For most Shopify stores, video ads come from two places: an afternoon of phone footage and the product photography already sitting in your admin. You do not need a studio or a creator budget to start, you need about six usable clips per product and a way to cut them into several different arguments. Be clear about one thing before you read further: Genyad has no Shopify app and no product-feed connector, so nothing here is about pulling your catalogue in automatically. It is about what to film, what to animate from stills, and what to skip.
What should you shoot with a phone?
Aim for one afternoon, one product, and a shot list you do not deviate from. Six to eight clips of 8 to 15 seconds each is enough raw material for a dozen different ads, because the variation comes from how they are ordered and scripted, not from how many you shot.
The list that consistently earns its place:
- Hands in frame, product in use. The single most useful shot in ecommerce. No face needed. Film it three times: slow, normal, and one very tight version.
- The first-touch or unwrap. Two takes, one from above and one at table height.
- A scale reference. The product next to a hand, a mug or a doorway. Returns and support tickets in small stores are dominated by "it was smaller than I thought".
- Texture at 20 centimetres. Fabric weave, powder, grain, stitching. Lock focus manually or the phone will hunt.
- The problem shot. Whatever the customer's situation looks like before your product exists. Filmed badly on purpose is fine.
- Two or three spoken lines, three takes each. Even if you plan to use a generated voiceover, a real person saying a real sentence is the most reusable footage you own.
Technical settings that actually matter on a phone: lock exposure and white balance before each take (tap and hold, then drag), film in 4K if the storage allows so you can reframe from 16:9 to 9:16 without softening the image, turn the grid on, and put the light source in front of the product rather than behind you. Overhead kitchen spotlights are the most common reason phone footage looks cheap. A window at 45 degrees fixes almost everything.
Frame every shot so it survives a crop: keep the product in the middle third with headroom, so the same clip works at 9:16, 4:5, 1:1 and 16:9 without decapitating anything.
Which gaps should you generate, and what does it cost?
Some shots are not worth filming and some are impossible with a phone. Those are the ones to generate. Animating an existing product photo into a short motion clip costs 2 credits per second in Genyad, so this is a decision with a price attached and it should be made shot by shot.
| Gap in your footage | Best fix | What it costs |
|---|---|---|
| Product has photography but no video at all | Animate two hero stills into 3 to 4 second motion clips | 6 to 8 credits per clip |
| Need a slow push in on the packaging | Animate the pack shot you already have | 2 credits per second |
| Need a clean rotating product on white | Film it. A turntable is cheaper than generation | Nothing but your time |
| Product in a setting you cannot access (beach, snow, restaurant) | Generate a short establishing clip | 2 credits per second, use 3s not 8s |
| Ten colourways of the same item | Photograph them, animate one, reuse the animation logic | 1 animation, then stills in the edit |
| Person using the product | Film a friend, or commission a creator | Genyad has no avatars, so generation is not an option here |
The rule that keeps the bill sensible: generate establishing and hero motion, film everything with a human in it. A 4 second animated clip at 8 credits is good value for a hero product. Doing the same across 60 SKUs is not, and the pricing page makes that arithmetic easy to check before you commit.
What does a first test set look like on the free plan?
The free plan gives 5 video ad variations plus one AI-generated video of up to 10 seconds, with no card and no watermark. That is genuinely enough to answer one question: does our footage work as an ad at all.
Spend the five variations like this rather than making five versions of one idea:
| Variation | Hook | Body | Purpose |
|---|---|---|---|
| 1 | Problem stated out loud | Product in use, then result | Baseline. Most stores' best performer |
| 2 | The result first, in reverse | How it happened | Tests whether the payoff can carry the open |
| 3 | Price or unit comparison | Product reveal | Tests whether you compete on value |
| 4 | The scale or texture close-up, no words for 2 seconds | Spoken claim over use footage | Tests a visual hook against a verbal one |
| 5 | The objection named directly (sizing, shipping, "does it actually work") | Proof, then offer | Usually your best retargeting ad |
Five files, four different arguments. Run them for a week at whatever your normal daily budget is, look at three second view rate and click-through rather than at sales, and you will know which argument to build the next fifteen around. There is a fuller version of this workflow on our ecommerce video ads page, and the image to video ad generator page covers the animation side.
If the test works and you want a real rotation, Starter is €29 for 15 credits and Growth is €99 for 65 credits, with one standard variation costing one credit. Credits do not expire, which matters for a seasonal store that runs hard for six weeks and then goes quiet.
What should a small store skip?
Skip the brand film. Skip the logo animation. Skip the second studio day. Skip 30 second cuts, which no prospecting audience on Reels or TikTok will finish.
Also skip the instinct to give every SKU its own video. At a small store's budget, three or four category-level arguments carry the long tail better than 40 thin per-product ads, and they fatigue more slowly because each one is seen by fewer people.
On which note, refresh cadence depends on what you sell. Our 2026 fatigue benchmark, a synthesis of published platform and agency figures rather than our own experiment, puts days to a 40 percent CTR decline at 9 for food and beverage, 12 to 14 for fashion, about 18 for beauty and DTC, and about 21 for electronics. A coffee subscription store and a lighting store shooting the same afternoon still need different production rates afterwards.
What Genyad does not do for Shopify
There is no Shopify app, no product-URL import, and no product-feed or CSV-driven template rendering. You cannot point us at a collection and get 200 ads back. Uploading footage is manual, and that is the honest shape of the tool: it turns a library of video you already own into fresh variations, each with its own script, shot selection, voiceover, captions and export.
If catalogue-scale automated rendering is the actual requirement, a feed-driven template renderer is the correct category and Plainly does that properly. If you want generated social posts built from a product feed, Predis.ai is closer to that. We also do not publish to Meta or TikTok directly, do not produce static banners, and do not score creative against a predicted performance model.
Frequently asked questions
Does Genyad integrate with Shopify?
No. There is no Shopify app, no product-URL import and no product-feed or CSV rendering. You upload footage and product photography yourself, and Genyad builds ad variations from that library. For catalogue-scale automated rendering from a feed, a template tool like Plainly is the right fit.
Can I make video ads from product photos alone?
Yes. Stills can be animated into short motion clips at 2 credits per second, so a 4 second clip costs 8 credits. It works well for hero products and pack shots. It is not a substitute for footage of a person using the product, which is still what earns the first two seconds in a cold audience.
How many video ads does a small Shopify store need?
Fewer than a DTC brand running heavy prospecting. Start with five variations covering four different arguments, then build out to 8 or so once you know which argument works. A small audience reaches fatiguing frequency slowly, so volume matters less than argument variety.
What can I actually test for free?
The free plan includes 5 video ad variations and one AI-generated video of up to 10 seconds, with no card required and no watermark on the export. That is enough to run a week-long test of four distinct arguments before deciding whether to buy credits.