How to Get Title Safe and Action Safe Areas Right

Title safe and action safe areas are the invisible margins inside a video frame that tell you where on-screen text and important action can live without being cropped by a display or buried under a player's controls. They sound like a technical footnote, yet they decide whether a viewer sees the whole of a presenter's name in a lower third, whether a price in a product ad survives the trip to a television, and whether the call to action at the end of a vertical video sits behind a like button. When they are ignored, nothing breaks in the edit suite. The problem only appears later, on someone else's screen, where it is expensive to fix.

This guide is for marketers, in-house video teams and business owners who commission motion graphics, and for the designers and editors who build them. It works through the misconceptions that most often cause text to be clipped or hidden, explains what the published standards actually say, and sets out the practice that keeps graphics readable across broadcast, web players and social feeds. The figures used here come from the European Broadcasting Union's recommendation on safe areas for 16:9 production; where a number depends on a specific platform or device, we say what drives it rather than inventing a figure.

The short version: safe areas are not a relic, they are not one box, and they do not work the same way in a 9:16 frame as in a 16:9 one. The longer version, with the reasoning and the fixes, follows.

What title safe and action safe areas measure

A safe area is a rectangle, centered in the frame, defined as a percentage of the picture's width and height. Anything inside it is expected to be visible on a correctly set-up display; anything outside it might not be. Two rectangles matter in practice. The larger one, usually called action safe, is where essential movement and picture content should stay: a product being demonstrated, a presenter's face, the ball in a sports replay. The smaller one, called title safe or graphics safe, is where text, logos and anything the viewer must read should stay.

The European Broadcasting Union's EBU R 95 recommendation on safe areas for 16:9 television production sets the action-safe area at 93% of the width and height of the picture and the graphics-safe area at 90% of the width and height. Because the percentage applies to each dimension, the margins are not equal in pixels: a 16:9 frame is wider than it is tall, so the side margins are larger than the top and bottom margins. SMPTE publishes a parallel standard for widescreen display, SMPTE ST 2046-1, Safe Areas for Widescreen Image Display, which is the reference North American broadcasters and many post houses work to. If you deliver to a broadcaster, their delivery specification is the document that governs, and it will usually point to one of these standards.

Turning the EBU percentages into pixels is simple arithmetic: subtract the percentage from 100, halve it, and multiply by the dimension. For a 1920 by 1080 frame, the graphics-safe margin is 5% of 1920 on each side, which is 96 pixels, and 5% of 1080 at top and bottom, which is 54 pixels. The action-safe margin is 3.5% on each side, which works out to 67.2 pixels horizontally and 37.8 pixels vertically; most people round those to 67 and 38.

Two things are worth noticing in that table. First, the margins scale with resolution, so a template built at 1080p and upscaled to UHD keeps its proportions; a template rebuilt at UHD by eye often does not. Second, the difference between the two boxes is small, under 30 pixels per side at 1080p. That thin band is where many problems start, because a designer who snaps a lower third to the action-safe line rather than the graphics-safe line has placed text in the one zone the standard reserves for picture, not words.

Safe areas also apply beyond broadcast. Web players draw scrub bars, titles and share buttons over the frame. Social platforms lay usernames, captions, and like, comment and share buttons over the bottom and right of vertical video. Presentation software, conference screens and signage players can scale or crop content. The underlying idea stays the same even where no formal standard exists: decide which parts of the frame might be lost or covered, and keep essential information out of them.

Myth one: flat-panel screens made safe areas obsolete

Safe areas were born in the era of cathode-ray tube televisions, which deliberately overscanned the picture, pushing its edges past the visible glass to hide the unstable borders of an analog signal. Because each set overscanned by a different amount, broadcasters needed a conservative inner zone where content was guaranteed to show. It is easy to conclude that, with digital signals and flat panels that can map pixels one to one, the problem has gone away.

Consider where a piece of video actually ends up. A television may still have an overscan or "zoom" picture mode enabled out of the box, or set by a previous owner, and viewers rarely change it. A hotel or venue screen fed through a set-top box or scaler may trim the edges. A web player shows a control bar across the bottom whenever the viewer moves the mouse or taps the screen. A social platform crops a 16:9 upload into a square or 4:5 preview in the feed. A presenter drops your video into a slide deck with a template border around it. None of these involves a CRT, and every one of them removes or covers the edge of the frame.

