Marketing

Marketing production with genmedia#

Use this skill when the user wants a campaign system, not one isolated asset. Load references as needed:

• references/campaign-patterns.md
• references/workflows.md
• references/examples.md

Load model-routing alongside this skill for default endpoint choices.

Marketing output should be executable: a brief, asset matrix, genmedia command plan, downloaded files, and clear defects. Do not add fake claims, fake social proof, fake partner logos, fake award badges, or invented legal copy.

Inputs to collect#

Only ask when missing details would change the asset plan.

• Objective: launch, acquisition, retargeting, activation, education, event.
• Audience: persona, market, use case, awareness level, objections.
• Offer: product, feature, bundle, trial, waitlist, event, promotion.
• Channels: paid social, organic social, landing page, email, display, app.
• Required assets: stills, video, thumbnails, hero image, carousel, banner.
• Brand rules: colors, logo use, typography, tone, taboo visuals, competitors.
• Source media: product images, screenshots, app UI, people, logo, prior ads.
• Claims: exact approved copy, proof points, disclaimers, compliance limits.
• Budget preference: final-quality, draft exploration, or mixed pipeline.

Genmedia workflow#

1. Start from routed endpoint IDs.

   genmedia models --endpoint_id openai/gpt-image-2 --json
   genmedia models --endpoint_id fal-ai/nano-banana-pro/edit --json
   genmedia models --endpoint_id fal-ai/nano-banana-2 --json
   genmedia models --endpoint_id bytedance/seedance-2.0/image-to-video --json
   genmedia models --endpoint_id bytedance/seedance-2.0/text-to-video --json
   genmedia models --endpoint_id veed/fabric-1.0 --json

   Use text search only as fallback discovery for a missing channel role:

   genmedia models "marketing banner image typography" --json
   genmedia models "social ad video product campaign" --json
   genmedia docs "image generation text rendering" --json

2. Inspect the selected endpoint before planning exact payloads.

   genmedia schema <endpoint_id> --json
   genmedia pricing <endpoint_id> --json

3. Upload all source media.

   genmedia upload ./product.png --json
   genmedia upload ./logo.svg --json
   genmedia upload ./screenshot.png --json
   genmedia upload ./creator.jpg --json

4. Build the campaign matrix before generating.

   Minimum matrix:

   • Hook asset: earns attention in the first frame.
   • Proof asset: shows product, result, process, feature, or evidence.
   • Context asset: places product in audience use case.
   • Conversion asset: clean end frame with safe space for external copy.

5. Run still assets synchronously when quick.

   genmedia run <endpoint_id> \
     --prompt "<marketing asset prompt>" \
     --download "./outputs/marketing/{request_id}_{index}.{ext}" \
     --json

6. Run video assets async and download from status.

   genmedia run <endpoint_id> \
     --prompt "<campaign video prompt>" \
     --async \
     --json
   
   genmedia status <endpoint_id> <request_id> \
     --download "./outputs/marketing/{request_id}_{index}.{ext}" \
     --json

7. Use schema fields exactly. Do not pass guessed flags. Mirror names such as image_urls, image_url, reference_image_url, aspect_ratio, duration, quality, seed, prompt, or negative_prompt.

Brief build order#

Build the campaign brief before prompts:

1. Product truth: what the product is and what must stay accurate.
2. Audience tension: problem, desire, objection, or trigger moment.
3. Campaign promise: user-approved benefit or visible outcome.
4. Asset roles: hook, proof, context, conversion, retention, reminder.
5. Channel specs: crop, runtime, safe zones, text needs, file count.
6. Variation axis: audience, setting, proof point, visual metaphor, offer.
7. Guardrails: claims, logo, legal, copy, product fidelity, banned imagery.

Prompt build order#

Write every asset prompt in this order:

1. Asset role and channel: launch hero, Meta ad, TikTok hook, email header.
2. Product or subject invariant: exact product, UI, feature, person, logo rule.
3. Audience context: where, who, problem state, usage moment.
4. Visual system: camera, lighting, color, composition, texture, motion.
5. Copy handling: no generated text, safe space, or exact provided wording.
6. Variant axis: what differs from the other assets.
7. Guardrails: no fake claims, no extra logos, no distorted product or UI.

