Text Legibility in Motion: The Decisions That Matter

Text legibility in motion is the craft of keeping on-screen words readable while they move, appear, sit over changing footage and then disappear. It covers every piece of type in a video: headline supers, lower thirds, price callouts, product claims, legal lines, kinetic typography, data labels and the end card with the URL. A frame can contain text that is technically present and still fail, because the viewer did not have enough time, enough contrast, enough size or enough stillness to take it in.

It matters to anyone whose message depends on words on screen. Marketers running sound-off social ads rely on text to carry the offer. In-house teams producing explainers, event openers and training videos use type to anchor key ideas. Agencies and editors inherit templates and footage they did not design and have to make the words work anyway. When text is unreadable, the cost is not aesthetic: the offer, the date, the claim or the call to action simply does not reach the viewer, and the viewer is left mildly frustrated with the brand.

This field guide moves from fundamentals to advanced practice. It starts with the four forces that break legibility in video, then works through reading time, type choice, contrast, size and placement, animation, and delivery compression, before a worked example, special cases and a review process you can adopt as a house standard.

Why Text Breaks Down in Motion: The Four Forces

Print designers control one variable at a time: the page does not move, the background does not change and the reader decides how long to look. Motion removes all three controls. Understanding the forces that act on type in video makes every later decision easier, because each technique in this guide exists to counter one of them.

Time

The viewer does not choose how long text stays on screen; the editor does. Reading requires the eye to fixate on words in sequence, and a line of text needs a certain number of fixations no matter how bold or beautiful it is. If the text leaves before those fixations are complete, the message is lost. Time is the most common failure and the cheapest to fix, which is why it gets its own section below.

Background

Over live footage, the background behind a letter can shift from sky to shadow to a white shirt within a second. A white headline that reads perfectly against a dark wall can vanish when the camera pans across a window. Unlike print, you cannot check contrast once; you have to check it across the whole duration of the text.

Movement

Moving text forces the eye to track and read at the same time. Tracking is something the visual system does well; reading while tracking is not. Text that drifts slowly can still be read, but text that is scaling, rotating, blurring or bouncing while it carries information asks the viewer to do two jobs at once, and most viewers will do neither well.

Degradation

Between your export and the viewer's screen, the video is usually re-encoded by the platform, scaled to fit a phone, and viewed in varying light. Each step removes fine detail. Hairline serifs, thin strokes, tight tracking and small colored text are the first casualties. What looked crisp in your project file can arrive soft, blocky or smeared.

Every decision that follows maps onto one of these forces: hold durations counter time, scrims and shadows counter background, disciplined animation counters movement, and robust type choices counter degradation.

Reading Time: How Long On-Screen Text Needs to Stay

Viewers need time proportional to the text length. That sounds obvious, yet the most common legibility complaint in review sessions is still "I couldn't read that." The reason is that the person who wrote the text already knows what it says; they recognize it rather than read it, so it always feels long enough to them.

The read-it-twice rule

The most reliable working standard is to hold text long enough for a viewer to read it twice at a relaxed pace. The first pass decodes the words; the second pass lets them register, and it covers viewers who looked at the screen a beat late. When a reviewer who has never seen the script can read a super twice before it leaves, the timing is right.

To turn that into numbers at the planning stage, many teams use a simple estimate: count the words, assume a comfortable reading pace of roughly two to three words per second for display text in video, double it for the second read, and add around half a second for the eye to find the text after it appears. Treat this as a starting point to be checked with fresh eyes, not a standard. Unfamiliar words, numbers, product names, URLs and non-native readers all need more time than the estimate suggests.

What counts as a word

Numbers and codes are expensive. A price like $1,249.99, a date range, a phone number or a promo code each takes longer to read than an ordinary word, because the viewer has to process each part. Count them as two or three words each when estimating. URLs are worse: a viewer who might want to type one needs time to memorize it, which is closer to a full second per segment.

Budgeting time in a script

The practical consequence is that on-screen text competes for seconds with everything else in the edit. In a 15-second spot, a 12-word super held for the read-twice duration can consume well over half the runtime. That is a scripting problem, not an animation problem, and it should be solved before anyone opens a timeline. Cut the words, split one super into two sequential beats, or move secondary information to the caption or the landing page.

