--- title: "Image-to-Video Google Ads Scripts — Photo Workflow" description: "The 5-step image-to-video workflow for Google Ads: photo audit, 4-beat I2V ad script, Kling 2.6 vs Wan 2.7 cost math, specs, and 8 product templates." canonical: "https://aicontentdrop.com/blog/image-to-video-google-ads-script-ai-content-drop" source: "https://aicontentdrop.com/blog/image-to-video-google-ads-script-ai-content-drop" --- Most Google Ads video briefs start with the wrong question. Teams ask "what should the ad look like?" when they already have a drawer full of product photos, packshots, e-commerce stills, and lifestyle shoots sitting in a Drive folder. If your brand has been operating for more than a quarter, you have seed material. The question isn't what the ad should look like — the question is how to turn the assets you already own into motion that holds up in the Google Ads asset library next to your search and display creatives. That's the case for image-to-video (I2V) in a Google Ads workflow. You keep brand fidelity — the actual bottle, the actual packaging, the actual model your audience has seen on your PDP. Google's asset quality scoring rewards consistency across campaign creatives, which means your video ad performs better when the hero frame matches the photo that's already converting on Performance Max. And because a seed-locked I2V pass costs roughly the same as a text-to-video generation, you're not paying extra for brand accuracy — you're paying the same price for dramatically more lift. This post is the workflow, the script skeleton, the eight templates by product type, and the math for running this at $49 a month through AI Content Drop. Image-to-Video Google Ads Script — product-reveal rotation — generated with Wan 2.7 text-to-video on AI Content Drop. ## Why Image-to-Video Beats Text-to-Video for Google Ads Product Campaigns Text-to-video is the right tool for early concept exploration, editorial B-roll, and campaigns where the product is conceptual rather than physical. For most Google Ads product advertisers, it's the wrong tool. Three reasons, in order of commercial weight. **Brand fidelity.** A text prompt describing "frosted glass bottle with rose-gold cap and serif wordmark" will produce a plausible bottle. It will not produce *your* bottle. Any reviewer with the actual SKU in hand can spot the difference in a second. Seed an I2V generation with your real packshot and the bottle stays exact — wordmark, cap geometry, material finish, all preserved. For any regulated or trademarked category (supplements, alcohol, cosmetics with efficacy claims, electronics with model numbers), this isn't a preference; it's a legal requirement. **Google Ads asset quality scoring.** Google grades assets partly on how coherent they are with the rest of the account — logo treatment, color palette, subject continuity. When your Video ad hero matches the images in your Sitelink extensions, your Performance Max asset group, and your Demand Gen creative, the quality signal reads higher and your CPMs tend to drop. Text-to-video rolls the dice on palette every generation. I2V locks it. **Cost efficiency per finished ad.** A Wan 2.7 I2V or Kling 2.6 I2V pass runs roughly 40-55 credits on AI Content Drop. Starting from a photo you already paid for means your marginal cost per finished ad is just the motion pass. On text-to-video, you pay the same motion credits but also spend generations iterating toward a frame that looks like your product, which most of the time it doesn't. For hero shots that need total brand control, reach for Wan 2.7 T2V when no seed exists; for brand-locked reveals on existing product photos, reach for Wan 2.7 I2V or Kling 2.6 I2V. These three models do different jobs — cover all of them in the [AI Content Drop video marketplace](https://aicontentdrop.com/best-ai-video-generator). ## The Image-to-Video Google Ads Workflow in 5 Steps Five steps, roughly 25 minutes end-to-end for a first ad, 15 minutes per subsequent ad once the template is dialled in. ### Step 1 — Audit your seed photos Not every product photo animates well. Before you spend credits, screen the candidates. Minimum 1920 pixels on the long edge — anything less and the I2V pass has no detail to interpolate from. Lighting direction readable in a single glance (is the key from the left, top, right?). And critically, cropability: can this photo survive both a 16:9 crop for YouTube In-Stream and a 9:16 crop for Shorts without losing the subject? The photos