Model routing#

• Text-heavy campaign key art, posters, landing heroes, app/UI visuals, and ads with exact copy: openai/gpt-image-2 at quality=high.
• Premium product or brand stills: openai/gpt-image-2, then fal-ai/nano-banana-pro, then fal-ai/nano-banana-2.
• Edits from source assets: fal-ai/nano-banana-pro/edit, then openai/gpt-image-2/edit, then fal-ai/bytedance/seedream/v5/lite/edit.
• Fast visual exploration: fal-ai/flux-2/klein/9b.
• Product reveal or social video: bytedance/seedance-2.0/image-to-video.
• Text-to-video concept or brand film: bytedance/seedance-2.0/text-to-video.
• Creator or spokesperson ad: load ugc and use veed/fabric-1.0, veed/fabric-1.0/text, or fal-ai/creatify/aurora.
• Multi-shot narrative: load storytelling.
• Single polished product asset: load commercial.

Quality bar#

Before returning:

• Every asset maps to a campaign role and channel.
• Product, UI, logo, and packaging are stable.
• Claims are user-supplied, observable, or removed.
• Text is either exact and schema-supported or reserved for external editing.
• Variants differ by one clear axis, not random style drift.
• Safe zones support overlays, cropping, and platform UI.
• Output paths come from downloaded_files[].
• The final answer includes a campaign manifest: asset role, endpoint, request id, local path, prompt summary, and defects.

If the campaign feels generic, reduce the asset count and make each role more specific before generating more variants.

Supporting file: references/campaign-patterns.md#

Marketing campaign patterns#

Use these patterns to turn a request into an asset matrix before generating.

Hook, proof, context, conversion#

The default paid-social set.

Awareness ladder#

Match the asset to audience awareness.

Channel matrix#

Variation axes#

Vary one axis at a time.

• Audience: beginner, expert, buyer, operator, creator, founder.
• Problem: time, cost, complexity, trust, quality, status, risk.
• Proof: demo, screenshot, texture, testimonial, process, comparison.
• Setting: desk, bathroom, kitchen, gym, store, studio, outdoor, event.
• Tone: direct, expert, warm, premium, playful, practical, urgent.
• Format: still, video, creator, carousel, before-after, product detail.

Claim discipline#

Never invent:

• review quotes
• user counts
• revenue or conversion numbers
• clinical, financial, legal, safety, or performance claims
• partner logos, award badges, app store ratings, press logos

When claims are missing, use observable language: "shows the workflow", "cleaner setup", "fewer steps on screen", "product texture visible".

Supporting file: references/examples.md#

Marketing examples#

Adapt these examples to the user's real product, brand rules, files, and selected model schema.

SaaS launch kit#

Fitness product paid social#

4:5 paid social still for a compact black recovery device, product unchanged
from uploaded reference, athlete's hand places it beside a gym towel and water
bottle, crisp side light, high contrast but natural, large clean space in upper
third for external headline, no generated text, no performance claims

Beauty email header#

wide email header image, uploaded moisturizer jar centered on pale stone,
soft morning light, subtle water texture in background, premium but natural,
product label unchanged, no extra objects blocking packaging, no text, clean
negative space on left for email headline

App feature visual#

landing-page hero visual for a productivity app, supplied screenshot placed
inside a clean laptop mockup, organized desk, daylight, realistic scale,
abstract UI details only if no screenshot is supplied, no fake readable text,
calm modern workflow mood

Campaign manifest shape#

Supporting file: references/workflows.md#

Marketing workflows#

Campaign asset set#

1. Write a one-paragraph brief.
2. Build an asset matrix with role, channel, crop, endpoint, source media, and variation axis.
3. Generate the minimum complete set: hook, proof, context, conversion.
4. Inspect outputs before expanding variants.
5. Return a manifest with local paths and defects.

Launch kit#

Use when the user needs broad go-to-market visuals.

1. Create one hero still.
2. Create one product or feature detail.
3. Create one social hook video or motion reveal.
4. Create one clean conversion/end frame.
5. Optional: create one UGC creator variant through the ugc skill.

Paid social variant test#

1. Keep product, offer, and crop constant.
2. Generate 3-5 variants that differ by hook angle only.
3. Use exact output naming with {request_id}_{index}.
4. Score each output for thumb-stop, product clarity, claim safety, and crop.
5. Do not mix model changes and concept changes in the same test unless the user specifically wants a model shootout.

Landing-page visual system#

1. Define hero, proof, feature, and conversion sections.
2. Generate visuals with consistent palette, lighting, and product treatment.
3. Leave text and UI copy for the frontend unless the model is explicitly selected for text rendering.
4. Return assets in page order.

Creator ad package#

1. Use ugc for the talking-head or faceless creator asset.
2. Use this skill for campaign framing, channel matrix, hook variants, and conversion/end frames.
3. Keep the same proof and offer across creator variants.
4. Return creator clips and supporting marketing assets together.