The practical consequence is that the safe area now does two jobs. It is still insurance against physical cropping, and it is also a layout rule that keeps text away from the parts of the frame most likely to be covered by an interface. The second job is arguably the more important one for most marketing video, which is watched far more often in players and feeds than through a broadcast chain.

What changes and what does not

What has genuinely changed is the tolerance for picture content near the edge. On a broadcast chain with modern displays, it is reasonable to let backgrounds, environment and camera movement use the whole frame, because losing a few percent at the edges no longer costs the viewer anything meaningful. What has not changed is the rule for information. If a viewer needs to read it, recognize it or act on it, it belongs inside the graphics-safe area, and on platforms with heavy overlays it belongs inside an even smaller zone that you define for that platform.

A useful way to explain this to stakeholders who want the logo "right in the corner, like a big brand" is to show them the same frame played in a browser with the controls visible and on a phone in the target app. The corner that looked clean in the edit suite is almost always occupied by something else.

Myth two: title safe and action safe are one margin with two names

Editors often switch on a single safe guide and treat it as the boundary for everything. Some software shows both rectangles by default and people learn to ignore the outer one; some templates are built to one line only. The result is either text sitting in the action-safe band, or picture content squeezed pointlessly into the graphics-safe box.

The reason for two lines is that the cost of losing something differs by content type. If the edge of a presenter's shoulder is cropped, the viewer loses nothing. If the first letter of their surname is cropped, the lower third is wrong. If a phone number loses its last digit, it is useless. Text has zero tolerance for partial loss, so it gets the more conservative margin. Action gets a little more room because the eye can infer what is happening from a partly cropped movement.

How to use both lines in practice

• Snap text to the inner line. Set up your templates so that the outer edge of every text box, logo and badge aligns to the graphics-safe boundary, not to the frame edge and not to action safe. For a 1080p frame that means a minimum 96-pixel inset from left and right and 54 pixels from top and bottom.
• Treat backing panels separately from their text. A lower-third bar can extend beyond the graphics-safe line, even to the frame edge, as long as the name and title inside it do not. A bar that bleeds off the left edge is a legitimate design choice; the text on it must still start inside the safe area.
• Keep the subject of the shot inside action safe. When you compose a demo, frame a product or animate a hero object, make sure the part that carries meaning stays inside the 93% box through the whole move, including the start and end of any camera push or pan.
• Do not use the outer line as a design grid. Aligning a column of text to action safe leaves it about 29 pixels closer to the edge at 1080p than the standard allows for graphics. That is small enough to look fine on your monitor and large enough to be clipped on a display that overscans.

If you tend to hand work between applications, check that each one uses the same definition. Editing and compositing tools let you change the percentages of their safe guides, and a team where one person's guides are set to one standard and another person's to a different one will produce graphics that shift slightly between versions. Agree the numbers once, write them into the project template, and keep them there.

Myth three: anything inside the safe area will be readable

Safe areas answer one question: will this be visible? They say nothing about whether it will be legible. A line of 16-pixel type sitting comfortably inside graphics safe can still be unreadable on a phone, and a thin white serif over a busy background can vanish even at full size on a television.

The trap is that safe placement and readable size pull in opposite directions near the edges. Once you have inset everything by 5% on each side, there is less room for text, and the temptation is to shrink it to fit a long job title or a legal line. On a large television at normal viewing distance, small type may still work; on a phone held at arm's length, the same text at the same proportion of the frame is tiny. The decisions that make text survive motion, from type size and weight to hold time, are covered in depth in our guide to text legibility in motion; the point here is that the safe area is where you start that thinking, not where you finish it.

Checks that go beyond placement

• Size for the smallest screen you deliver to. If the same master goes to broadcast and to phones, size type so that it is comfortable on the phone and still restrained on the television. That usually means fewer words per line rather than smaller words.
• Protect contrast near the edges. Graphics near the frame edge often sit over the busiest part of the image, such as foliage, crowds or set dressing. Use a backing panel, a soft shadow or a gradient scrim rather than relying on white text over whatever happens to be there.
• Allow reading time. A lower third that animates in and out over a short hold may be technically visible but practically unread. The more text, the longer the hold.
• Keep moving text inside the box for its whole path. Text that slides in from off-frame passes through the unsafe zone, which is fine. Text that comes to rest there, or that is still settling there during the key reading moment, is not.

