← 1 Day ScriptAll programsHomeSearch
1 Day Script·FAQ·1:06:54

FAQ: How Long Should a Video Be (Platform, Objective, Viewer, and the 5% Rule) and How Much Does Quality Matter (the Message-Audio-Image Triangle)

Paul Xavier Instructor - Next Level Creators; walks through his own Facebook ad account and gear

The short version

  1. Length is set by three inputs - where it runs (Facebook feed up to 240 min, Story 120 s, YouTube skippable minimum 12 s, your own site unlimited), what you are asking for (click < opt-in < call < buy; the bigger the ask, the more irresistible the message must be), and who is watching and how well they already know you.
  2. Paul's own theory, labelled as unproven: longer builds trust like a courtship - 'I didn't ask my wife to marry me the first time I saw her' - provided the content is positive; negative content watched longer only deepens dislike.
  3. The real account: $77,928 spent, 230,964 reached, 1.4M plays, average watch 7 seconds, ~31,600 completions - and qualified sales meetings at $59.81 against a $200 ceiling. Under 5% of strangers finish any ad; shortening the ad so everyone finishes is 'counterproductive and a foolish practice.' The only universal answer: 'the length that generates the best results.'
  4. Quality triangle: message, audio, image, weighted 80/20. An iPhone-and-PowerPoint video 'generated over $2 million'; a Canon 70D video on his wife's massage site runs at 10x ROI. Chris Gardner in paint-stained clothes still gets the internship.
  5. Audio beats image: Mehrabian's 7-38-55, and a demo pairing a $4,000 audio chain with iPhone video against a $6,500 camera with iPhone audio. Priority: message (80% of effort), then audio, then image - rate each on a 0-10 believability scale; the only-implant-dentist-in-300-miles scores 10 on message and can shoot everything on a phone.

At a glance, three clicks deep

Skim here first: the closed row is the glance, open is the study card with the key points and timestamps, and the ↓ link drops to that concept's full write-up below.

01Three factors set video length: platform, objective, viewerLength = f(placement limits, size of the ask, viewer familiarity);

Length = f(placement limits, size of the ask, viewer familiarity); never a fixed number.

Three factors (p8850638 0:01, 0:03, 0:12)

'Value video' umbrella (p8850638 0:01)

Facebook 240 min / Story 120 s; YouTube 12 s min, skip at 5 s (p8850638 0:02-0:03)

Own site: no limit; requirements list below module (p8850638 0:02-0:03)

↓ Full write-up of this concept

02The commitment ladder: the bigger the ask, the more irresistible the message must beRank the ask by commitment;

Rank the ask by commitment; scale message strength (and often length) to the rung.

Examples low to high (p8850638 0:05)

Conversion share falls as commitment rises (p8850638 0:07)

Message strength must scale with the ask (p8850638 0:08)

↓ Full write-up of this concept

03Paul's theory: length builds trust like courtship - if the content is positiveLonger exposure to positive, proof-rich content compounds trust;

Longer exposure to positive, proof-rich content compounds trust; longer exposure to weak content compounds dislike.

Explicitly unproven (p8850638 0:10)

Marriage analogy (p8850638 0:10-0:11)

Negative content watched longer backfires (p8850638 0:11-0:12)

↓ Full write-up of this concept

04The 5% rule: fewer than one in twenty strangers finish an ad, and that is fineOptimise for the buying minority who finish, not for completion rate;

Optimise for the buying minority who finish, not for completion rate; judge length by cost per qualified action.

Spend, reach, plays, 7-second average (p8850638 0:13-0:14)

25% / 50% / 100% funnel (p8850638 0:14)

$200 ceiling vs $59.81 actual (p8850638 0:15)

Under-5% rule; fleeting glance (p8850638 0:16-0:17)

Shortening for completion is foolish; door-to-door (p8850638 0:17-0:19)

Re-cut long ads into short hooks (p8850638 0:19, 0:27)

↓ Full write-up of this concept

05The only universal answer, and the three-step process for choosing lengthStarting certainty (0-10) -> required belief changes -> length that delivers them;

Starting certainty (0-10) -> required belief changes -> length that delivers them; split across videos if the platform forbids the length.

'The length that generates the best results' (p8850638 0:20-0:21)

Step 1 certainty scale (p8850638 0:21-0:22)

Step 2 messaging map (p8850638 0:22)

Step 3 length; platform caps -> potency or sequences (p8850638 0:22-0:23, 0:28)

↓ Full write-up of this concept

06The video quality triangle: message, then audio, then image (80/20)Message > audio > image;

Message > audio > image; 80% of effort on the message; a phone is enough for image if the other two corners hold.