that work best have generous margins on all four sides and the hero product off-center enough that reframing doesn't decapitate the composition. ### Step 2 — Write the 4-beat I2V ad script The skeleton is four lines. Seed description — not a prompt to generate an image, but a description of the photo you're feeding in, because the motion model uses your words to decide what can move. Voiceover hook — 15-25 words that run across the first six to eight seconds. Animation brief with timestamps — camera motion, subject motion, pacing cues. CTA frame — the final two-second beat where your Google Ads text overlay lands in post. Full anatomy in the next section. ### Step 3 — Generate with AI Content Drop Open the [video generator](https://aicontentdrop.com/best-ai-video-generator), upload the seed, pick Kling 2.6 I2V or Wan 2.7 I2V, paste the animation brief, ship. Short cost comparison: | Model | Best For | Strength | Cost / Pass | | --- | --- | --- | --- | | Kling 2.6 I2V | Human subjects, lifestyle, warm grade | Natural faces, soft motion | ~17 credits | | Wan 2.7 I2V | Product heroes, macro, clean studio motion | Sharp texture, controlled camera | ~42 credits | | Wan 2.7 T2V | Hero shots with no existing photo | Cinematic control from scratch | ~42 credits | ### Step 4 — Assemble in your editor Drop the rendered clip into your editor of choice (Premiere, CapCut, DaVinci — any NLE). Three layers: the I2V footage, the voiceover track, and the CTA text overlay. Keep the CTA text overlay in post — do not ask the model to render text, because it will not produce legible wordmarks reliably. Export at the Google Ads asset specs for your target placement (specs table in Section 7). ### Step 5 — Upload to Google Ads asset library Google Ads video ads run through YouTube. Upload the finished MP4 to your connected YouTube channel (unlisted is fine for ad use). Inside Google Ads, go to Tools & Settings → Shared Library → Asset Library → Videos, paste the YouTube URL, and it joins your reusable asset pool. From there you attach it to a Video campaign (Skippable In-Stream, Bumper, In-Feed) or to Demand Gen / Performance Max asset groups. Same MP4 can fuel multiple campaign types — this is why brand fidelity compounds so well on Google's side. ## The 4-Beat I2V Ad Script Skeleton Every template in Section 5 follows this exact structure. If you write your own, reuse it: *"Seed description: [two sentences describing the photo you're uploading — subject, framing, lens feel, lighting direction, palette]. Voiceover hook: [15-25 words, reads in 6-8 seconds]. Animation brief with timestamps: [camera motion 0-Xs, subject motion Ys-Zs, pacing notes]. CTA frame: [final 2-second beat — logo position, product hold, color wash, space reserved for Google Ads text overlay]."* - Beat 1 — Seed description. Two sentences that tell the I2V model what the still frame already contains. You're not generating an image; you're telling the motion model what to respect as it interpolates. Include subject, framing, lens feel, lighting direction, and palette. - Beat 2 — Voiceover hook. 15-25 words. This runs across seconds 0-6 of the ad. Neither Kling 2.6 nor Wan 2.7 I2V lip-syncs reliably on this workflow, so the hook is voiceover, not dialogue from an on-screen speaker. - Beat 3 — Animation brief with timestamps. Camera motion (push, pull, pan, hold, orbit) with second markers. Subject motion — what moves and when. Pacing cue so the payoff lands before second eight. - Beat 4 — CTA frame. The final 1-2 seconds. Describe what's on screen and where you're leaving space for your Google Ads text overlay (bottom-third for CTAs, lower-right for prices, top-banner for offers). ## 8 Image-to-Video Script Templates by Product Type Eight product categories, eight I2V script templates, each tuned to a specific Google Ads placement. Copy, swap variables, ship. ### 1. Apparel (Fashion) — 9:16 Shorts + Demand Gen *"Seed description: Model in an oversized cream wool coat walking through a narrow cobblestone alley at blue hour, 9:16 vertical crop, 35mm lens feel, teal-and-amber cinematic grade, soft streetlamp light. Voiceover hook: She hasn't turned around yet. Wait for it. Animation brief: Kling 2.6 handheld tracking shot behind subject 0-3s with slight bounce, wind moves hair and coat hem continuously, subject begins a half-turn over the shoulder 4-7s, face enters frame in partial profile 7-8s. CTA frame 8-10s: full face reveal, brand label visible on lapel, space reserved bottom-third for Shop Now overlay."