Tracked graphics deserve special mention. When text is attached to a moving object, as in a callout that follows a product or a label pinned to a location, the text's position is controlled by the footage, not by your layout. If the tracked object drifts toward the edge of the frame, the label goes with it. Plan tracked labels with an offset toward the center, or clamp their position so they cannot leave the graphics-safe area; our article on motion tracking for graphics goes into the techniques.

Myth four: broadcast safe guides work for vertical social video

Social video breaks the symmetry that broadcast safe areas assume. A broadcast safe area is a centered rectangle with equal margins on opposite sides, because the risk of cropping is roughly the same all around the picture. On a vertical feed, the risk is lopsided: the platform's own interface covers large and specific parts of the frame, weighted heavily toward the bottom and the right.

On most vertical feeds, the lower part of the frame carries the account name, the post caption, music or audio credits and sometimes a call-to-action button, while a column of engagement buttons runs up the right side. The top may carry a search bar, tabs or a progress indicator. The exact dimensions differ by platform, differ between an organic post and a paid placement on the same platform, and change when the platform redesigns its app. That last point matters: a safe-zone template downloaded once and never checked again will slowly drift out of date.

Building a platform safe zone you can trust

The dependable approach is to build your own overlay for each platform you publish to, rather than relying on a number from memory. Take screenshots of the platform's interface on at least two phones with different screen shapes, in the placement you will actually use, and bring them into your motion software as a guide layer at the size of your composition. Mark the areas covered by interface elements, then draw your own safe box inside what remains. For organic and paid placements on the same platform, make separate overlays, because ads often add a button or label that organic posts do not have.

Once you have those overlays, you will usually find that the practical text zone on a vertical frame is a band in the upper-middle part of the picture, pulled away from the right edge. That affects composition, not just typography: a presenter framed with their face low in the frame will have their chin under the caption; a product shot placed in the lower third of the frame will sit behind the account name. Our guide to getting social media motion assets right covers the wider set of decisions for feeds, and the article on animating for multiple aspect ratios explains how to design one piece so that it adapts across 16:9, 1:1, 4:5 and 9:16 without being rebuilt from scratch.

End cards deserve their own overlay pass. The final seconds of a social ad are where the call to action lives, and they are also when some platforms surface their own buttons or replay prompts. A web address or offer that falls behind an interface element in the last frame is the single most expensive safe-area mistake in paid social, because it hides the part of the video that the whole budget exists to deliver. See our piece on end cards and calls to action for how to structure that final moment.

Myth five: captions will fit wherever there is space left

Captions and subtitles tend to be added late, often by a different person or tool, and often by the platform itself. That means the graphics designer rarely sees them during the build, and it is easy to design a frame that assumes the bottom of the picture is free for a lower third, a price badge or a logo.

There are three kinds of captions to plan for. Open captions are burned into the picture by your team, so you control their position. Closed captions are delivered as a separate file and rendered by the player, usually at the bottom of the picture, with limited or no control over where they appear. Auto-generated captions on social platforms are positioned by the platform, sometimes with an option for the uploader to move them. Only the first is fully in your control, and even then the house style for captions tends to claim the bottom of the frame.

The simplest rule is to treat the bottom of the graphics-safe area as reserved for captions whenever captions are likely, which, given how much video is watched with the sound off, is most of the time. Put lower thirds a little higher than tradition suggests, or on the left side with enough vertical clearance that a two-line caption can appear below without touching them. When a graphic must appear low in the frame, such as a legal line on an ad, plan for captions to be suppressed or repositioned during that moment, and check with whoever creates the caption file that this is possible for the delivery format.

Handling burned-in captions and on-screen text together

When your team burns captions in, design the caption style as part of the graphics package rather than as an afterthought. Fix the caption box's position, maximum line count and maximum line length, and then place every other graphic around that box. For vertical social, the caption zone usually has to move up to clear the platform's own interface at the bottom, which pushes it toward the middle of the frame and competes with faces and product shots. That is a design trade-off, not a technical detail, and it is better settled in the storyboard than in the last round of revisions.

A good sign that captions and graphics are fighting is a lower third that has been nudged upward in one version and downward in another to avoid them. If that happens across a series, the underlying template needs a reserved caption band, not more nudging.

