What you should actually use: end screens, cards, or both

If you publish on YouTube and want viewers to keep watching your channel instead of drifting off to the suggested sidebar, you have two built-in tools: end screens and cards. They look similar at a glance but solve different problems, and most creators under-use them simply because the rules feel fiddly.

Here is the short version:

  • Use cards to point viewers at related content during a video, when something you mention deserves a clickable link right then.
  • Use end screens to send viewers somewhere specific after the video, in the final stretch.
  • Use both when the video is long enough and the content warrants it. They are not interchangeable.

The rest of this guide walks through the rules that actually matter, the placement traps that trip creators up, the limits per video, and a sensible workflow for deciding what to add to new uploads, what to retrofit on older ones, and when an automated template beats a hand-placed layout.

The rules that change your decisions

Before deciding what goes where, internalise the constraints. They are not suggestions — they shape what is even possible.

End screen basics

According to YouTube’s own help documentation, end screens live in the final 5 to 20 seconds of a video. Your video has to be at least 25 seconds long for an end screen to be available at all, and the editor can grey out without explanation on shorter clips.

You can place up to four elements on an end screen for standard 16:9 videos, though other aspect ratios can have a lower limit. At least one of those elements must be a video or playlist — you cannot fill all four slots with subscribe buttons and links.

End screens are also suppressed during other interactive elements like card teasers and video watermarks, and they are not available on videos set as Made for Kids, in the YouTube music app, flash videos, mobile web, or 360 videos. That last point matters: setting a video as Made for Kids disables end screens permanently for that upload, so the designation has real downstream consequences.

Cards basics

Cards are different. They appear as small teasers at any point in the video — not just the end. According to creator guides such as LabelWorx’s documentation, up to five cards can be added to a single video, and cards can feature a video, playlist, channel, or approved link.

Cards also stop working on Made for Kids videos, so the same COPPA caveat applies.

The practical difference at a glance

Decision End screen Cards
When does it appear? Final 5–20 seconds Anywhere in the video
Maximum elements 4 (16:9) 5
Minimum video length 25 seconds None stated
Placement freedom Constrained to a safe zone Free, but YouTube suggests last 20%
Best for “What to watch next” “Here is the thing I just mentioned”

The safe zone matters more than most creators realise. End screen elements can get hidden by the video title overlay, the suggested sidebar, or mobile UI chrome. YouTube’s editor visually marks this, but the rule of thumb is to keep elements well inside the frame and away from the bottom-right corner where the suggested video panel sits.

A workflow for choosing what goes where

Rather than asking “should I add an end screen to this video,” ask three smaller questions in order.

1. Is the video long enough to need one?

Below about 25 seconds, you have no choice — the option simply isn’t there. Between 25 seconds and a minute, an end screen is technically available but rarely useful because the viewer barely has time to act on it. Most creators only start seeing meaningful lift once videos run two minutes or longer, where a meaningful share of viewers reach the final stretch.

If your video is under a minute, focus on cards instead. They can appear earlier and don’t compete with outro graphics.

2. What is the single next thing you want the viewer to do?

This is the most important question and the one most creators skip. End screens work best when there is one obvious next step. Common choices:

  • Watch the next video in a series. Pick a specific video in the end screen editor and link to it.
  • Subscribe. Use the subscribe element.
  • Land on a curated playlist. Useful for evergreen videos because the playlist stays relevant without re-editing.
  • Send traffic off-platform. Use the link element, but only if you are in the YouTube Partner Program and the link is approved.

If you cannot name the single most important next step, your end screen will likely be cluttered with three mediocre options instead of one strong one.

3. Should you use the “best for viewer” option?

The end screen video element has three modes: a specific video, your most recent upload, or best for viewer (YouTube picks based on the viewer’s history). For broad-appeal content where the logical follow-up varies by viewer, best for viewer often outperforms a manual pick. For series content with a clear sequence, a specific video still wins.

A reasonable default for many creators is one specific video, one best for viewer, one subscribe button — three elements, one slot spare, and a less cluttered screen.

End screen best practices that actually move the needle

Based on guidance from multiple creator resources, a few patterns show up consistently:

  • Keep it to two or three elements. Four-element layouts often look cluttered on mobile and split attention. The Artlist guide and AKNMIMM’s deep dive both flag this.
  • Place elements away from the bottom-right corner where the suggested video sidebar lives, and away from the very bottom edge on mobile.
  • Pair the end screen with a verbal call to action. Saying “watch this next” out loud in the last ten seconds measurably improves click-through versus a silent end screen.
  • Match the visual style. A plain end screen that fades into a clean graphic with the clickable elements on top reads better than leaving raw footage running under the tiles.
  • Use a playlist element when you don’t have a single obvious next video. Playlists keep viewers in your ecosystem longer and don’t need re-editing every upload.

Cards: when and where to place them

Cards are the under-used half of this toolkit. End screens get the attention because they look bigger, but cards let you link to context at the moment something relevant comes up in the video — and that moment is when viewers are most likely to click.