* ### 2. Beauty / Skincare — 9:16 Shorts + In-Stream *"Seed description: A single frosted-glass serum bottle on a wet marble surface, soft morning light from window camera-left, pale pink and cream palette, 9:16 vertical, 85mm lens feel, shallow depth of field. Voiceover hook: The serum that sold out three times this year is back in stock for 72 hours. Animation brief: Wan 2.7 I2V slow push-in 0-4s toward a tight macro on the label, a single water droplet slides down the glass 1-3s, subtle bottle rotation 6-8s reveals the back label. CTA frame 8-10s: bottle centered, soft glow wash, lower-third reserved for Limited Restock price tag."* ### 3. Food / Beverage — 16:9 Bumper + In-Stream *"Seed description: Extreme close-up macro of a burger being lifted off a wooden board, melted cheese mid-stretch, sesame bun in crisp focus, 16:9 horizontal crop, 100mm macro lens feel, tungsten warm key light. Voiceover hook: Six seconds. Cheese at second four. Line outside by second six. Animation brief: Wan 2.7 I2V slow vertical lift 0-4s with cheese stretching further, steam rises continuously, subtle rotation of the bun 4-5s reveals the stack. CTA frame 5-6s: restaurant logo bottom-right, warm glow wash, top-banner reserved for Order Now extension."* Because this is a 6-second Bumper, every beat is tighter. Lose the push-in, live entirely inside the lift-and-stretch motion. ### 4. Tech / Electronics — 16:9 In-Stream Skippable *"Seed description: A sleek laptop on a dark walnut desk with dashboard UI visible on screen, ambient violet LED accent from above, matte black coffee mug releasing steam, 16:9 horizontal, 35mm lens feel, moody cinematic lighting. Voiceover hook: The dashboard our ops team refused to share. We made them. Here's the 20-second tour. Animation brief: Wan 2.7 I2V slow push-in 1-4s toward the screen, hand enters frame and hovers over trackpad 4-6s, coffee steam rises throughout, camera drifts slightly left 6-10s revealing more of the UI. Subtle click gesture 10-12s. CTA frame 12-15s: product logo overlay space top-left, dashboard glow wash, bottom-third reserved for Start Free Trial overlay."* Skippable In-Stream rewards 15s as a strong second-choice duration after 30s. The ad has to hook inside 5 seconds before the Skip button appears; hence the push-in starts at second one. ### 5. Home Goods — 16:9 Discovery + 9:16 Shorts *"Seed description: A linen-upholstered armchair in a sunlit minimalist living room, hardwood floor, a single draped throw and a ceramic floor lamp, soft afternoon light from a tall window camera-left, 9:16 vertical crop (re-export 16:9 for Discovery), 50mm lens feel, warm neutral palette. Voiceover hook: The chair half the room talked about. Here's how it moves when you sit down. Animation brief: Kling 2.6 I2V slow orbit 0-3s revealing the full silhouette, fabric texture catches changing light 3-5s, subtle dust motes in the sunbeam 5-7s. CTA frame 8-10s: chair centered, warm glow, lower-third reserved for price + Free Shipping Google Ads overlay."* ### 6. Supplements / Wellness — 16:9 Bumper + Demand Gen *"Seed description: A supplement bottle on a clean white desk next to a hydrated glass of water, a journal open showing a checked morning routine, 16:9 horizontal, diffused daylight from overhead, soft neutral palette, 50mm lens feel. Voiceover hook: The afternoon routine three-quarters of our customers reorder without being asked. Animation brief: Wan 2.7 I2V slow push-in 0-3s toward the bottle, pages of the journal lift gently in a breeze 2-4s, subtle light shift warms the palette 4-5s. CTA frame 5-6s: bottle centered, brand color wash, bottom-third reserved for offer overlay (keep overlay claim-compliant — no efficacy language)."* Supplement ads live and die on Google Ads policy compliance. Keep the voiceover ritual-focused ("routine," "reorder") rather than outcome-focused ("reduces," "boosts"). The frame does the implying; the copy stays safe. ### 7. Pet Products — 9:16 Shorts + In-Feed *"Seed description: A medium-sized dog mid-chew on a beef dental treat, eyes bright, soft natural daylight through a kitchen window, 9:16 vertical, 50mm lens feel, warm wood floor background, a branded treat pouch out of focus behind. Voiceover hook: Three seconds of chewing. Twelve days of fresher breath. You decide what's worth it. Animation brief: Kling 2.6 I2V hold 0-1s on the chew, subtle head tilt 1-3s, dog looks up toward camera 3-5s, slight tail-wag motion throughout. CTA frame 8-10s: pouch slides into focus foreground, dog calm behind, lower-third reserved for Buy Now + Subscribe & Save overlay."