Short answer: As long as the argument needs and no longer. We write to four lengths: an ADHD Cut of 10 to 15 seconds for simple, impulse products and ready buyers, Standard DR of 20 to 35 seconds as the default, Deep Narrative of 40 to 55 seconds for skeptics and stories, and Long Form of 60 to 180 seconds for considered purchases and narrative formats. The right length comes from the price, the awareness stage and how much belief the viewer still needs.
"What length should our ads be?" is one of the most common questions we get, and the usual answers are rules of thumb about attention spans. The better question is: how many beats does this argument need, and how fast can a real person say them? Length is one of the picks in the Script Vibe system. When it is not picked, the system defaults to Standard DR and adjusts only when the idea needs it.
The brands in the examples are made up.
What determines the right length for a video ad?
Length is set by the number of beats the viewer needs before they believe, multiplied by how long each beat takes to say.
The problem: Teams pick a length first ("everything is 15 seconds") and then cram or pad the script to fit. A crammed script rushes the proof and the viewer does not believe it. A padded script repeats itself and the viewer leaves.
The solution: Count the jobs the script has to do. A buyer who already wants the product needs one reason and the offer: two beats. A skeptic who has tried three fixes needs the failed attempts, a realization, a mechanism, a first win and a transformation. That is not a 15 second job.
Three inputs decide it:
- Price and consideration. A low price, low risk impulse product can be decided in seconds. A considered purchase needs the viewer to think through an argument, which takes time. See peripheral vs central route.
- Awareness stage. The earlier the stage, the more ground to cover. A Most Aware viewer needs the offer. An Unaware viewer needs the pain, the cost, the mechanism and the proof first. See entry stage.
- Framework. Each framework has a natural length. Jake 1.0, 1.1 and 3.0 run about 35 seconds; Jake 2.0 runs about 55. Full beats are in the Jake ad script frameworks.
How many words fit in each length?
Plan on about 2.5 spoken words per second for a direct response read, and about 2 per second for a calm read.
| Length option | Seconds | Words at 2.5 per second | Best for | Usual framework |
|---|---|---|---|---|
| ADHD Cut | 10 to 15 | About 25 to 37 | Impulse products, Most Aware offers, one reason ads | Jake 1.0 |
| Standard DR (default) | 20 to 35 | About 50 to 87 | Most ads, most stages | Jake 1.0, 1.1, 3.0 |
| Deep Narrative | 40 to 55 | About 100 to 137 | Skeptics, saturated markets, testimonial stories | Jake 2.0 |
| Long Form | 60 to 180 | About 150 to 450 | Considered purchases, founder content, story formats | 2.0, full funnel one shot |
The word count is the most useful number on the page. If a writer hands you a "30 second script" with 140 words in it, it is a 55 second script, and either the creator will rush it until it sounds like a disclaimer, or the editor will cut it in a way nobody planned. Count words before anyone films.
What is an ADHD Cut?
An ADHD Cut is a 10 to 15 second ad built on one reason: a peripheral cue that stops the thumb, a single clear central beat, and a close. It is a full ad, not a trailer.
It works when the decision is small. Nightfold, a $38 weighted sleep mask that blocks the light leaks normal masks let in at the nose, fits.
0 to 3s: "This is my bedroom with every light on." 3 to 6s: (mask on, screen goes black) "And this is with Nightfold." 6 to 11s: "Normal masks leak light at the nose. This one seals there, and the weight just settles you." 11 to 14s: "$38, and I'm keeping it."
About 35 words. One visual cue, one reason, the price, a close. Notice what is missing: no backstory, no agitation, no transformation beyond a single line. For an impulse product to a warm viewer, that is enough. For a skeptic, it is not.
The ADHD Cut fails when the product needs belief it does not have time to build. A $289 purchase in 15 seconds usually reads as a pitch.
What is Standard DR length?
Standard DR is 20 to 35 seconds and our default, because it gives a script room for every rung of the transformation ladder: feature, benefit, outcome and identity.
At 35 seconds you have a hook, a handoff line that makes it personal by about 5 to 8 seconds, a problem with real stakes, the mechanism, a piece of proof and an earned close. That covers Problem Aware viewers, most Product Aware viewers and, with Jake 3.0, cold Unaware viewers. Every example in the frameworks article except Jake 2.0 is a Standard DR script.
A good test: read the script out loud at a natural pace with a timer. If you have to speed up to hit 35 seconds, it is too long. If you find yourself adding "and honestly" and "like, seriously" to fill time, it is too short, or the idea is too thin.
What is a Deep Narrative ad?
Deep Narrative is 40 to 55 seconds and exists for one reason: skeptics need to see their own failed attempts before they will believe a new answer.
Jake 2.0 lives here. The extra 20 seconds over Standard DR are not filler. They buy the journey (named things the person tried and why each failed), the realization (the shift in understanding), and a first win that is small enough to believe. Take those out and you have a 1.0 script with a slower start.
Groundswell, a $289 raised garden bed kit with tool free corners and a built in drip line, is a natural Deep Narrative product. The buyer has probably already tried pots, a cheap kit, or a reminder app. They need to hear those named, then hear why "it waters itself" changes the problem from "I need to be better at gardening" to "I need a garden that doesn't depend on me." The full Groundswell script is in the frameworks article.
