How to Make a Brand Video for LinkedIn in {{year}}

The spec for LinkedIn

Here is the fast, practical checklist that maps cleanly to LinkedIn's feed behavior and B2B attention patterns:

Aspect ratios: 1:1 square for broad desktop-mobile reach, 9:16 vertical for mobile-first, 16:9 only if the story genuinely benefits from wide framing. Default to 1:1 or 9:16.
Duration cap: ship 20 to 45 seconds. Performance drops hard past 60 seconds for brand intros. Keep one message per video, keep it short.
Autoplay and audio: LinkedIn autoplays muted in feed, viewers turn sound on only if the hook lands. Design for sound-off first with captions and bold on-screen text.
Captions: always-on, high contrast, 2 lines max. Open captions baked in or upload an .srt. Make sure the first line carries the hook, not a generic speaker label.
Thumbnail or poster frame: bold headline, strong brand color, a face or clear product shot. Do not rely on a random frame to sell the click.
Export: 1080x1080 or 1080x1920, H.264 MP4, 24 to 30 fps, variable bitrate with a clean keyframe cadence. Keep motion blur and text animation subtle for legibility.
Safe zones: keep key text inside the central 1:1 box even in vertical exports, so the edit survives cropping everywhere.

The structure that works

Use a beat map that respects LinkedIn's short attention and muted autoplay while still delivering brand clarity. This 35 to 45 second layout has shipped reliably:

0 to 2 seconds - Hook in text plus face or product: one high impact line that names the outcome. Example overlay: "Turn a 10 day onboarding into 2 hours."
2 to 7 seconds - Tension in context: name the pain with a specific metric or scenario. Example: "Your pipeline stalls when handoffs span 6 tools."
7 to 15 seconds - Proof snippet: one stat, one logo strip, or a mini demo clip. Anchor it with a caption like "4 teams live in under 48 hours."
15 to 25 seconds - Solution reveal: show the product in use, map the journey in 3 steps. Voice or captions should be instruction-ready, not hype.
25 to 35 seconds - CTA and value framing: state who this is for, what to do, where to click. Example: "Ops leaders, get the playbook link in comments."
35 to 45 seconds - Optional kicker: quick second proof or feature montage at a slower motion speed, then end card with brand mark.

Editorial guardrails:

One promise per video, one CTA. If you have multiple promises, split them into a series.
On-screen text should express the narrative on its own. Treat audio as an upgrade, not a dependency.
Compose in a square-safe grid. Keep captions, logos, and lower thirds within the middle 80 percent of the frame.

Hooks that earn attention

Hooks for LinkedIn should be concrete, useful, and B2B fluent. Use these formulas with examples:

Outcome plus time compression

Formula: [Outcome], [time frame].
Examples:
- "Ship your QBR in 30 minutes, not 3 days."
- "Hire 3 SDRs in 2 weeks with one repeatable funnel."

Contrarian POV that fixes a common habit

Formula: "Stop [popular but weak tactic]. Do [specific better tactic]."
Examples:
- "Stop demoing features. Demo outcomes in the first 15 seconds."
- "Stop gating PDFs. Gate repeatable templates instead."

Single metric reveal with a direct benefit

Formula: "[Metric] that changed [result]."
Examples:
- "One calendar rule that cut no-shows by 41 percent."
- "Three labels that cut bug triage time by half."

Mini case study in one sentence

Formula: "[Customer] went from [pain] to [result] by [method]."
Examples:
- "Acme Ops went from 10 tools to 1 workflow by templating handoffs."
- "Northbound HR went from 12 email threads to 1 checklist with auto reminders."

Checklist teaser

Formula: "If you do [X, Y, Z], you will get [benefit]."
Examples:
- "If you plan the hook, the proof, and the CTA, your LinkedIn watch time jumps."
- "If your captions are high contrast, short, and local to the action, more viewers switch audio on."

Brand + voice

One-off hits feel good. Consistency wins. A strong brand kit and a stable voice will compound reach and trust faster than any single viral video. The kit enforces recognizability, reduces creative thrash, and makes iteration safe. The voice makes your sentences sound like the same person, even when a different teammate records the talking head.

Build and apply a kit that includes:

Color tokens with accessible contrast pairs, primary and accent roles.
Typographic scale for titles, captions, and annotations that survives 1:1 and 9:16 exports.
Motion grammar: entry, hold, exit speeds, easing, and how lower thirds behave.
Graphic system: safe areas, logo placement, corner radius, background grain or solid, watermark rules.
Audio palette: two tracks that fit your brand tempo, plus bed volume standards and ducking rules.
Voice-of guidelines: person, tense, jargon tolerance, sentence length, favorite verbs, words to avoid.