Typeface, Weight and Spacing Choices That Survive Motion

Typeface selection for video follows different priorities than typeface selection for print or even for the web. The type has to survive scaling, motion blur, compression and a small screen, often all at once. Brand guidelines written for print do not always anticipate this, so part of the job is translating a brand's type system into a version that works on screen.

Weight

Thin type degrades after platform encoding, so use weights that survive compression. Light and thin cuts that look elegant on a poster tend to break up in video: strokes a pixel or two wide get averaged into the background by the encoder, and the letterforms lose their edges. As a working habit, set body-length supers in regular or medium weights and headlines in semibold or bold. If the brand insists on a light weight, reserve it for large display words that are held still over a clean background.

Letterform structure

Sans-serif faces with open apertures, generous x-heights and clearly distinguished characters are the safest choice for text that must be read quickly. Watch for confusable characters: capital I, lowercase l and the numeral 1; the letter O and zero; rn and m. These matter most in codes, URLs and prices, exactly where errors are costly. Condensed faces save horizontal space but reduce the counters inside letters, which fill in quickly when the image softens.

Tracking and leading

Slightly looser tracking than you would use in print helps letters stay separate after compression and scaling. Tight tracking causes adjacent strokes to merge into a smear when the encoder blurs detail. Line spacing should be generous enough that ascenders and descenders from adjacent lines do not touch, especially when a drop shadow is added, because shadows effectively thicken the letters.

Case and line length

All-caps lines read more slowly than sentence case because the word shapes are uniform rectangles. Use caps for short labels of one to three words, and sentence case for anything longer. Keep lines short: two lines of four to six words each are much easier to take in at a glance than one long line running across a 16:9 frame, and they also fit vertical formats without re-breaking.

Contrast Over Footage: Scrims, Shadows and Plates

Text over footage needs contrast support. The web accessibility community has a well-defined baseline for contrast, and it is a useful floor for video too: the W3C guidance on minimum contrast for accessible text sets a ratio of at least 4.5:1 for normal text and 3:1 for large text. Video has no formal equivalent for most marketing work, but borrowing these numbers gives teams an objective way to settle arguments about whether a super is readable enough.

The difficulty is that footage has no single background color. A contrast check on one frame tells you little about the next. The goal is therefore to create a stable local background behind the text, so that the contrast ratio holds regardless of what the footage does.

The techniques, from lightest to heaviest

• Soft drop shadow. A diffuse shadow a few pixels wide, at partial opacity, separates light text from mid-tone backgrounds without looking like an effect. It fails over very bright or very busy areas.
• Glow or outer stroke. A tight dark halo around light letters keeps edges defined over busy texture. Keep it subtle; heavy strokes look dated and thicken the type.
• Gradient scrim. A soft darkening gradient across the lower or upper portion of the frame, sitting between footage and text. This is the workhorse for lower thirds and social captions because it is almost invisible as a design element while guaranteeing contrast.
• Solid or translucent plate. A box or bar behind the text. It guarantees contrast completely and becomes part of the graphic language, which suits news, data and brand systems.
• Localized grade. Darkening or defocusing the footage in the region behind the text, which keeps the image natural while quieting it. Depth-of-field treatment is a subtle version of this.

Choosing among them

Decide by how much the background varies across the hold, not by what looks nicest on the first frame. Over a locked-off interview with a plain wall, a soft shadow may be enough. Over handheld street footage with bright signage, you need a scrim or a plate. Color choices matter too: saturated brand colors often have lower luminance contrast than they appear, and the guide to color in motion design explains how to check luminance rather than hue when pairing type and backgrounds. Where the aesthetic calls for softening the footage behind a headline, shallow depth-of-field techniques can quiet the background without a visible box.

Size, Placement and the Phone Screen

Most social video is watched on phones, and a great deal of it is watched in feeds where the video occupies only part of the screen. Judging legibility on a large monitor only is one of the classic mistakes: a 27-inch display at arm's length makes text look several times larger than it will appear on a handset, so everything feels readable in the edit suite and nothing does on the train.

