How to Write Better Prompts for Seedance (24 Examples)

Coding Liquids blog cover featuring Sagnik Bhattacharya for writing better prompts for Seedance.
Coding Liquids blog cover featuring Sagnik Bhattacharya for writing better prompts for Seedance.

The difference between mediocre and impressive Seedance output is almost always the prompt. The model is capable — the question is whether your prompt gives it the right instructions.

The techniques here are written for the current Seedance release and carry across versions, because they target how the model reads a prompt, not the quirks of any one release.

I teach Flutter and Excel with AI — explore my courses if you want structured learning.

This guide covers prompt structure, keyword strategies, and practical patterns that produce consistently better results.

Follow me on Instagram@sagnikteaches

Quick answer

Structure your prompt in order: subject → action/motion → camera movement → style → lighting → quality modifiers. Be specific about motion, use filmmaking terms for camera direction, and keep to 2-3 elements rather than cramming in everything.

Connect on LinkedInSagnik Bhattacharya

Who this guide is for

  • You want to improve the quality and consistency of your Seedance output.
  • Your prompts produce inconsistent or unexpected results.
  • You are new to Seedance and want to start with effective prompt patterns.

Prompt structure that works

A well-structured prompt follows a consistent order. Seedance weights earlier parts of the prompt more heavily, so put the most important elements first.

Subscribe on YouTube@codingliquids
  • 1. Subject: who or what is in the scene
  • 2. Action/motion: what moves and how
  • 3. Camera: camera movement direction and speed
  • 4. Style: visual style, aesthetic, genre
  • 5. Lighting: lighting type and direction
  • 6. Quality: resolution, detail level, production quality

The master prompt template

Rather than writing each prompt from a blank page, drop your details into this fill-in-the-blank skeleton. It hits every dimension Seedance cares about, in the order the model weights them:

[Subject + appearance], [doing what action], [camera movement + speed], [visual style or genre], [lighting + time of day], [quality modifiers].

Filled in, that becomes:

A weathered fisherman mending a net on a wooden dock, hands moving slowly and deliberately, camera gently dollies in from a low angle, documentary realism, soft overcast morning light, shallow depth of field, cinematic, 4K.

The subject and action sit at the front because Seedance weights the opening of the prompt most heavily. The quality modifiers sit at the end because they tune the shot rather than define it — they are the last thing the model should worry about, not the first.

Being specific about motion

'The camera moves' is vague. 'Camera slowly dollies forward through the scene' is specific. Seedance produces dramatically better results with specific motion descriptions.

Include direction (forward, right, up), speed (slowly, gradually, quickly), and type (dolly, pan, orbit, track). These are the three dimensions of motion specificity.

Keywords that improve quality

Certain keywords consistently improve output quality. These are not magic words — they guide the model toward higher-quality generation modes.

  • Quality: 'cinematic', 'professional', 'high-quality', '4K'
  • Lighting: 'golden hour', 'studio lighting', 'dramatic shadows'
  • Depth: 'shallow depth of field', 'bokeh', 'foreground blur'
  • Style: 'photorealistic', 'anime style', 'oil painting style'
  • Motion: 'smooth', 'fluid', 'deliberate', 'gentle'

24 copy-paste prompts for Seedance

These are complete, balanced prompts you can paste straight into Seedance and then adapt. Each one follows the master template above and stays inside the 2-3 action limit, so the model has room to render the shot cleanly instead of fighting competing instructions. To reuse any of them, swap the subject and keep the camera, lighting, and quality scaffolding — that scaffolding is what makes the output look deliberate rather than accidental.

Product & commercial

1. Perfume bottle

A glass perfume bottle on a polished marble surface, a single drop of water sliding down the glass, camera slowly orbits left around the bottle, clean commercial style, bright studio lighting with soft reflections, shallow depth of field, cinematic, 4K.

2. Sneaker on a pedestal

A white running shoe floating above a concrete pedestal, the laces drifting slightly as it rotates, camera slowly orbits right at a low angle, premium sportswear advert style, crisp studio lighting with a soft rim light, shallow depth of field, cinematic, 4K.