* ### 8. B2B / SaaS — 16:9 Discovery (Dashboard-to-Motion) *"Seed description: A flat-on-desk composition of a laptop open to a dashboard, a paper notebook beside it with a highlighted KPI, a coffee mug, 16:9 horizontal, soft office daylight, 35mm lens feel, muted palette with a violet accent from the screen. Voiceover hook: The quarterly review that used to take a day now takes one click. Watch the handoff. Animation brief: Wan 2.7 I2V slow top-down push 0-3s toward the dashboard, the KPI number ticks up subtly 3-6s, a hand enters frame and taps the trackpad 6-9s, notebook page turns gently from airflow 9-11s. CTA frame 12-15s: logo overlay space top-left, dashboard glow, bottom-third reserved for Book a Demo Google Ads overlay."* For B2B, the dashboard-to-motion frame is the one that reliably converts on Demand Gen because it lets prospects see the product working without a talking head. Keep any legible text in the UI blurred in the seed and composite real UI recordings over the Wan 2.7 pass in post if label accuracy matters. ## Source Photo Quality Checklist Before you burn credits on a motion pass, run your seed photo through this list. Failing any one item is usually fixable; failing three or more means pick a different seed. - Resolution. Minimum 1920px on the long edge. Recommended 2560px or higher. Below 1920 and the I2V model has nothing to interpolate. - Lighting direction. A clearly readable key (left, right, top, front). Flat lighting produces flat motion — the ad lands lifeless. - Margin / bleed for crop. At least 15% margin on all four sides around the hero subject so the frame survives both 16:9 and 9:16 recropping. - Color accuracy. White-balanced. Wildly color-cast photos confuse the motion model and drift further as the animation progresses. - Model releases for human subjects. Required for any paid ad. Agency shoots should include I2V / AI-motion derivative rights; if the release is old, re-sign. - Packaging legibility. Wordmarks and warnings readable at 100% resolution. Blurry packaging in the seed stays blurry in the motion — often worse. - Subject count. Three or fewer named subjects in the seed. Kling 2.6 and Wan 2.7 I2V both degrade past three tracked entities. - Trademark accuracy. For any trademarked packaging, use the actual production photo. Do not let a generative model invent the wordmark. ## Google Ads Asset Upload: Technical Specs That Matter Google's public specs are more permissive than the practical ones. Here's the working set for 2026: | Spec | Minimum | Recommended | | --- | --- | --- | | Resolution | 720p | 1080p (1920x1080 or 1080x1920) | | Codec / Container | H.264 MP4 | H.264 MP4, ~12 Mbps for 1080p | | Frame Rate | 24fps | 30fps | | Audio | AAC 128kbps stereo | AAC 192kbps stereo | | File Size | Up to 256GB via YouTube | <200MB for quick asset library sync | | Bumper Duration | ≤6 seconds | 6s exact, cut to the frame | | Skippable In-Stream | ≥11 seconds | 15s or 30s | | In-Feed / Shorts | No hard cap | 15-30s | Most I2V clips render at 5-10 seconds. For Bumpers, cut to the frame at 6.0s exactly — Google rejects 6.01s. For 15s Skippable, splice two I2V passes with a voiceover bed over the transition. For 30s, three passes plus a text outro card. ## Credit Math: How Many I2V Ads Can $49/mo Generate? Each finished Google Ads I2V unit is one motion pass plus minor editor assembly (voiceover is cheap; text overlay is free). Using the lower of the two I2V models (Kling 2.6 at ~17 credits) as the working average: | Plan | Monthly Credits | Finished I2V Ads / Month | Use Case | | --- | --- | --- | --- | | Starter — $19 | ~500 | ~12 ads | Solo brand, 3 Google Ads campaigns | | Professional — $49 | ~1500 | ~37 ads | Small DTC, weekly creative refresh | | Ultra — $99 | ~3500 | ~87 ads | Agency running 5-10 client accounts | | Enterprise Max — $299 | ~12000 | ~300 ads | Network or in-house performance team | Full tier math and credit top-ups on the [pricing page](https://aicontentdrop.com/pricing). If your workflow leans on Wan 2.7 I2V for packshot precision instead of Kling 2.6, multiply ad counts by roughly 0.8 to account for the slightly higher pass cost. ## Related Reading on Google Ads Scripting This post is the image-to-video slice of a three-part cluster. For the broader script-writing discipline, start with [how to write a winner Google Ads script with AI Content