Triangle and the spirals (p8850637 0:00-0:01)

Pareto anecdotes (p8850637 0:02-0:03)

$2M iPhone/PowerPoint video; 70D massage video 10x (p8850637 0:12)

Pursuit of Happyness; Blair Witch; format emotions (p8850637 0:07-0:10, 0:23-0:25)

Mehrabian 7-38-55 (p8850637 0:16-0:17)

$4,000 audio vs $6,500 camera demo (p8850637 0:28-0:29)

↓ Full write-up of this concept

07Three believability scales and the dentist test: message 10/10 on an iPhoneScore message, audio, image 0-10 for believability;

Score message, audio, image 0-10 for believability; invest where the score is low and the stakes are high; message first.

Message first, 80% of effort (p8850637 0:31-0:32)

Radiator thought experiment (p8850637 0:32)

Three independent scales (p8850637 0:33-0:34)

Dentist example (p8850637 0:34-0:35)

Self-invented while grading client campaigns (p8850637 0:35)

↓ Full write-up of this concept

The concepts in full

01

Three factors set video length: platform, objective, viewer

There is no right length. There is a right length for this placement, this ask, and this viewer.

'The three key factors which we consider before selecting the ideal length': the platform ('where are you going to distribute it?'), the objective ('how do you know when your video is successful?'), and the viewer ('who is the ideal video viewer that you are communicating to'). 'Value video' covers 'a commercial, a edutainment video, a video sales letter, a promo video, a corporate video, a video advertisement, a webinar', and a campaign 'can consist of an infinite number of value videos.' Platform numbers as of recording: 'a Facebook video ad placed in the newsfeed can be up to 240 minutes long maximum, while a Facebook story ad can only be 120 seconds'; YouTube in-stream skippable ads 'have to be a minimum of 12 seconds' and are skippable 'after about five seconds'; on your own site with driven traffic 'there are no restrictions.' A kept-up-to-date requirements list sits below the module.

Why it matters

The first question to ask a client who says 'make it 60 seconds': for which placement, and asking for what?

02

The commitment ladder: the bigger the ask, the more irresistible the message must be

A click is cheap to earn. A credit card on the first watch is not. Write accordingly.

'Different objectives require different levels of commitment from your viewer.' Low: click a link, read a free article, opt in for an ebook or case study; high: call the company, book an appointment, fill a survey, 'buy now' with a card, mail a cheque. 'The lower the objective and level of certainty, the weaker your messaging and content can be while still achieving results. Then the higher the objective and level of certainty, the more irresistible your messaging and content will need to be.' Lower asks convert a larger share; length is one of the levers for making a message more irresistible.

Why it matters

Explains why the same client needs a 15-second retargeting cut and a 20-minute VSL - they sit on different rungs.

03

Paul's theory: length builds trust like courtship - if the content is positive

He flags it himself: no proof, just a couple of million dollars of observation.

'I don't have definitive proof to substantiate that longer videos are the ones that sell more. I'm merely sharing what I've observed from producing dozens of value videos in different markets and generating a couple million dollars in sales.' The analogy: 'I didn't ask my wife to marry me the first time I saw her... We built a relationship and trust over a long period of time.' So 'the longer your value videos are, the stronger the relationship and trust with your company and offer can become' - room for demonstrations, case histories, myth-busting, testimonials - with the caveat that 'if your messaging content is perceived as negative by the viewer, the more time they spend watching... the more they will dislike your company.' Contrary to 'pretty much every other marketer' preaching 60 seconds or less.

Why it matters

An honest example of stating a belief as a belief - the kind of hedge Paul's own reports should carry.

04

The 5% rule: fewer than one in twenty strangers finish an ad, and that is fine

Average watch time: seven seconds. Cost per qualified sales meeting: $59.81. Both numbers are true, and only one matters.

Live in the ad account: 'we spent $77,928.46 to generate 230,964 people who viewed our advertisements... 1.4 million video plays... average video watch time only 7 seconds' on ads 'up to 20 minutes long.' Drop-off: 103,000 watched 25%, 59,000 half, 'at 100%, we're at just 31,000' (31,598). Willing to pay 'up to $200 to get one phone call'; actual 'qualified sales meetings for just $59.81.' Generalised: 'A general rule of thumb when advertising to complete strangers is less than 5% of viewers will watch your video ad from start to finish' - most give 'a fleeting glance' or 'two to five seconds.' Therefore 'shortening your video ad to a length where everyone will watch it is actually counterproductive and a foolish practice' - it forces generic messaging and 'being funny instead of direct and irresistible.' The door-to-door analogy: you would not cap your words with a serious in-person prospect. Pro tip: long ads are re-cut into several short ads with different hooks.

