# The Teaser Schedule That Actually Builds > Teasers fail by revealing too little for too long. The four-post arc across three weeks, and how much of the song each one should give away. [All guides](/blog) # The Teaser Schedule That Actually Builds Published 27 August 2026, 00:00 UTC · updated 12 August 2026, 21:30 UTC · by Loopin Most teasers withhold. That is backwards — anticipation comes from wanting more of something you already like, not from being shown less of something you have not heard. ## Give away more each time, not less Post one: eight seconds. Post two: fifteen. Post three: thirty, including the chorus in full. Post four: the whole first minute if it is good. By release day people should already know how the song goes. The fear is that people will not stream a song they have heard. In practice, familiarity is what makes them press play — the song that gets skipped is the unfamiliar one, not the one they half-know. ## Day −21: the eight seconds, with no context The loudest bar, a text hook, no mention of a release. This post is a test, not an announcement: you are finding out whether the section works while there is still time to pick a different one. If it underperforms, post a different eight seconds three days later. Two or three of these before you commit is the cheapest audience research available. ## Day −14: fifteen seconds and the first hint Extend the winning section and add the date. Still no artwork, still no link — one new piece of information per post is the pace that keeps people reading. This is also when the pre-save should exist, mentioned once at the end rather than as the point of the post. ## Day −7: thirty seconds, artwork, and the story The full chorus, the artwork, the date, and one specific sentence about the song. This is the anchor post of the campaign — the one people screenshot and the one you point back to. Everything before it was testing. Everything after it is repetition of what worked, which is exactly as it should be. ## Day −2 and −1: repeat the winner No new material. Post the strongest clip again with the date on screen. Almost nobody saw it the first time, and the people who did are not annoyed — they are the ones who already like it. Then keep going after release. The teaser arc is three weeks; the campaign is six. Most artists stop at the moment the song finally exists, which is when most people are first able to act on it. ## Frequently asked questions ### How long before release should I start teasing? Three weeks. Eight seconds at day -21, fifteen at -14, thirty with the artwork at -7, then repeat the strongest clip at -2 and -1. ### How much of the song should I give away? More each time, ending with the full chorus or the first minute. Familiarity is what makes people press play — the song that gets skipped is the unfamiliar one. ### Should I post the same clip more than once? Yes. Almost nobody saw it the first time, and the people who did are the ones who already like it. Repeating the winner beats inventing new material nobody has tested. ## Related reads - [The 30-day pre-release timeline](/blog/the-30-day-pre-release-timeline) - [How to pick which 8 seconds to post](/blog/how-to-pick-the-clip-that-goes-viral) - [How to use a countdown properly](/blog/how-to-use-a-countdown-properly) --- Source: https://onloopin.com/blog/the-teaser-schedule-that-works