Drop](https://aicontentdrop.com/blog/write-winner-google-ads-script-ai-content-drop) — it covers hook construction, CTA framing, and placement strategy across all ad types. For text-to-video hero shots when you don't have seed photos yet, see our [Wan 2.7 script templates](https://aicontentdrop.com/blog/winner-google-ads-video-script-templates-wan-2-7). For the craft-deep version of the I2V workflow on a single model, the [Kling 2.6 workflow walkthrough](https://aicontentdrop.com/blog/how-to-write-kling-2-6-image-ad-script-workflow) and the [Kling 2.6 script bible](https://aicontentdrop.com/blog/kling-2-6-script-bible-what-it-rewards) pair well. For ten ready-to-ship Kling 2.6 I2V scripts, the [copy-paste artifact drop](https://aicontentdrop.com/blog/winning-kling-2-6-image-ad-scripts-copy-paste) is the fastest starting point. The broader [image-to-video ads guide](https://aicontentdrop.com/blog/image-to-video-ai-ads) and our [AI product photography piece](https://aicontentdrop.com/blog/ai-product-photography-revolution) fill in the seed-generation half of the pipeline. ## FAQ ### Can I use my existing product photos? Yes, and for most Google Ads use cases it's the recommendation. Your PDP hero shots, lifestyle shoots, and packshots are already optimized for brand fidelity. Run them through the Section 6 checklist first — resolution, margin, lighting direction. If the photos pass, they become your seed library. If they fail, shoot new stills specifically for I2V or generate replacements with Nano Banana Pro / Wan 2.7 Pro. ### What resolution do my seed images need? Minimum 1920 pixels on the long edge; recommended 2560 or higher. Below 1920 the motion model has insufficient detail to interpolate and the rendered clip will look softer than the seed. If your seed is borderline, upscale it first (Nano Banana Pro handles upscales cleanly) before feeding to the I2V model. ### Kling 2.6 I2V vs Wan 2.7 I2V — which to pick? Kling 2.6 for anything with a human subject — faces, lifestyle, reaction shots, walking motion, subtle expression change. Warmer, softer aesthetic that reads well on Shorts and Demand Gen. Wan 2.7 for packshot macros, product heroes, and controlled studio motion — sharper texture, more precise camera control, slightly more expensive. If the ad is 60% face, Kling. If the ad is 60% product, Wan 2.7. Both plug into the same script skeleton. ### Can I keep brand logos readable through motion? Partially. Both I2V models preserve logos acceptably when the logo is stationary in the seed and the motion pass is short (≤6s) and gentle (slow push, subtle rotation, light shift). For aggressive motion or logos in fast-moving areas of the frame, composite the logo back in post from a clean brand asset — most professional editors do this anyway because it guarantees trademark accuracy. ### Can I run these straight to Google Ads Editor? Google Ads Editor imports campaign structure, not video files. Workflow: export the finished MP4 from your NLE, upload to your connected YouTube channel (unlisted is fine), grab the YouTube URL, then either attach the URL to a campaign in the browser UI or reference it from a CSV import via Editor. AI Content Drop doesn't publish directly to Google Ads — the YouTube-as-hosting-layer step is a Google requirement, not a tool limitation. ## Ship the Seed Batch Pull ten product photos that clear the Section 6 checklist. Pick three of the eight templates closest to your category. Write three scripts using the 4-beat skeleton. Generate in the [video studio](https://aicontentdrop.com/best-ai-video-generator) — Kling 2.6 for human-led, Wan 2.7 for product-led — layer voiceover and CTA text overlay in your editor, upload to YouTube, attach inside Google Ads. If you'd rather iterate conversationally, drop a seed into the [Chat-to-Ads Studio](https://aicontentdrop.com/) and ask for three I2V script variations tuned to a specific placement. Three finished 9:16 Shorts creatives cost about 120 credits — roughly one working-day's ration on Professional. The hard work is choosing which seed to animate and which template to apply. The generator handles the rest. ### About the Author This piece was assembled by the AI Content Drop creative team. We ship Google Ads I2V campaigns across apparel, beauty, food, SaaS, and pet verticals every week. Questions about adapting a template to a specific product, placement, or asset library configuration? Reach us through the [Chat-to-Ads Studio](https://aicontentdrop.com/).