3. Coffee being poured

A ceramic cup on a wooden counter, dark coffee streaming in as steam curls upward, camera slowly pushes in from a top-down three-quarter angle, cosy cafe advert style, warm directional morning light, shallow depth of field, cinematic, 4K.

4. Smartwatch reveal

A matte-black smartwatch on a brushed-metal stand, its screen lighting up to show the time, camera slowly tracks right past the device, sleek tech product style, cool studio lighting with subtle reflections, shallow depth of field, cinematic, 4K.

Nature & landscape

5. Misty pine valley

A misty pine valley at sunrise, low fog drifting between the trees, camera slowly cranes upward to reveal the horizon, epic nature documentary style, warm golden-hour backlight, deep depth of field, cinematic, 4K.

6. Waves at dusk

Waves rolling onto a black-sand beach at dusk, foam sliding back into the sea, camera slowly tracks along the shoreline, moody coastal documentary style, deep blue twilight with a faint orange horizon, deep depth of field, cinematic, 4K.

7. Desert dunes

Wind-rippled desert dunes under a vast sky, sand streaming off the crests in the breeze, camera slowly cranes up to reveal the dune field, sweeping nature documentary style, low golden-hour side light casting long shadows, deep depth of field, cinematic, 4K.

8. Northern lights

Green aurora rippling above a snow-covered valley, the lights shifting slowly across the sky, camera holds steady on a wide locked-off shot, awe-inspiring nature style, cold starlit night with a faint snow glow, deep depth of field, 4K.

People & character

9. Chef plating a dish

A young chef plating a dish in a busy kitchen, steam rising as she garnishes the plate, camera gently tracks right at eye level, warm hospitality style, soft window light with practical lamps behind her, shallow depth of field, cinematic.

10. Street musician

A street musician playing an acoustic guitar on a city corner, his head nodding to the rhythm, camera slowly arcs around him, candid documentary style, warm late-afternoon light with long shadows, shallow depth of field, cinematic.

11. Runner at dawn

A runner crossing an empty bridge at dawn, breath visible in the cold air, camera tracks alongside at a steady pace, energetic sports style, cool blue morning light with a warm sunrise behind, shallow depth of field, cinematic.

12. Painter in a studio

An elderly painter adding a brushstroke to a canvas in a sunlit studio, her hand moving slowly and precisely, camera gently pushes in over her shoulder, intimate portrait style, soft window light from the left, shallow depth of field, cinematic.

Anime & stylised

13. Rooftop cherry blossoms

A teenage girl standing on a rooftop as cherry blossoms fall, her hair and scarf drifting in the breeze, camera slowly pulls back to reveal the city, 2D anime style, soft pastel afternoon light, gentle motion, high detail.

14. Mecha cockpit

A pilot strapped into a glowing mecha cockpit, the control panels flickering as she grips the controls, camera slowly pushes in on her face, dynamic anime style, cool blue panel light with warm alert flashes, gentle motion, high detail.

15. Forest spirit

A small forest spirit drifting between ancient trees, glowing motes floating around it, camera slowly tracks forward through the undergrowth, hand-painted Ghibli-inspired style, dappled green afternoon light, gentle motion, high detail.

16. Cyberpunk street

A lone figure in a long coat walking down a rain-soaked neon street, reflections shimmering in the puddles, camera slowly follows from behind, cyberpunk anime style, saturated pink and blue neon at night, gentle motion, high detail.

Food & drink

17. Pasta twirl

A fork twirling fresh spaghetti on a rustic plate, steam rising from the sauce, camera slowly pushes in at a low three-quarter angle, appetising food advert style, warm directional light from the side, shallow depth of field, cinematic, 4K.

18. Chocolate drizzle

Molten chocolate drizzling over a glazed dessert, the surface catching the light, camera slowly orbits the plate, indulgent food advert style, soft top light with a warm glow, shallow depth of field, cinematic, 4K.

19. Cocktail pour

An amber cocktail being poured over ice in a crystal glass, the liquid swirling as it settles, camera slowly tracks right past the glass, upscale bar advert style, moody warm light with a soft backlit glow, shallow depth of field, cinematic, 4K.

