How long should ad music be: as long as the ad, and every length costs the same

Cut the ad first. The runtime is a result of the script and the pace, not something to decide in advance for the music's benefit.

Then pick the nearest length at or above it. A 28 second ad takes the 30 second track.

The price is 50 credits at every length, so a 15 second track is not a saving and a 120 second one is not better value.

There is no trimming inside the module, so generating long and cutting is a job for whatever tool you finish the ad in.

A track written for the length it plays at ends properly. A trimmed one stops, and a listener notices a stop.

That is the whole argument for matching, and it costs nothing to follow.

Music written for thirty seconds arrives at thirty seconds. It resolves, and the ad feels finished.

A sixty second track faded out at thirty does not resolve. The listener registers an interruption even when they cannot say what happened.

On a direct response ad the last two seconds are where the instruction lives, and an unresolved bed under an instruction weakens it.

Fades are the usual workaround and they are a compromise rather than a fix.

Since the price is flat, the compromise buys nothing, which is unusual and worth taking advantage of.

Generate the length you need. If the ad changes length later, generate again for another 50 credits.

Two minutes is the longest single track available, so anything longer has to be assembled outside the product.

That makes this module right for ads, intros, stings and shorts. Long-form background is a different purchase.

If you upload weekly long-form video, a subscription library will be cheaper and less work than joining generated pieces.

For a 9:16 ad the ceiling is irrelevant, because almost no ad in a vertical feed should be approaching two minutes anyway.

The practical range for social ads is 15 to 60 seconds. Most of what gets made lands between 20 and 40.

So four of the five lengths cover almost everything, and the fifth exists for the occasional long piece.

A sung line carries roughly sixty words a minute, against a hundred and fifty spoken.

So a 30 second sung track is about thirty words, and pasting more than that means lines get rushed or dropped.

Count the words before generating. That check is free and it prevents the most common disappointment with sung returns.

For an instrumental, leave the lyrics box empty and the word budget stops mattering.

If the ad has a voice-over, the track should be instrumental anyway. Sung and spoken words compete for the same attention.

Two tracks for one ad, at 50 credits each, is a normal and cheap answer to that problem.

So a sung track has two ceilings running at once: the seconds you picked, and the words those seconds can hold.

The second ceiling is the one people hit. A paragraph looks short on a page and long in a mouth.

Read the words aloud at singing speed and time yourself. That measurement beats any word count rule you could memorise.

Decide the length question once and stop revisiting it. That is the useful discipline for anybody making ads weekly.

Cut the ad, read the runtime off the export, take the nearest length at or above it. That rule handles every case.

It removes a decision from the week, which is worth more than it sounds when you are shipping four ads.

The one exception is a sung track, where the word budget rather than the runtime is the binding constraint.

Two ads want no track at all. A wall-to-wall voice-over with no pauses has nowhere for music to sit, whatever length you pick.

A demonstration where the product makes a sound is stronger with the real audio than with a bed over it.

Play the cut muted, then voice only, before you spend anything. If voice only holds, the music is optional.

Four plain limits sit around all of it. No editing after generation, no stems, no tempo slider and no trimming.

A track longer than two minutes is not available here. A library is the right purchase for that.

The track is generated for you rather than licensed from a catalogue. Confirm the usage terms that apply to your account before running one behind paid spend.