Sizing text for the smallest screen it will reach

Instead of setting a point size, think in terms of text height as a fraction of frame height. A useful habit is to define minimum sizes per format in your templates and test them on a real phone: body supers large enough to read comfortably at arm's length in a feed, with headline text substantially larger. Exact values depend on the typeface's x-height and weight, so the test matters more than the number. Once you find sizes that work for your brand face, write them into the template so every editor inherits them.

Safe areas and platform overlays

Broadcast has long used title-safe margins to keep text away from frame edges. Social platforms add a second layer: interface overlays for usernames, captions, buttons and progress bars that cover parts of the frame, particularly along the bottom and right edge of vertical video. These overlays change as platforms update their apps, so check each platform's current guidance before a campaign, and build a safe-zone guide layer into your project that you toggle on during review.

Aspect ratio versions

A super designed for 16:9 rarely survives a crop to 9:16 or 4:5. Long single lines get cut off or have to shrink below readable size. Plan text layouts per aspect ratio from the start, re-breaking lines and repositioning blocks rather than scaling the whole graphic. This is also where responsive template systems pay off; the article on motion graphics templates for editors covers how to build templates that re-flow text safely across formats.

Viewing distance for other screens

Phones are not the only small screen. Presentation screens seen from the back of a room, digital signage in a corridor and museum displays viewed from a few meters away all shrink the effective size of text. The same logic applies: define the farthest realistic viewer and test at that distance. For the specific constraints of rooms and venues, see motion graphics in presentations.

Animating Text Without Fighting the Reader

Motion is what makes text in video feel alive, and it is also what makes it hard to read. The governing principle is simple: avoid animating text while it must be read. Animation belongs on the way in and the way out, and during the hold the text should be still or moving so gently that the eye can read without tracking.

The three phases of a super

Think of every piece of on-screen text as having three phases. The entrance draws attention and establishes where the text is. The hold is where reading happens. The exit clears the frame for what comes next. Most legibility problems come from the entrance or exit eating into the hold, or from motion continuing through the hold.

Entrances that help reading

Good entrances are short and settle quickly into a stable position. A fade combined with a small positional move, a mask reveal, or a quick scale from slightly larger all work because the text reaches its final shape fast. Word-by-word or line-by-line reveals can actually help comprehension if they follow the natural reading order and each segment stays still once it lands. Character-by-character scrambles, heavy blurs and elastic overshoots look energetic but delay the moment the text becomes readable, which effectively shortens the hold.

The hold

During the hold, keep the text still. A very slow drift or scale, the kind used to keep a frame from feeling frozen, is usually fine provided it is slow enough that the letters do not appear to move relative to each other. Avoid simultaneous motion elsewhere in the frame that competes for attention: a particle burst or a fast camera move behind a super pulls the eye away from the words. The article on particle effects in motion design discusses how to keep decorative motion subordinate to the message.

Exits

Text that flies in and straight out is one of the listed mistakes for good reason. The exit should begin only after the full read-twice hold, and it should be quick and simple. A fast fade or a mirrored version of the entrance is enough. Do not start the exit early to make room for the next beat; move the next beat instead.

Easing and motion blur

Ease text into its resting position so it decelerates and stops cleanly; linear moves that stop abruptly create a visible jolt that draws the eye to motion rather than content. Motion blur on moving text looks natural during fast entrances, but make sure it resolves to a sharp frame before reading starts. In After Effects, the Adobe After Effects documentation explains how motion blur samples and shutter angle affect the result, which is worth understanding when blurred text arrives soft at the start of a hold.

Text attached to moving objects

Labels that follow a product, a person or a location in the shot are a special case: the text must move because its anchor moves. Keep the label itself as stable as possible by smoothing the tracking data, locking rotation and avoiding scale changes, and give it a plate or scrim because its background will change constantly. The guide to motion tracking for graphics covers how to stabilize tracked elements so attached text stays readable.