20. Fresh bread

A baker tearing open a crusty loaf of bread, steam escaping from the soft inside, camera slowly pushes in on the break, rustic artisan style, warm bakery light from a side window, shallow depth of field, cinematic, 4K.

Abstract & artistic

21. Gold ink in water

Liquid gold ink blooming through clear water, tendrils unfurling in slow motion, camera holds steady in macro close-up, abstract fine-art style, dramatic side lighting against a black background, smooth fluid motion, 4K.

22. Smoke swirls

Wisps of white smoke curling and folding through still air, the shapes shifting slowly, camera holds steady in macro close-up, minimalist abstract style, a single hard light against a black background, smooth fluid motion, 4K.

23. Paint splash

Drops of vivid paint colliding and splashing in slow motion, ribbons arcing outward, camera holds steady in extreme macro, energetic abstract style, bright even studio light against white, smooth fluid motion, 4K.

24. Light particles

Glowing particles of light drifting and swirling through darkness, gathering into a slow vortex, camera slowly pushes in, dreamy abstract style, soft warm glow against deep black, smooth fluid motion, 4K.

What to avoid in prompts

Long, complex prompts with many competing elements produce worse results than short, focused prompts. Seedance cannot optimise for everything at once.

Avoid: contradictory motions, more than 2-3 action elements, describing things that should not happen (negative prompts are less effective than positive direction), and technical jargon the model does not understand.

Building a prompt library

As you find prompts that work well, save them as templates. A good prompt library lets you produce consistent results faster.

Organise templates by use case: product shots, landscapes, character shots, abstract/artistic. Adapt the template for each new generation rather than writing from scratch.

Iterating effectively

Generate 3-4 variations with the same prompt first. If all are bad, the prompt needs work. If some are good, you might just need to generate more and select the best.

Change one thing at a time when iterating. If you change the motion, style, and lighting simultaneously, you cannot tell which change helped.

Worked example: evolving a prompt from basic to polished

Basic: 'A woman in a garden.' (Vague, no motion, no style direction.)

Better: 'A woman in a flower garden, camera slowly dollies forward.' (Adds motion.)

Good: 'A woman in a sunlit flower garden, camera slowly dollies forward, shallow depth of field, golden hour lighting, cinematic.' (Adds style, lighting, quality.)

The final prompt produces dramatically better results because it gives Seedance specific guidance on every important dimension.

Common mistakes

  • Writing very long prompts that try to describe every detail.
  • Not using specific motion terms — 'camera moves' vs 'camera slowly dollies forward'.
  • Changing multiple elements between iterations, making it impossible to learn what works.

Step by step: build a Seedance prompt that works

  1. Write one subject sentence. "A red fox walks through tall grass at dawn." Keep it under 15 words. No adjectives yet.
  2. Add one camera instruction. "Camera slowly dollies left as the fox moves forward." One camera move per clip.
  3. Add one lighting line. "Warm golden-hour backlight, long shadows, soft haze." Lighting sells the shot more than detail words.
  4. Add one motion-intensity cue. "Gentle motion, natural pace." Tells Seedance not to overdrive the scene.
  5. Generate at low resolution first. Use draft quality to test the idea. Only re-render at full quality once the shot looks right.
  6. Revise one element at a time. If lighting is wrong, only change the lighting line. Changing three things at once makes it impossible to tell what helped.

Troubleshooting table

SymptomLikely causeFix
Subject drifts or morphs mid-clipPrompt has too many competing descriptorsCut to one subject sentence. Move style details to a separate style line.
Camera moves too fastGeneric camera verb without qualifierUse "slowly dollies" or "gently pans" — add a speed word.
Scene feels flatNo lighting directionAdd time of day + light source ("late afternoon side light").
Output is overly busyToo many objects in the promptRemove background elements. Seedance handles 1-2 subjects well, 5+ badly.

For the full beginner workflow, read the Seedance beginner tutorial. For motion-specific fixes, see motion intensity settings.

When to use something else

For specific camera movement techniques, see cinematic camera movement. For fixing motion problems, see fixing bad motion in Seedance.