Deep Narrative only holds attention if the emotional pacing is real. A 55 second script with no friction, no doubt and an instant miracle is just a long ad.
When does Long Form make sense?
Long Form, 60 to 180 seconds, fits considered purchases, founder content, and story formats where the viewer is watching a narrative rather than an ad.
Three common uses:
- Full funnel one shot for a considered product. One video that walks Problem Aware to Most Aware with full proof. For Groundswell this might open on a dead garden, show the full afternoon build in real time with no tools, explain the drip line, answer the "do I need a water hookup" style objections the brand has actually heard, and close on the first harvest.
- Founder content. A founder answering real questions (what problem did you set out to solve, how does it solve it, what makes it different) can hold attention for minutes if they are credible and specific.
- Trigger event animations and story formats. A first person narration over animation, built from a real customer moment, often runs 60 to 180 seconds. See output modes.
The risk is obvious: long form multiplies weak moments. Every beat has to earn the next one. A useful check is to ask, at every 10 second mark, "would I keep watching if I knew this was an ad?"
How do you cut a long script into a short one?
Cut from the middle, never from the ends.
- Keep the hook. It is the only line every viewer hears.
- Keep the handoff line. The sentence that makes it personal is what turns a watcher into a thinker.
- Keep one mechanism line. The single sentence that explains why it works.
- Keep the close. An earned close in a short cut is still an earned close.
- Cut in this order: second and third failed attempts, expanded outcomes, agitation detail, then the first win.
Here is the Copperbean idea (cold brew concentrate, brewed slow for a long steep then filtered twice, one bottle makes a week of coffee) cut from Standard DR to an ADHD Cut:
0 to 3s: "Stop pouring half your homemade cold brew down the sink." 3 to 8s: "Copperbean's brewed slow, then filtered twice. That's where the bitterness goes." 8 to 13s: "One bottle is my whole week. I'm done paying for coffee twice."
About 30 words, and it still has a hook, a mechanism, an outcome and a close. What it lost is the person. Shorter cuts trade the relatable human for speed, which is why they suit warm viewers who already trust the brand.
Reveal timing moves with the cut. Reveal guides in our system are proportions of the ad, so a product revealed at 40% of a 35 second script lands around 14 seconds, and in a 15 second cut around 6.
Should you test different lengths?
Yes, as a deliberate pick, holding everything else constant.
If length is one of your two picks, keep the angle, the creator and the framework the same and change only the length. Otherwise you will not know whether the shorter ad won because it was shorter or because it said something different. Judge the result on outbound CTR (the share of viewers who click through to your site) and on purchases together. Short cuts often win on outbound CTR and lose on purchases when the product needs belief. Long cuts often show the reverse.
What mistakes do people make with ad length?
- One length for every ad. A blanket "15 seconds max" rule starves considered products.
- Writing past the word count. The script that is 40 words too long gets rushed or cut badly in the edit.
- Cutting the close to save time. A short ad with no close is a trailer.
- Padding a thin idea to fill 55 seconds. If the idea only needs 20, make a great 20.
- Silent openings. Audio starts at the first frame, and something on screen should change in the first second and a half, whatever the length.
Key takeaways
- Length should follow the argument: price, awareness stage and framework decide it.
- Plan on about 2.5 words per second for a direct response read, and count words before filming.
- ADHD Cut is one reason in 10 to 15 seconds; Standard DR at 20 to 35 seconds is the default.
- Deep Narrative exists for skeptics and the failed attempts that win them over; Long Form fits considered purchases and story formats.
- Cut from the middle: keep the hook, the handoff, one mechanism line and the close.
- Test length on its own, with angle and creator held constant.
FAQ
Is 15 seconds the best length for Facebook ads? Only for some products. A 15 second cut works when the decision is simple or the viewer is already warm. Considered purchases and cold, skeptical viewers usually need more runway to believe.
How many words is a 30 second ad script? About 75 words at a direct response pace of 2.5 words per second, and about 60 at a calm pace. If your draft is much longer, it will not fit without rushing.
Does length change which biases I use? It changes how many you can use well. A short cut has room for one hook bias and one close bias. A long script can carry a full stack across hook, body and close. See the cognitive bias stack.
Can one script be filmed once and cut to several lengths? Yes, if it is written with that in mind. Film the long version with clean breaks between beats, then cut down by removing middle beats, keeping the hook, handoff, mechanism and close.
If you want to know whether your ad lengths match what your products and audiences need, we offer a free Meta account audit. It includes a full account audit plus net new reach and cumulative reach reports. You can request one at our contact page.
Check what you learned
Tap an answer to see if you got it right.
1. How long should a script be?
2. What's the default length?
3. When cutting, where do you cut from?
Ready for the full test? Take the Ad Scripts exam and get certified. Want this run on your account? Get my free audit.
Module 6: Ad Scripts · article 10 of 21
Next: Peripheral vs Central Route: How Viewers Process Facebook Ads
Previous: What Makes a Good Ad Hook? The Hook Types That Stop the Scroll
About the author
Jake Mehani, founder of Creative Analytics
Jake has spent 15 years running Meta and paid social ads for consumer brands. Meta published his RBX Active Advantage+ case study, he taught the DMAnc Meta and TikTok ads masterclass, and he spoke at Motion's 2024 Creative Strategy Summit. His senior team runs media buying, creative and measurement like an in house team.