Myth six: everything in the frame has to stay inside the safe area

Once teams learn about safe areas, some overcorrect. Every element is pulled inside the graphics-safe line: backgrounds are framed with visible borders, camera moves stop short of the edge, and the frame starts to look like a picture inside a picture. That is a different failure, and it wastes the parts of the screen that most viewers do see.

The logic of a safe area is that the edge of the frame is uncertain: it may or may not be visible. The right response to uncertainty is to put nothing essential there, not to put nothing there at all. A background that stops at the safe line will show a visible edge on every display that does not crop, which today is most of them. Build full-frame backgrounds, let set extensions and gradients reach the edge, and treat the area between the frame edge and the safe line as a place for atmosphere.

The same principle applies to reframing. When you animate for several aspect ratios from one master, it is useful to think of three nested zones: the full canvas, which must look complete; the frame for each crop, which varies; and a shared safe core where all essential content sits for every crop. Designing to that shared core lets one animation produce a 16:9, a square and a vertical version with the least rework. When graphics are composited into live-action footage, the full-frame rule matters even more, because a graphic that stops short of the edge while the footage continues will look pasted on; our guide to blending graphics with live footage covers that side of the craft.

Where the edge is part of the design

Some formats use the edge deliberately. Screen-recorded interface demos may show a full application window edge to edge; kinetic type sequences may push words off-frame as a transition; a brand system may use a bleeding shape along one side. These choices are fine as long as the information itself, the words the viewer must read, the button they must see in a demo, the logo they must recognize, comes to rest inside the safe area. For interface demos in particular, consider zooming into the relevant region rather than showing a full-screen capture where the key control sits at the edge; the same principles that govern good interface animation apply directly.

A worked example: one lower-third package for broadcast and social

The following is an illustrative example, not a client project. It shows how the numbers work when a single set of graphics has to travel to a 1080p broadcast deliverable and to a vertical social cut.

Imagine a brand producing a two-minute interview film. The master is 1920 by 1080 for a television spot and a website player. A 1080 by 1920 vertical version will be cut for social feeds. The graphics package includes a two-line lower third (name and role), a logo bug, burned-in captions for the social cut, and an end card with a web address.

Step one: the 16:9 master

Using EBU R 95 figures, the graphics-safe box at 1080p runs from x = 96 to x = 1824 and from y = 54 to y = 1026. The action-safe box runs from roughly x = 67 to x = 1853 and y = 38 to y = 1042. The designer sets both as guides in the composition template.

The lower third's backing bar is allowed to bleed from the left frame edge. The name text starts at x = 96, not at the frame edge and not at x = 67. The bar's bottom edge is placed well above y = 1026, leaving a band at the bottom of the graphics-safe area for closed captions the website player may display. The logo bug sits at the top right with its right edge at x = 1824 and its top edge at y = 54, which keeps it inside graphics safe while staying clear of the corner where some players put a menu or share icon. The end card's web address is centered, set large, and kept well inside the box.

Step two: the vertical cut

For 1080 by 1920, the same 5% arithmetic gives a graphics-safe box from x = 54 to x = 1026 and y = 96 to y = 1824. That is the starting point, not the answer. The designer imports screenshots of the target platform's interface, captured on two phones in the placement being bought, and draws the covered regions: a band at the bottom holding the account name and caption text, a column of buttons on the right, and a narrower band at the top. The remaining clear area is then the real safe zone for this platform.

In this illustrative case, the clear zone leaves the lower third nowhere near the bottom of the frame. It moves to the upper-middle portion of the picture, left-aligned at x = 54 or further in, and its width is limited by the right-hand button column rather than by the 5% margin. The burned-in captions move up above the platform's bottom band. The designer checks that the presenter's face, framed in the vertical crop, does not sit where the captions now are, and adjusts the crop to lift the eyeline. The logo bug moves to the top left, because the top right is close to interface elements on this platform. The end-card web address is repositioned into the clear zone and enlarged so it can be read on a phone.

Step three: verification

Exports are checked in three ways: in the software with guides on, on a television or large monitor with its picture mode set as a viewer's would likely be, and uploaded privately or previewed in the platform's own ad tool on at least two phones. Anything that touches an interface element or reads poorly at arm's length goes back for a fix. The time cost is modest, typically part of one review round, and it replaces the far more expensive discovery that the web address in a paid campaign was covered for its entire run.

