Product launch video: one recording, every launch-day cut

A launch is one date, and everything ships on it#

Distribution is normally something you can stagger. Post the YouTube walkthrough this week, cut a LinkedIn clip next week, add the docs loop when a ticket asks for it. A launch takes that slack away. On launch day the Product Hunt post, the announcement page, the email blast, the social thread, and the changelog entry all go live inside one window, each pointing at the others, and the urgency you are trading on, the "it's live today" and the upvotes and the reshares, does not reschedule if you slip.

That compression is the actual problem. Counted as separate productions, a launch kit is five or six scripts, recordings, voice tracks, and edits, all due the same morning, which is how a team ends up at 11pm with three of the six finished. The evergreen distribution case, one recording spread across many channels, still applies, but a launch adds two things a slow rollout never has to handle: a deadline that cannot move, and a set of cuts that must agree with each other because they are all on screen at the same time. The way out is the ordinary distribution move made stricter: one recording, cut per launch surface, planned before you hit record instead of assembled after.

The launch-day cut list: what goes live where, and how long#

A launch does not need every surface a product ever uses. It needs a specific, smaller set, the surfaces that actually carry a launch, and each wants the recording in a particular length and shape. Here is that kit as one spec.

Four rows carry a spec worth pinning down. The Product Hunt gallery does not accept an upload; it takes a public YouTube URL, so its video inherits YouTube's encoding: an MP4 in H.264 High Profile, AAC at 48kHz, 16:9 and 1080p at roughly 8 Mbps². That is the identical 16:9 file behind the hero and the deep cut, which is why the Product Hunt cut is close to free once the flow is recorded. The social clip has to work muted and in-feed: LinkedIn accepts 16:9, 1:1, 4:5, and 9:16, and puts the sweet spot at 15-30 seconds, short enough to clear every feed placement³, which is why the vertical and square framings get their own week-of-launch treatment. Length is a retention decision, not a house style: Wistia's data puts videos under one minute at about a 52% average engagement rate and product videos near 50%, with 1-5 minute explainers still watched past the halfway mark⁴, so the hero loop stays under a minute and the deep cut spends its extra minutes only where a viewer chose to watch.

The last row is the one that breaks the "one recording feeds everything" rule, and it is worth naming. Hacker News takes no video. A Show HN is a link to "something you've made that other people can play with," ideally with no signup wall, and blog posts, sign-up pages, and landing pages are explicitly off topic⁵. If Hacker News is on your list, the deliverable there is a URL that works, not a cut of the film, a clean reminder that a launch kit is a set of surface-specific assets rather than one file pasted six places.

The run-of-show: choreographing the cuts across 24 hours#

The part no "product launch video examples" roundup gives you is the order. A launch is not a pile of clips published whenever each finishes rendering; it is a timed event, and the assets fire in sequence. Product Hunt runs its leaderboard on a 24-hour cycle in Pacific time, and its own guide is blunt that "posting at 12:01am PST gives people a full 24 hours" on the board¹. That single line anchors the day: the Product Hunt cut has to be public before midnight Pacific, and everything else keys off it.

Here is the kit mapped to the clock for a US-timed launch.

Writing this down matters because every asset in the middle column points at another: the email drives traffic to the Product Hunt page, the LinkedIn clip links the announcement, the recap thanks the people who upvoted. If a feature gets renamed or a price changes between the day you recorded and the day you publish, that inconsistency is now visible on six surfaces at once, the exact failure a slow rollout never risks. Product Hunt lets you schedule a launch up to a month ahead, which is enough runway to record the master and cut the kit without panic¹, as long as the kit can absorb a late change without a reshoot.

The 11pm change is the whole argument for a re-renderable kit#

Every launch has one. A pricing tier gets a new name, legal wants a word off the plan card, a beta badge comes off a feature the night before the date. In a kit of separately recorded clips, that one edit lands on all six deliverables at once: re-record the hero, re-cut both social versions, re-shoot the Product Hunt walkthrough, rebuild the email GIF, re-render the YouTube upload, and re-sync every caption track, at 11pm, against a date that will not move for you.

The fix is to stop storing the demo as footage and start storing it as instructions. When the flow lives as a committed spec, the narration and the click targets and the pacing all held as text, the late change is a one-line edit and every deliverable rebuilds from it, because the six cuts were only ever the same source rendered at different sizes and lengths, each trimmed to what its placement rewards. aidemo, our own open-source engine, works exactly this way: a coding agent writes the spec, the engine replays it in a real browser, and the midnight rename becomes a re-render instead of six reshoots. Its honest boundaries: it drives a browser and nothing else, there is no timeline you drag clips around on, and the spec is agent-written rather than clicked together by hand in an editor. That makes it a poor pick for a one-take screencast and a strong one for a kit that has to absorb a change arriving hours before it ships.

What a launch cut carries that an evergreen demo doesn't#

A launch cut and a standing product demo come off the same recording, but they open differently, because they answer different questions. An evergreen demo answers "what is this, and why should I care." A launch cut answers "what is new, and why now." That reorders the first five seconds: the launch cut leads with the capability that did not exist last week rather than the product's name or a logo card, and it carries a dated ask, "live on Product Hunt today" or "shipping this week," that a permanent demo cannot make, because a permanent demo has no today. The playbook for the underlying master is the same either way; the launch is a framing and an ask laid over it.

That timestamp is also the launch cut's expiry date, and planning for it saves money. The social clips and the email GIF are disposable by design: they did their job inside the launch window and can be archived after. The survivors are the hero loop and the deep walkthrough, and if you strip the "today" framing they become the standing demo on the marketing page and the evergreen feature-announcement clip for the next release. Build the kit so the durable cuts outlive the day and the disposable ones were cheap enough that their short life costs nothing. That is the line between a launch that costs one recording and a launch that costs six.

Sources#

1. Product Hunt - Preparing for your launch (24-hour Pacific cycle, 12:01am timing, scheduling, video guidance)
2. YouTube Help - Recommended upload encoding settings
3. LinkedIn Help - Video ad specifications (aspect ratios and length)
4. Wistia - What's the optimal video length? (engagement by length)
5. Hacker News - Show HN guidelines

FAQ#

What should a product launch video include?#

Lead with what is new, not with what the product is: the first five seconds should show the capability that did not exist last week, then the one workflow that proves it, then a dated call to action such as "live on Product Hunt today." Keep the launch-day hero loop under a minute, burn in captions because feed and gallery views start muted, and end on a single ask rather than a menu of links.

How many videos do you need for a product launch?#

One recording, cut several ways, not several videos. A typical launch kit is a sub-60-second hero loop for the announcement page, a Product Hunt gallery cut (a YouTube link), a 15-30 second vertical and square social clip, an email GIF, and a 1-3 minute deep walkthrough for YouTube. They come off the same captured flow at different lengths and aspect ratios, so the marginal cost of the fifth cut is a re-encode, not a reshoot.

What is the difference between a launch video and a regular demo video?#

A demo video is evergreen and answers "what is this and why should I care"; a launch video is time-boxed and answers "what is new and why now." The launch cut leads with the just-shipped capability, carries a dated ask tied to the launch surface, and has a short shelf life: after the launch window the disposable social clips are archived and only the hero loop and deep walkthrough live on as the standing demo.