With HyperVids, the per-project brand kit captures exactly those pieces, then applies them across short-form, talking-head, explainer, or audiogram formats so every LinkedIn brand video looks and sounds like you without rewriting the rules each time.

Captions + accessibility

Captions on LinkedIn are not optional. They are the backbone of comprehension and attention in a muted feed. Use these concrete rules:

Always-on: open captions for the hook and core lines. You can add an .srt if you prefer closed captions for the full transcript.
Contrast: white text on a near-black bar at 70 to 85 percent opacity or black text on a brand-light bar that meets at least a 4.5:1 contrast ratio for normal text, 3:1 if your caption font is large.
Line length: 28 to 32 characters per line, 2 lines max. Shorter lines scan faster.
Font size: for 1080 height, target 48 to 64 px equivalent with 1.2 line height. Increase slightly for vertical since the text is closer to thumbs.
Position: sit the caption block 12 to 14 percent above the bottom edge. Maintain 10 percent side margins to avoid UI and engagement overlays.
Speaker clarity: use inline cues like "CEO:" only if multiple voices appear. Otherwise devote line one to the content, not the label.
Language: sentence case, no ellipses, no all caps. Write short clauses that complete ideas independently of sound.
Post text: add 1 to 2 lines that summarize the value and link to the resource. It improves accessibility and helps screen readers.

A sample HyperVids prompt

Drop one clear prompt so the system can generate the right beats, captions, and exports for LinkedIn:

Brand kit:
- Colors: Primary #0A84FF, Accent #111827, Light #F9FAFB
- Typography: Inter Bold for headers, Inter Regular for captions
- Motion: Fast in 200ms, hold, exit 200ms, ease-in-out
- Caption style: White on 80% black bar, 2 lines, 32 chars line max
- Logo: Top right, 64px, 80% opacity watermark

Prompt:
"LinkedIn brand video, 35s, 1:1 and 9:16 outputs.
Audience: RevOps managers at mid-market SaaS.
Hook: 'Turn 10-day onboarding into 2 hours.'
Structure: hook, pain, proof (48h rollout), solution demo (3 steps), CTA 'Grab the onboarding checklist in comments.'
Format: talking-head + screen capture.
Tone: practical, developer-friendly, no fluff.
Include open captions for hook and key lines, export poster frame with headline + brand mark."

In HyperVids, that single input uses the /hyperframes skill with your existing Claude CLI subscription to produce a time-coded script, motion templates aligned to the kit, open captions, 1:1 and 9:16 renders, and a poster frame. You can swap the hook or CTA, re-render, and keep the same kit automatically.

Common failure modes

Slow open: anything that starts with a logo sting before the promise loses attention. Hook first, logo later.
Overlength: a 90 second brand story in feed is rarely watched. Break it into a series and link a playlist.
Caption clutter: long lines, small fonts, low contrast. Tighten to two short lines and raise contrast.
All demo, no bridge: showing features without naming the outcome forces viewers to guess why it matters.
Unclear CTA: ending on a generic "learn more" with no destination wastes the watch. Use a concrete next step.
Mixed brand rules: each video uses different fonts, colors, and motion speed. It dilutes recognition.
Unsafe framing: text or faces pushed to edges that get cropped on mobile. Compose square-safe.
Muted music mismatch: tracks that fight the voice or feel off-tempo. Use a small, branded palette.
No poster frame: relying on a random frame makes the click depend on luck, not your story.
Wall-of-text posts: 10 lines of post text before the video link discourages clicks. Use 1 to 2 lines and a link.

Conclusion

LinkedIn rewards short, specific brand videos with strong hooks, proof, and clear calls to action. Compose for sound-off first, keep captions legible, use a repeatable beat map, and protect your brand kit across exports. When your system is tight, iteration becomes a weekly habit instead of a quarterly project. HyperVids makes the kit, captions, and multi-ratio outputs feel automatic so you can focus on sharper hooks and better proof.

FAQ

How long should a LinkedIn brand video be?

Target 20 to 45 seconds for feed. If you need more depth, publish a short series, link the next clip in comments, and keep each clip focused on one promise.

Square or vertical, which performs better?

Vertical wins on mobile, square balances desktop and mobile. If your audience skews mobile, lead with 9:16. If it is mixed, export both and let the post or ad placement choose the ratio.

Do I still need captions if I use on-screen text?

Yes. On-screen title text is not enough. Captions carry the full sentence-level meaning that survives muted autoplay and scanning, they also support accessibility guidelines.