everything at the wall" template — it confuses viewers and lowers click-through across the board. Less is more. For building binge-able sequences, how to start a YouTube channel covers the planning side.

A consistent end screen layout helps viewers learn what to expect, which improves click rates over time. The trap is thinking consistency means pointing to the same videos forever — it doesn't. Design one template you reuse channel-wide (same element positions, same closing plate, same voiceover pattern), then swap in different target videos for every upload. Consistency in design, variety in destination. The 'best for viewer' element personalizes per viewer — a beginner gets another beginner tutorial, an advanced viewer gets advanced content — so it often beats hand-picked slots. Hand-picked wins when you have a clear sequence (part 2 follows part 1) or want to push a specific new video. Many creators mix both: one 'best for viewer' slot plus one hand-picked slot for the latest upload.

Here's the trap nobody warns you about: end screens count as part of your video for retention. If your video is 10 minutes and the end screen runs the last 20 seconds, viewers who drop off during it still register as a retention dip. The fix is to make sure those seconds aren't dead air. Keep your voiceover going — recap the takeaway, tease the next video by name, and explicitly invite the click: "If you found this useful, the next thing I'd watch is the video on the right where I break down [topic] step by step." I've seen creators boost average view duration by 8 to 15 percent just by recording a tighter, more directive end-screen voiceover. Avoid abrupt music cuts and dead pauses. And never end screen on top of important visual information — if you're still showing a critical chart or product shot, transition to a calmer plate first. Done well, end screens often increase total session time even if they slightly dip per-video retention. YouTube SEO for beginners has more on retention metrics.

Shorts don't get traditional end screens — the format is too short and the player too different for the overlay system. What you get is the auto-loop and the channel handle at the bottom. The closest analog to an end screen is the last 1 to 2 seconds of the Short, where you verbally or textually direct viewers to a longer video. The pattern that works: deliver a complete, satisfying punchline in 30 to 50 seconds, and have the last 1-2 seconds say "Full breakdown on the channel." Then drop a direct link in the description and pinned comment. Pinned comments on Shorts get unusually high read rates because viewers scroll slowly while the loop plays. Don't treat Shorts as an afterthought — but don't bolt traditional end-screen logic onto them either. Different format, different funnel. For how Shorts monetize and feed long-form, see YouTube Shorts monetization.

In rough order of frequency: One — end screens promoting videos older than a year that no longer represent your best work; they compound, so a stale video drags your perceived quality down for new viewers. Cycle them quarterly. Two — subscribe button as the only element; subscribing is friction, clicking another video is easier and more rewarding, so always pair the subscribe button with at least one video element. Three — cards at random timestamps with no contextual relevance; throwing them at minute 3, 5, and 7 with no anchor to what's on screen wastes the slots. Four — end screens layered over busy footage, making both the overlay and the underlying video harder to read; design a clean closing plate. Five — forgetting to update end screens when you publish new videos. Your back catalog is evergreen real estate: viewers find old videos through search and recommendations constantly, and for a creator earning from home every refresh is essentially free extra income. Block out an hour every quarter to refresh end screens on your top 20 viewed videos so they point to current best work — one of the highest-leverage maintenance tasks on a channel.

Every percentage point of click-through compounds across every future upload, which is why this tiny corner of YouTube Studio matters disproportionately for a one-person operation. Studio gives per-element data in the Reach tab under End screens and Cards. The numbers to watch: end screen element click rate (typically 5 to 15 percent for well-placed elements), card click rate (typically 1 to 5 percent), and impression-to-click ratio across both. If one element consistently outperforms, double down and make that pattern your default for similar videos. If your overall end screen click rate is under 3 percent, you have a copy or design problem, not a placement problem. Also track how end screens affect session start position — clicks into a second and third video build session length, which the algorithm rewards. Test on a phone before publishing, because mobile is the majority of US watch time and tap targets are tighter there; avoid tiny text, busy backgrounds, and overlapping elements. Some creators keep a simple spreadsheet of end screen variant and click rate by video, refining monthly. That operator discipline is what separates channels that grow from channels that plateau. For the analytics that actually matter, YouTube analytics explained walks through every report worth opening.