Best moments for a card

  • When you mention a tool, person, or video by name and want to give viewers a clickable link without breaking the experience.
  • At the natural transition between sub-topics, when a viewer might want to jump to a related piece.
  • Near the end (last 20%) as a soft fallback when viewers bail before the end screen.

Best practices

  • Don’t add more than you need. Three to five cards spread across a 10-minute video is plenty. More than that and the right-side panel becomes a wall of teasers.
  • Match the card timing to the script. A card about a tool you mentioned two minutes ago, fired at the five-minute mark, is more useful than a card fired at random.
  • Use the link card sparingly. Cards with external links require Partner Program approval and can look spammy if every video has one.

Should you retrofit end screens and cards on old videos?

This is the question the topic brief specifically calls out, and the answer depends on three filters.

Filter 1: Is the old video still getting views?

If a video is still pulling in search traffic or recommended sidebar impressions, it is worth spending ten minutes adding an end screen — especially one that points to a relevant specific video or playlist. If the video is dormant, the time is better spent on your next upload.

Filter 2: Is the video long enough?

Anything under 25 seconds cannot have an end screen at all. Videos between 30 seconds and a minute rarely benefit. Focus retrofit effort on videos over two minutes.

Filter 3: Has your channel evolved since the video was published?

If your current best content is significantly better than what you linked to originally, retrofitting with a specific video element pointing to the newer work can rescue old videos from sending viewers to weaker follow-ups. The “best for viewer” mode partially handles this automatically, but a manual pick gives you control over narrative flow.

A practical retrofit workflow:

  1. Pull your top 20 videos by views from the last 12 months.
  2. For each, check whether it has an end screen with at least one element.
  3. Add or update the end screen on the top performers you missed.
  4. Skip the bottom of the list unless you have spare time.

Manual placement vs. templates: when each makes sense

YouTube Studio offers pre-built end screen templates that group elements together. They are useful when:

  • You are processing many videos quickly and want a consistent baseline.
  • You don’t have a strong opinion about layout and just want something on screen.
  • You are retrofitting older videos at scale.

Manual placement is worth the extra minute when:

  • The video is part of a series and needs a specific next video pointer.
  • You want to test a layout hypothesis (for example, subscribe on the left vs subscribe on the right).
  • The visual composition of your outro demands custom positioning to avoid covering a face or product.

There is no real “automated tool” outside YouTube Studio that adds end screens for you at the API level — the feature is creator-side, not bulk-editable by third-party apps. Some scheduling and analytics tools surface end screen performance data, but the placement itself happens in YouTube Studio.

A/B testing end screens honestly

YouTube does not offer a built-in A/B testing tool for end screens in the way it does for thumbnails. What you can do is a manual rotation:

  1. Run variant A (for example, specific video + subscribe) for two to four weeks.
  2. Swap to variant B (for example, playlist + subscribe) for the same window.
  3. Compare end screen element click-through rates in YouTube Studio analytics.

Because end screen click-through rates are noisy and depend on traffic patterns, treat small differences as noise. A real signal usually takes a few hundred end screen impressions per variant to show up.

Frequently asked questions

How many cards can you add to a YouTube video?

Up to five cards per video, based on creator-side documentation. In practice, three to four well-placed cards outperform five random ones.

How many end screen elements can a YouTube video have?

Up to four elements for standard 16:9 videos. Other aspect ratios may have lower limits.

What are the end screen safe-zone dimensions?

YouTube’s editor visually marks the safe zone. The general rule is to keep elements inside the frame and away from the bottom-right corner (where the suggested video panel sits) and the very bottom edge (where mobile controls overlay content). Approximate tile sizes are 615×345 for video/playlist elements and 294×294 for subscribe, channel, and link elements.

Should I add end screens to old YouTube videos?

Yes — but only to the videos still getting meaningful views and that are at least 25 seconds long. Prioritise your top 20 by recent traffic and skip the long tail unless you have spare time.

Do end screens and cards work on Shorts?

The traditional end screen and card tools documented here are designed for long-form 16:9 videos. Shorts use different end-of-video mechanics (the loop, the swipe-to-next behaviour) and do not support the four-element end screen editor in the same way.

Do end screens affect watch time or revenue?

End screens can keep viewers on your channel longer, which is the metric YouTube rewards. They do not directly change CPM or ad rates, but the session-length effect is real and worth optimising for.

A repeatable workflow you can actually run

If you take one workflow away from this guide, make it this:

  1. On upload, in YouTube Studio, open the Editor and add an end screen with three elements: one specific video (the strongest next step), one best-for-viewer slot, and one subscribe button. Skip the fourth slot deliberately — less is more.
  2. Add two to four cards at the moments in the script where you mention the linked thing by name.
  3. Once a quarter, pull your top 20 performing older videos and retrofit any that are missing end screens. Skip the bottom of the list.
  4. Once a quarter, swap one element on a few videos (for example, change a specific video pointer to a playlist) and watch the click-through numbers for a few weeks.

That is enough process to capture most of the upside without turning end screens into a weekly chore. The goal is not perfection — it is a consistent, low-friction default that keeps viewers moving from one of your videos to the next.

Sources