Compression and Delivery: What Platforms Do to Your Type

Your export is rarely what the viewer sees. Social and video platforms re-encode uploads to fit their own bitrate ladders, then serve lower-quality versions when bandwidth is limited or the player is small. Encoding is designed to preserve what matters to natural images and discard fine, high-contrast detail, which is exactly what small text is made of.

How compression attacks letterforms

• Detail loss at low bitrates. Thin strokes and small counters get averaged away, so lighter weights fade and letters fill in.
• Block and ringing artifacts. Hard edges against smooth backgrounds produce faint halos and blockiness around letters, especially during motion when the encoder has fewer bits per frame.
• Color bleed. Most delivery formats store color at a lower resolution than brightness. Red text on a blue background, or any pairing that relies on hue rather than brightness difference, can bleed and smear at the edges.
• Motion penalties. Frames with lots of movement consume more of the bit budget, so text over a fast camera move or particle storm gets fewer bits and looks worse than the same text over a static shot.

Defensive choices

The defenses mirror the attacks. Use weights with sturdy strokes. Separate text from backgrounds by brightness, not just hue. Keep backgrounds behind text calm, since a scrim or plate reduces the detail the encoder has to spend bits on. Avoid fine textures and gradients inside the letterforms themselves. And keep text still during the hold, because static regions of the frame compress far better than moving ones.

Export practice

Deliver at the resolution and frame rate the platform expects rather than relying on the platform to convert; every conversion is another chance for softness. Export at a generous bitrate so the platform's re-encode starts from clean material. Then do the most important step: upload a private test, watch it on a phone through the actual app, and screenshot any text that looks soft. The review is only meaningful after the platform has done its worst.

Worked Example: Auditing an Illustrative 20-Second Social Spot

The following example is illustrative, not a client project. It shows how the principles above turn into specific, measurable fixes on a typical piece of work: a 20-second vertical ad for a fictional home-fitness app, designed to play with sound off, with five supers over handheld lifestyle footage.

The first cut

The editor's first cut placed five supers across the 20 seconds. The type was the brand's light weight, white, with no shadow or scrim, set at the same size used in the 16:9 master. Each super had an energetic character-scramble entrance and a fly-out exit. Reviewed on a desktop monitor, the cut felt dynamic and every line seemed readable to the team, who had written the script.

The audit applied the working estimate from earlier: about two and a half words per second for display text, doubled for the read-twice rule, plus half a second to acquire the text. Numbers and the URL were counted as extra words. The table compares the time each super needed with the time it actually had once entrances and exits were subtracted.

Summed, the supers needed roughly 25.7 seconds of hold time in a 20-second spot, which confirmed the real problem was the script, not the animation. The revision cut super 2 to five counted words (about 4.5 seconds), cut the offer in super 4 to "Free for 14 days" (also about 4.5 seconds) and moved the progress-tracking line out of the video and into the post caption. The four remaining supers needed about 16.4 seconds of combined hold, which fit inside 20 seconds once entrances were shortened to around a third of a second each, leaving a little air between beats.

The visual fixes

The light weight was swapped for medium on body supers and bold for the offer. A gradient scrim was added across the lower 40 percent of the frame, because the footage included bright windows and a white kitchen that pushed contrast well below the 4.5:1 floor on several frames. Text was re-set for 9:16 with lines broken to four words or fewer, sized up, and moved above the platform's bottom overlay zone. Character scrambles were replaced with short fade-and-rise entrances, and every exit was a plain fade that started only after the hold completed.

The verification

The revised cut was uploaded privately and watched on two phones by three people who had not seen the script. Each read every super aloud before it left, twice in most cases. The one remaining complaint, that the URL on the end card was hard to memorize, was solved by holding the end card a further second and removing a decorative particle loop behind it. None of these changes required new footage, and all were cheaper than a second round of paid media spend on an ad that could not be read.

Special Cases: Kinetic Type, Data, Captions and Localization

The fundamentals apply everywhere, but some kinds of on-screen text bend the rules or add constraints of their own.

