Turn product screenshots or a product page into a product video that explains a feature or a workflow, with captions, voiceover and music.
Free to preview and edit. You only pay when you export.
0/5000
Feature launches, release updates and first introductions built from product screenshots with template motion. Select a poster to play the video.
16:9 · 19s
A hook, two zoomed screenshot scenes with a focus crop, a feature list and a closing step.
16:9 · 13s
A problem scene, one representative workflow and a closing step for viewers who have never used the product.
16:9 · 10s
Names the change, shows where to find it and the new result, without repeating the full pitch.
9:16 · 8s
A tight crop on one control, with a short title and caption sized for a phone screen.
1:1 · 9s
One feature, one result and one next step in a square frame.
16:9 · 9s
Opens on the problem, introduces what is new and shows a benefit supported by the supplied materials.
Sample videos are composed from screenshots and template motion. They are arranged demonstrations, not screen recordings.
The same source image before and after it becomes a scene. The frame adds a caption, a focus crop and a cursor highlight; the picture in it is the source.
The screenshot supplies the interface. Control names, labels and the visible state come from the source, not from the video. Check that the frame still shows the state the caption describes.
A caption, a focus crop on the key control, a cursor highlight and the motion between scenes. These guide attention. They do not add behavior the screenshot does not show.
Whether the label is readable at the size you will publish, whether the crop shows the state the narration describes, and whether the cursor motion implies a click you never captured. Call an arranged sequence composed, not recorded.
The audience decides how much context the opening needs. Choose the message first, then the ratio and length that fit where the video will play.
16:9 · 19s
Start with the task the feature changes, then show the entry point, the feature state and the result. Three well-chosen screenshots tell a clearer story than a tour of every menu. Give each scene one job, and confirm the final message matches the version people can use when the video is published.
16:9 · 13s
For viewers who have never used the product, open with the problem, show one representative workflow and end with a single invitation to continue. Pick screenshots that reveal the structure of the product without a full tour, and keep screen text short enough to read at the pace of the video.
9:16 · 8s
Existing users need what changed, where to find it and how it affects their work, not a new pitch. Crop the screenshot to the changed control so it stays readable in a tall frame, and say so if the change depends on a plan or rollout, so the video does not imply universal access.
Three steps in the creator above. The project quote is shown before you export.
Attach up to 8 images, documents or audio files, or paste a URL. Video clips are not accepted. Say who will watch, what they should understand and the one action to take, and put exact names, dates and numbers in the brief.
Pick 16:9, 9:16 or 1:1 and a target of 15, 30, 45, 60 or 90 seconds. The default is 30 seconds in 16:9 without narration. The target is a budget: narration and revisions can change the final length.
Check the claims in the outline and storyboard before you judge transitions, then export a 1080p MP4. The base rate is 5 credits per second, so 30 seconds starts at 150 credits before narration and illustrations. The project quote shows the exact amount. Download the file and play it to the end.
Your screenshots stay the evidence. Potto adds the crops, captions, motion and formats around them, and you review every scene before you export.
01
Attach up to 8 images, documents or audio files, or paste a URL. Each screenshot becomes a scene, so the interface in the video is the interface you supplied, in the state you captured.
02
A focus crop zooms to the control that matters, and a cursor highlight marks where to look. They guide attention without adding behavior your screenshot does not show, so crop to the changed control and let the caption name it.
03
Pick the shape for where the video will play: 16:9 for a site or a channel, 9:16 for stories and vertical feeds, 1:1 for feed posts. After you change the shape, check that the crop still shows the control at the new size.
04
Read the outline and storyboard against your source, and fix any claim that does not match before you judge the transitions. The project quote shows the cost, with a base rate of 5 credits per second, before you export a 1080p MP4.
A product video starts with clean screenshots and ends as a finished video. These pages cover the tasks on either side.
What a product video is, which materials it accepts, and what to check before you publish it.
A product video explains a product, a feature or a workflow to a specific audience. Potto AI makes it as motion graphics: your screenshots and other materials move through template-based scenes with captions, optional narration and music, rendered as an MP4. Choose one product message and check that your sources support it.
A product launch video shows how a product or feature works. Potto AI can arrange screenshots into an explanatory sequence, but that is not automatically a continuous recording of the interface. Use genuine screen recording when exact interaction behavior or timing needs to be demonstrated as evidence.
Screenshots are the core material: add images of the screens you want to show, plus documents or audio files, or paste a product page URL, with up to 8 attachments in one request. An existing video clip is not supported. A URL supplies source material, not permission to invent features the page does not show, so put exact names, prices and numbers in the brief.
Use current screenshots that show the relevant state, crop out unrelated or private information, and identify the control or change the viewer should notice. Verify feature names and availability. Build the video around one message the source supports, such as a feature's starting state, action and result, and name the audience and the next action in the brief.
Choose 16:9, 9:16 or 1:1, and a target of 15, 30, 45, 60 or 90 seconds. The default is 30 seconds in 16:9 without narration. The target is a budget, so narration and revisions can change the final length. After changing the shape, check the text and screenshot crop at the new size: a desktop screenshot can become too small in a tall frame or a narrow page column.
You can add screenshots, a link or a brief and pick the format before signing in. Starting the product video opens the sign-in dialog, and your sources and settings stay in the workspace.
Drafts, outlines and storyboards are free to redo. If a render or export fails, the project shows the reason and you can retry; credits are only charged for a finished export. Use the export cancel action to stop work in progress.
Planning, storyboards and chat edits are free. Exporting costs 5 credits per second, so a 30-second product video is 150 credits. The export button shows the total before you confirm.
No. Potto AI arranges your screenshots with template motion, so cursor highlights and transitions guide attention but are not a capture of a live session. Review the downloaded MP4 from start to finish, and use a real screen recording when exact interaction timing must be shown as evidence.
Add your screenshots or a product page, choose the format, and review the export quote before you render.