Your product is done. The demo video is the only thing between you and launch.

Why the video is where launches get stuck

You can build a product on four platforms alone and still freeze in front of a demo. It is not a skill problem, it is a category problem: the tools you know are for making software, and the tools for making video were built for people who edit video all day. So you look for someone to do it. The agencies quote a price that makes sense for a company and none for a side project, the freelancers on Fiverr want a brief you do not know how to write, and the AI generators you try first invent screens that are not your app. Weeks pass. The product is still done.

The way out is to lower what "demo video" means. A demo is a screen recording of the real product with the boring parts removed and the important parts made readable. That is all. The cinematic feel, the zooms, the clean cursor, the captions: that part is now done by software while you record.

Step one: three lines, not a script

Open a text file and write three sentences. What hurts today. What your product does about it, as a verb. What the person has when it is done. For an invoicing app: "Chasing late invoices by hand eats a morning a week. Here you set the reminder once, and it follows up for you. You get paid, and the morning back." Read it out loud. If it sounds like a person, keep it. If it sounds like a website, make it shorter.

You do not memorize it. You look at the three lines while you record and you say them in your own words. The pauses and the "so, here" are fine, and the ones that are not fine get cut later.

Step two: a clean screen

Close everything that is not the product. Turn on do-not-disturb so a message does not pop up over your pricing page. Set the browser zoom to 110 or 125 percent so text is readable before any zoom. Put the app at the exact starting point of the story, and log in as a user with real-looking data, not "test test 123".

In Ciakko, pick what to record: the whole screen, one window, or an area you draw. For a web app, the browser window without the taskbar is usually right. Add the webcam bubble if you want your face in the corner. You can move it or remove it after recording, so the decision is not final.

Step three: record it once, mistakes included

Press record, wait for the countdown, and walk through the three lines while you click through the product. Click deliberately: every click is where the zoom will land. Type what a real user would type. When something goes wrong, do not stop. Say "let me do that again" and do it again. That sentence will be gone in two minutes.

Two minutes of raw recording is plenty for a sixty-second demo. If you went past five, you are giving a tour, not a demo. Pick the one thing and record again.

Step four: the editor opens with the work done

Stop the recording and the editor opens with the zooms already placed on the moments you clicked and typed. This is the part that used to be the afternoon. Now it is a check: scrub through, watch the camera land on the right things, drag a block if it framed the wrong button, delete one if it zoomed on nothing. Press a number key to set the zoom level. Two minutes.

Then the cuts. Split at the "let me do that again", delete the part before it. Let the app speed up the stretches where nothing happened. Pick a background, a color or a gradient, round the corners. If you spoke, let it write the captions, and click any word it got wrong.

Step five: export for where the launch happens

Wide at 1080p for the landing page and YouTube. Vertical for Shorts and Reels, from the same project, with the camera keeping the action in frame. Square for a feed post. The export runs on your GPU and the file is yours: upload it to your own site, embed it in the README, attach it to the Product Hunt launch. No link that expires, no logo in the corner.

What this will not be

It will not be a motion-graphics piece with a 3D phone spinning over a gradient. Those cost what they cost for a reason, and for a launch they are usually the wrong thing anyway: people want to see the product, not an animation of it. A demo of the real product, made by the person who built it, with the camera following the clicks, is what the good launches actually use. Watch the top ten on Product Hunt in any week and count.

The cost of doing it this way

Ciakko is $89, once. The trial is the whole product for 14 days without a card and without a watermark, which means you can make this launch video before deciding whether to pay. We wrote down what the other routes cost on the demo video cost page, with the numbers founders posted, so you can compare with your eyes open.