Putting it together: one prompt, three briefs

The real test of a prompt is how well it adapts. Here is the perfume-bottle prompt (#1 above) reworked for three different briefs without starting from scratch. Only the variable parts change — the subject, the camera-and-quality scaffolding, and the one-move-per-clip rule all hold steady.

Brief: a fast, punchy social ad

A glass perfume bottle on a polished marble surface, the cap lifting off as light flares across the glass, camera snaps in quickly then settles, energetic commercial style, bright high-key studio lighting, shallow depth of field, cinematic, 4K, vertical 9:16 framing.

Brief: a slow, high-end hero shot

A glass perfume bottle on black velvet, a single drop of water sliding down the glass, camera drifts in almost imperceptibly, restrained luxury style, low-key lighting with one soft key light and deep shadows, shallow depth of field, cinematic, 4K.

Brief: a lifestyle context shot

A glass perfume bottle on a sunlit bathroom shelf beside fresh flowers, gentle morning light shifting across the scene, camera slowly pans right, warm lifestyle style, soft natural window light, shallow depth of field, cinematic.

What stayed constant: the subject, the "shallow depth of field, cinematic" quality block, and a single camera move per clip. What changed: the pace word, the lighting key (high-key vs low-key vs natural), the setting, and the style label. That is the whole skill — keep the scaffolding, vary the brief. If you would rather start from a still you already have, the same scaffolding applies to image to video prompts.

A six-point check before you hit generate

Generations cost time and credits, so run every prompt past this list first. If it fails a line, fix that line before you spend a render on it.

  1. One clear subject? If you cannot name the single main thing in one breath, the model cannot lock onto it either.
  2. One specific action? Not "moving" but "mending a net" or "sliding down the glass".
  3. Exactly one camera move, with a speed word? "Slowly dollies in", never "dynamic camera" or two moves at once.
  4. A lighting line with a time or a source? "Golden-hour backlight", "soft window light from the left".
  5. A style label? "documentary realism", "2D anime", "abstract fine-art" — one phrase that fixes the look.
  6. Quality modifiers last, not first? "cinematic, 4K" belong at the end, tuning the shot rather than defining it.

Five out of six is usually enough to get a usable clip. A prompt that passes all six rarely produces a truly bad one — when a result disappoints, it almost always traces back to a line you skipped, which is exactly what the motion intensity settings and camera guides help you tighten.

Gemini Omni prompt patterns

For Google's video model, the companion guide is Best Gemini Omni Prompts for AI Video. Use it to compare camera, motion, style, and reference prompt structures across tools.

Frequently asked questions

How long should a Seedance prompt be?

One to three sentences, usually 20 to 40 words. That is long enough to name the subject, one action, a camera move, lighting and a style, but short enough that the model is not juggling five competing instructions. If a prompt runs past about 50 words, it is normally hiding two ideas that belong in separate clips.

What order should I write a Seedance prompt in?

Subject, then action, then camera movement, then style, then lighting, then quality modifiers. Seedance weights the opening of the prompt most heavily, so the subject and what it is doing go first; quality words like "cinematic, 4K" go last because they tune the shot rather than define it.

Why does Seedance ignore part of my prompt?

Almost always because the prompt asks for too much at once. Seedance handles one subject and two to three actions reliably; beyond that it starts dropping or blending instructions. Cut competing descriptors, keep one camera move per clip, and move style detail into its own clause.

How do I get smoother, slower camera movement?

Pair a specific camera verb with a speed word — "camera slowly dollies in" or "gently pans right" — instead of "camera moves" or "dynamic camera". A generic verb with no qualifier is the usual cause of jerky, too-fast motion.

Do quality keywords like "cinematic" and "4K" actually help?

Yes, but as tuning, not magic. Terms like "cinematic", "shallow depth of field" and "golden-hour lighting" nudge the model toward higher-quality generation modes. Keep them at the end of the prompt, and do not rely on them to rescue a weak subject or a vague action.

What is the fastest way to fix a bad result?

Change one element and re-generate. If the motion is wrong, edit only the camera line; if the shot looks flat, add a lighting line. Changing several things at once makes it impossible to learn which change actually helped.

Related guides on this site

These guides cover camera techniques, motion troubleshooting, and platform setup for Seedance.