Setting up safe guides in the software you already use

The first practical step is simple: turn on safe guides in your editing or motion software and keep them on. Every mainstream tool has them. In Adobe After Effects, title/action safe guides are available from the composition panel's grid and guide options. In Adobe Premiere Pro, safe margins can be displayed in the Program Monitor. DaVinci Resolve offers safe-area overlays in the viewer, and Final Cut Pro can show title and action safe zones in its viewer. The menu names change between versions, but the feature is there in each.

What matters more than where the switch is, is what the guide represents. Each application ships with default percentages, and those defaults do not necessarily match EBU R 95, SMPTE ST 2046-1 or your broadcaster's specification. Check the preferences and set the values to the standard you are delivering to. If you deliver to European broadcasters, set action safe to 93% and graphics safe to 90%. If a client's delivery specification gives different figures, those win, and they should be recorded in the project template so nobody has to remember them.

Building platform overlays as reusable assets

For social, go further than the built-in guides. Create a guide layer for each platform and placement at its native composition size, marked with the covered regions and your own clear zone, and store it in a shared library alongside your templates. Label each overlay with the date it was last checked against the live app. Set it as a guide layer (or its equivalent) so it never renders into the export by accident, a mistake that is easy to make with a semi-transparent image layer and awkward to explain to a client.

For work that also goes into slide decks, stream layouts, exhibition screens or signage, the same logic applies with different overlays. A presentation template may place a title bar over the video; a streaming layout will place a webcam frame and chat panel; a signage player may crop for a specific panel. Our article on digital signage animation covers what to ask about each environment before you design for it.

What to do instead: the safe-area routine for every project

Most safe-area problems are prevented by a short, consistent routine rather than by expertise. The list below turns the corrections from each myth into a set of checks that can sit in a project template or a review sign-off.

Two of these checks deserve emphasis because they are the ones most often skipped under deadline. The first is previewing on real devices. A composition viewed on a calibrated desktop monitor, at full resolution, with nothing overlaid, is the most forgiving possible view of the work. It is where a design is least likely to fail and so where failures are least likely to be noticed. The second is re-checking platform overlays. Platforms change their interfaces without notice to advertisers, and an overlay that was accurate at the start of a campaign can be wrong a few months later.

The routine also helps with approvals. When stakeholders review cuts in a browser, it is worth reviewing the social versions with the relevant overlay visible, so that feedback such as "make the logo bigger and move it to the corner" can be tested against the real interface in the meeting rather than after export.

When title safe and action safe areas need specialist help

Most teams can handle safe areas for a single destination with the routine above. The work gets harder, and the cost of mistakes rises, in a few specific situations.

• Simultaneous delivery to broadcast and social. A single graphics package that must satisfy a broadcaster's technical review and several social placements needs a template system designed for both, with a shared safe core and per-destination variants. Retrofitting a broadcast package for vertical feeds after the fact usually means redesigning most of it.
• Text that keeps getting hidden. If campaign after campaign comes back with covered calls to action or clipped lower thirds, the issue is usually the template and the review process rather than individual designers. A structured audit of templates, overlays and review steps fixes it at the source.
• High-stakes information on screen. Regulated claims, legal lines, pricing and technical animation, such as the kind covered in our guide to mechanism of action animation, often have to remain fully visible and readable for a required time. Losing part of them to a crop or an overlay is not just a design problem.
• Unusual displays. Exhibition installations, stream overlays, video walls and signage all crop, scale or overlay content in their own way, and each needs a specific safe-zone plan.

If one of those describes your project, our animation and motion graphics team builds graphics packages with safe-zone templates for each delivery destination and checks exports on real devices before they ship. You can also browse more practical guides in the motion graphics and animation articles.

Where this comes from

• European Broadcasting Union — EBU R 95 Safe areas for 16:9 television production
• SMPTE — ST 2046-1 Safe areas for widescreen image display

The figures and practices above come from the sources listed.

Working on something like this?

We take on Motion Graphics & Animation work for teams who want it done once, properly. Tell us what you are building and we will tell you honestly whether we are the right studio for it. Start a project.

Where to go next

• After Effects Expressions, Done Properly
• Keyframe Interpolation and the Graph Editor: What Actually Works