Kinetic typography

In kinetic type pieces, text motion is the content. The rule of keeping text still while it is read is relaxed, but not abandoned: the words that carry meaning should land and settle before the next movement starts, while connective words can move more freely. Synchronizing reveals to a voiceover or music track helps enormously, because the audio tells the viewer where they are in the sentence. The article on sound design for motion graphics covers how audio cues can pace text reveals.

Numbers and data labels

Figures carry the highest reading cost per character. When numbers count up to a final value, the count should end early enough that the final value holds still for a full read; viewers remember the resting number, not the animation. Use tabular (fixed-width) numerals so digits do not jitter horizontally as they change, and keep units and labels visible throughout. Live results graphics push these constraints hardest, as described in election results graphics, where many numbers update at once and must stay legible under pressure.

Captions and subtitles

Burned-in captions for sound-off viewing are text in motion too, and they follow the same principles: short lines, sturdy weights, a plate or strong shadow for contrast, and timing that lets viewers read at a comfortable pace. They also compete for the same screen area as designed supers, so decide early whether captions sit above or below the graphic zone and never let the two overlap.

Localization

Translated text rarely fits the space designed for the source language. Some languages run considerably longer than English, and others use scripts with different vertical proportions or reading directions. Longer text needs longer holds, and a layout that relied on tight fit will break. Build templates with room to expand and time budgets that can stretch. The guide to localizing motion graphics covers planning text layouts and timing for multiple languages from the outset.

Building a Legibility Review Into Your Workflow

Legibility problems are easy to fix early and expensive to fix late, especially after footage is graded and media is booked. The most effective teams treat legibility as a named checkpoint in the production process rather than something reviewers might notice.

1. Script with a time budget. Estimate hold time for every line of on-screen text before editing starts and cut words until the total fits the runtime with room for entrances and exits.
2. Lock type standards in the template. Set minimum sizes, weights, line lengths, scrim defaults and safe-zone guides per aspect ratio, so every editor starts from legible defaults.
3. Check contrast across the whole hold. Scrub every super, find the worst frame and verify it against your contrast floor; add or strengthen a scrim where it falls short.
4. Review at real size and speed. Play the cut at full speed on a phone, at arm's length, with sound off if that is how it will be watched.
5. Use fresh eyes. Ask someone who has not read the script to read each super aloud as it appears.
6. Test after platform encoding. Upload privately, view through the platform's app on a phone and fix anything that has softened.
7. Record what you learned. Update the template and style guide so the same problem does not recur on the next project.

Handing standards to developers and other teams

When animated text will be rebuilt in an app, website or game engine rather than delivered as video, the legibility standards need to travel with the design. Specify hold durations, easing, minimum sizes and contrast treatments explicitly rather than leaving them implied in a video reference. The guide to motion specs for developer handoff explains how to document these so they survive implementation.

A comparison of common treatments

The table below summarizes where each contrast treatment tends to fit. It is a planning aid based on the trade-offs discussed above, not a measured benchmark.

When to Bring In Specialist Help

Many teams handle straightforward text well once they adopt the practices above. Bring in help when campaigns rely on on-screen text: sound-off social ads where the words carry the offer, multi-format campaigns that need text to re-flow across several aspect ratios, multilingual rollouts, data-heavy explainers, and any work with regulated claims or disclaimers. In each of these, a legibility failure is a direct failure of the campaign rather than a cosmetic flaw.

A specialist motion team brings three things that are hard to build quickly in-house: tested templates with legibility standards baked in, an eye trained to spot contrast and timing problems before review, and a delivery process that checks results after platform encoding. If you are evaluating outside support, look at how a studio's reel handles text specifically: whether supers hold long enough, whether they stay readable over busy footage and whether vertical versions were properly re-laid out. Our overview of motion graphics production describes how we approach text-heavy work across formats.

Whatever you decide, write your standards down. Minimum sizes, weights, hold formulas, scrim defaults and a phone-based review step are easy to document and make every future project start from a legible baseline.

Where this comes from

• W3C Web Accessibility Initiative — Understanding SC 1.4.3 Contrast (Minimum)
• Adobe Help Center — After Effects User Guide

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

• A Practical Guide to Animating Icons
• Healthcare Explainer Animation: What Actually Works