You have a product photo. Maybe it is a listing shot. Maybe it is something a founder took on a kitchen counter with a phone. Either way, it is sitting flat in a folder while your competitors run video.
Video ads are not optional anymore. On YouTube, Shorts placements now average a 1.24% CTR, ahead of skippable in-stream at 0.65% and non-skippable at 0.21% (Digital Applied, "YouTube Ads Benchmarks 2026," published 2026). Static image feeds are the format everyone already has. Video from one photo is the format most small and mid-size advertisers still skip, because they think it needs a shoot.
It does not. This is a step-by-step walkthrough of turning a single product photo into a launch-ready video ad, using a model that is live on the Seedance 2.0 marketplace pipeline today. Every prompt below is the exact prompt used. Every output is the exact output it produced, not a mockup.
What you need before you start
- One clean product photo. Plain background, good lighting, product fills most of the frame. A phone photo works if it is in focus.
- A one-line description of what the product does.
- Five minutes.
That is the whole materials list. No camera, no actor, no editor.
Source photo used for every example below: a ceramic pour-over coffee dripper on a plain cream backdrop.
Step 1: Generate the raw video from the photo
The model doing the work here is Seedance 2.0, ByteDance's text-and-image-to-video model, callable directly through the Coinis marketplace pipeline in image-to-video mode. It takes your still photo plus a motion prompt and renders a short cinematic clip. It does not just pan the camera. It adds real physical motion (steam, light shifts, product rotation) that a photo alone cannot show.
Product used for this walkthrough: a ceramic pour-over coffee dripper on a plain cream backdrop.
Prompt used:
"Slow orbit around the product, soft steam rising from a coffee cup beside it, warm morning light shifting across the ceramic surface, shallow depth of field, 6 seconds, cinematic product commercial style, no text overlays."
What this demonstrates: Seedance reads the still image for product shape and material, then invents physically plausible motion around it (the steam, the light shift) rather than a static push-in. This is the difference between a photo with a Ken Burns effect and an actual video ad.
Settings: 720p, 9:16, 6-second duration, image-to-video mode, source: the single product photo above.
Generated output above: real Seedance 2.0 render from the exact prompt, no edits.
A second pass, same photo, different motion prompt, shows the range available from one source image:
Prompt used:
"Product sits still center-frame, camera locked off, background softly blurs from cream to warm amber as if the sun is rising behind it, single ray of light sweeps across the ceramic surface left to right, 6 seconds, no camera movement, minimalist product film style."
What this demonstrates: you do not have to move the camera to make a photo feel alive. A locked-off shot with light and background motion alone reads as premium and is often the safer first ad variant to test, since it does not risk a weird-looking orbit artifact on a tricky product shape.
Generated output above: real Seedance 2.0 render from the exact prompt, no edits.
Step 2: Skip the manual edit, run it through an Ad Studio template instead
Raw video output is a clip. An ad is a clip plus hook copy, captions, pacing, and a CTA, packaged to the spec of the platform you are running it on. That assembly work is exactly what Coinis Ad Studio templates do in one pass, without you touching an editor.
Two Ad Studio templates apply directly to a single product photo:
Product-shot template (cinematic preset): feed in the same product photo, the template runs the generation and assembles it into a finished product-commercial-style ad, matched to spec, ready to upload. This is the direct next step after Step 1 above. It is not a separate model. It is the templated version of the same generation pipeline wrapped in a launch-ready structure.
UGC-ad template: same product photo, different output shape entirely. Instead of a cinematic product shot, this template generates a creator-style clip: a presenter holding or reacting to the product, on-screen captions in kinetic or karaoke style, paced like a native TikTok or Reels post rather than a commercial. UGC-style creative consistently gets tested against static ads right now because audiences respond to the creator feel over the polished-spot feel, particularly on Reels and TikTok (RevenueCat, "UGC Ads for Apps," 2026).
The point of running both: you are not choosing between the AI model and the ad platform. Marketplace access gets you the raw generation. Ad Studio gets you from that generation to a file you can actually launch, with the hook, captions, and platform spec already built in. That packaging step is what a bare model API does not do for you.
Step 3: Pick your test set, not just your favorite
Do not ship one video ad. Ship three: the orbit version, the locked-off version, and the UGC-style version from the template. Creative fatigue sets in fast on any single winning ad, and having variants ready before you launch, not after CTR drops, is the difference between a proactive refresh and a reactive scramble. See our guide on combating creative fatigue on Instagram ads for the rotation cadence that keeps CTR from decaying.
If you are running this specifically as a Facebook or Instagram video ad, match the clip length and caption placement to platform spec before you launch. Our Facebook video ad guide and Shopify-specific video ad walkthrough cover the spec details that make or break approval.
FAQ
Do I need a professional product photo to start?
No. A clean, well-lit phone photo with the product filling most of the frame works. The model reads shape and material from the image. It does not need studio lighting, but blurry or cluttered photos produce worse motion.
How long does generation take?
A single 6-second clip from Seedance typically completes within a couple of minutes end to end (image-to-video is an async job with a short poll window). Running the same photo through an Ad Studio template adds the assembly step but not a meaningfully longer wait.
What is the difference between generating directly and using a template?
Direct generation gives you a raw clip: the model, the motion, nothing else. A template runs that same generation and then assembles it into a launch-shaped ad: captions, pacing, and format matched to the platform, so what comes out the other end is something you can upload immediately, not something you still need to edit.
Can I use this for more than one product photo?
Yes. Once you have a working prompt structure for one product, reusing the same motion-prompt pattern (orbit, locked-off, or UGC preset) on a new photo is the fastest way to build a variant set across a catalog without writing every prompt from scratch.
Does this replace video ads that use real actors?
Not always, but for a first test of video-format viability on a product that currently only has static creative, it removes the cost and turnaround barrier entirely. Many teams use it exactly that way: test the format cheaply here, invest in a full shoot once you know video outperforms static for that specific product.
Turn your product photo into a launch-ready ad today
You do not need a production budget to find out if video beats your current static ad. One photo, one prompt, and an Ad Studio template gets you from a still image to a launch-ready video ad, on the same platform where you can also generate the UGC or cinematic version to test side by side. Run yours on Coinis Ad Studio and see which version wins before you spend a dollar on a shoot.
Sources
Digital Applied, "YouTube Ads Benchmarks 2026: CPV, CPM, CTR by Industry," 2026. https://www.digitalapplied.com/blog/youtube-ads-benchmarks-2026-cpv-cpm-ctr-industry
RevenueCat, "UGC Ads for Apps: Which Formats Drive Conversion in 2026," 2026. https://www.revenuecat.com/blog/growth/ugc-ads-apps/
Isidora Matovic
Author
Social media enthusiast and a full time researcher. She takes digital presence very seriously and that is why you are always in touch in what is going on with us! Follow us for more posts like this.