AI Prompts for Runway Prompts for Marketing Videos

I am marketing [PRODUCT / SERVICE] to [AUDIENCE] with the core message: [MESSAGE]. Translate this into a Runway-producible video concept: the 3-5 shot sequence that tells the story visually (no dialogue dependence), what each shot must communicate, and the honest feasibility check: which shots AI video generates well (atmosphere, product beauty, abstract motion) versus which to rethink (specific hand interactions, legible text, complex choreography).

My brand benefit is abstract: [BENEFIT: SAVES TIME / REDUCES STRESS / BRINGS CLARITY]. Generate 8 visual metaphors that Runway can render: physical transformations, natural phenomena, spatial changes that embody the benefit. For each: the one-line concept, why it maps to my benefit, and the draft generation prompt. Abstract benefits need concrete images; that is the whole craft.

Define my brand’s visual world for Runway generations: [BRAND, INDUSTRY, PERSONALITY]. Specify: the color world (dominant palette and light quality), the environment language (minimal studio, urban, natural, domestic), the motion personality (slow and considered versus kinetic), and texture (clean digital, filmic grain, soft focus). Output as a reusable style block I append to every generation prompt for consistency.

Build the storyboard for a [LENGTH: 30-SECOND] marketing video: [PASTE BRIEF OR DESCRIBE GOAL]. For each shot: duration, the action or motion, camera behavior (static, slow push, orbit), how it connects to the next shot, and where text overlays or the logo land in post. Mark which shots come from Runway generation, which from image-to-video on product photos, and which need stock or live capture.

Video ads in [CATEGORY] all look the same: [DESCRIBE THE CLICHES YOU SEE]. Identify the category patterns worth breaking and design 3 counter-concepts Runway can execute: different pacing, unexpected environments, or a visual tone nobody in the category uses. For each: the concept, the risk, and the first test shot to generate before committing to the full video.

Convert my shot description into a strong Runway prompt: [DESCRIBE SHOT]. Structure it as: subject and action, environment and lighting, camera movement and framing, style and mood keywords, using concrete cinematic language (golden hour rim light, shallow depth of field, slow dolly-in) not vibes (beautiful, epic). Give me the main prompt plus 2 variations that adjust the variable most likely to miss: [MOTION / LIGHTING / FRAMING].

I have brand assets to animate: [DESCRIBE: PRODUCT PHOTOS, CAMPAIGN STILLS, BRAND IMAGERY]. Plan the image-to-video approach in Runway: which images will animate convincingly (depth, clear subject, motion potential) versus which will warp, the motion prompt for each chosen image (what moves, what stays anchored, camera behavior), and the subtle-motion rule: less movement usually reads as more premium.

My generations move wrong: [PROBLEM: TOO CHAOTIC / STIFF / WARPING]. Teach me the motion vocabulary that controls Runway: the camera terms (static, pan, dolly, orbit, handheld), the subject motion terms (drifts, unfurls, cascades, settles), and intensity modifiers. Then rewrite my failing prompt ([PASTE]) with precise motion language and explain each change.

My video needs [NUMBER] shots that feel like one film. Build the consistency strategy: the shared style block for every prompt (from my brand world), reference image usage to anchor the look, matching light direction and color temperature across prompts, and the generation order: establish the hero shot first, then prompt the rest to match it. Output the full prompt set for my shot list: [PASTE SHOTS].

My generation almost works but: [ISSUE: HANDS ARE WRONG / TEXT IS GARBLED / MOTION DRIFTS OFF-SUBJECT / THE VIBE IS OFF]. Give me the fix protocol in order: prompt adjustments that target this specific failure, whether to reroll versus revise, when to crop or trim around the flaw in editing instead of regenerating, and when to restructure the shot to avoid what AI video cannot yet do.

Plan the edit for my marketing video: [LIST GENERATED CLIPS WITH DESCRIPTIONS]. Define: the cut order and pacing (where fast cuts create energy, where a long shot builds premium feel), transition style (hard cuts almost always; note any exception), where the message text overlays land and for how long, logo placement, and total runtime for [PLACEMENT: FEED AD / LANDING PAGE / PRE-ROLL].

Brief the sound for my video: [DESCRIBE THE VIDEO AND MOOD]. Specify: music style and energy curve matched to the edit, whether voiceover carries the message or text does, the sound design moments that sell realism (ambient texture, a whoosh on the product reveal), and the mix priorities for sound-off viewing: the video must work muted, sound is the bonus layer.

Write the on-screen text for my video: [MESSAGE, OFFER, CTA]. Rules: maximum [NUMBER] words per card, readable in under 2 seconds each, message arc across the video (hook, value, proof, CTA), and safe-zone placement for [PLATFORMS]. Output the text cards in order with timing, plus the font and animation guidance matching my brand: [BRAND STYLE].

My video runs on [PLACEMENTS: FEED, STORIES, YOUTUBE, SITE]. Plan the aspect ratio strategy: generate in [RATIO] and reframe, or generate per ratio? Consider how my key shots crop ([DESCRIBE SHOT COMPOSITIONS]), which placements drive most impressions, and Runway generation settings per ratio. Output the ratio production plan with the least regeneration work.

Build my pre-publish checklist for AI-generated marketing video: the artifact scan (warped hands, morphing objects, flickering textures, garbled text in-scene), brand accuracy (colors, product details against real product), message clarity muted and with sound, platform spec compliance ([SPECS]), and the fresh-eyes rule: one person who has not seen it watches once and says what the video is about. If they miss the message, recut.

Adapt my finished video for each placement: [LIST: META FEED, REELS, YOUTUBE PRE-ROLL, LANDING PAGE]. For each: the length cut, the hook adjustment (pre-roll needs the brand earlier, feed needs the pattern interrupt first), text overlay changes, and whether any placement justifies generating an alternate opening shot in Runway rather than recutting.

Design a hook test: 3 alternate opening shots for the same video body. From my concept ([DESCRIBE]), define 3 different first-3-seconds approaches: [MECHANISMS: PRODUCT BEAUTY SHOT, HUMAN MOMENT, VISUAL METAPHOR IN MOTION]. Write the Runway prompt for each opener, matched to cut seamlessly into my existing edit. First-frame and first-second decisions dominate feed performance.

Read my video results: [PASTE METRICS: IMPRESSIONS, THUMB-STOP RATE, WATCH TIME CURVE, CTR, CONVERSIONS]. Diagnose by stage: weak thumb-stop is the opening shot, watch-time cliff at [SECOND] is that shot or pacing, clicks without conversions is a message-landing page mismatch. Prescribe: the specific shot to regenerate or edit change, not generic advice.

Plan my iteration sprint from the winning video: [DESCRIBE WINNER AND ITS METRICS]. Generate the variant agenda: same structure with a new environment (one prompt change), same shots with a different pace edit, a seasonal or offer refresh, and one wildcard concept. For each variant: the Runway work needed, effort estimate, and the hypothesis it tests. Ship [NUMBER] variants in [TIMEFRAME].

Review my Runway marketing video economics this quarter: videos produced ([NUMBER]), total production time and subscription cost versus the agency or shoot alternative ([QUOTE OR ESTIMATE]), performance versus my previous non-AI creative ([DATA]), and where AI video underdelivered. Verdict: scale up, hold, or hybrid (AI for volume and testing, shoots for hero campaigns). Justify with the numbers.