Why it matters

Kills the 'completion rate' vanity metric in client reporting; report cost per qualified outcome instead.

05

The only universal answer, and the three-step process for choosing length

how-to

Rate the viewer's certainty out of ten before they press play. The video's job is the distance to ten.

'There's an ongoing argument... video ads should be 30 seconds, minutes, hours' - the only universal answer is 'The ideal length of your video ad is the length that generates the best results.' Three steps: (1) rate the ideal client's certainty/commitment toward the offer on a 0-10 scale before watching; (2) map the messaging needed to move them - educate, build trust, establish credibility, dispel competitor myths; (3) produce 'at whatever length required' to close the certainty gap and trigger action. If a platform caps length, 'your video needs to become more potent' - credibility and social proof fast - or use several sequential value videos (see the Distribution training).

Do it in this order
Why it matters

A three-line brief format Paul can use for any video: starting certainty, what must change, therefore length.

06

The video quality triangle: message, then audio, then image (80/20)

Two million dollars from an iPhone and a screen-recorded PowerPoint. The camera was never the constraint.

'Those three components that make up the video quality triangle are message, audio and image', combined into either 'a positive spiral' or 'a downward spiral.' Weighting by Pareto - 'roughly 80% of the effects come from 20% of the causes' - with Paul's garden, wealth, roads and violence anecdotes. Image is secondary: 'The first one I ever produced in this market was shot with a iPhone and... screen recording PowerPoint slides. It generated over $2 million in sales'; a Canon 70D video on his wife's massage site has run 'two and a half, three years' producing '$300' clients for '$30' - 'a 10x ROI.' Chris Gardner (The Pursuit of Happyness) lands the internship in paint-stained clothes on message alone; The Blair Witch Project's shaky footage was a hit; formats carry emotion by design (iPhone-to-camera = authenticity, POV mountain-biking = adrenaline even muted). Audio outranks image: 'I would actually put more money into a better microphone than the camera' - Mehrabian's 7-38-55 (words, tonality, body language) - demonstrated with 'a Rode NT one... through the toft audio design soundboard, which is about $4,000' on iPhone video versus 'a Canon C100, which is a $5,000 body... L Glass 1.2 f stop Canon 50 mil... a $6,500 camera setup' recording audio through the iPhone; bad audio makes 'even five minutes... painful... unwatchable.'

Why it matters

The argument Paul needs when a client wants to spend the budget on a camera before they have a message or a microphone.

The video quality triangle: message first (80% of effort), then audio, then image MESSAGE 80% of your time AUDIO Mehrabian 7-38-55 IMAGE a phone is enough iPhone + PowerPoint slides -> "over $2 million in sales" Canon 70D on a massage site -> 10x ROI for three years $4,000 audio + iPhone video beats $6,500 camera + iPhone audio radiator + $100,000 of gear + no message -> nobody buys Rate each corner 0-10 for believability; the only implant dentist in 300 miles is a 10 on message and shoots on a phone
Message, audio and image with the 80/20 weighting and the proof points behind each corner.
07

Three believability scales and the dentist test: message 10/10 on an iPhone

$100,000 of camera gear pointed at a radiator with nothing to say sells zero radiators.

'Prioritize putting your time and money into the message first, foremost, and spend 80% of your time there.' Thought experiment: '$100,000 in camera equipment, massive crew, and a amazing sound team' shooting a radiator with no message - 'no one's going to buy the thing.' Rate message, audio and image each on an attract-vs-repulse believability scale (0-10); you do not need tens - 'it could be a 5, it could be a 6, a 7, an 8.' Dentist example: 'the only one offering dental implants in a 300 mile radius' - closest, affordable, 'see you tomorrow' - scores message 10/10, so audio and image can both be an iPhone and the ad still works with distribution. The scales were invented by Paul and Anthony 'grading them for our clients' - 'it didn't exist... we had to do it for ourselves first.' A silent product demo can still attract if the visuals alone transfer certainty. No skills lab in this module; questions go to the community.

Why it matters

A scoring sheet Paul can put in front of a client: three numbers, and the budget follows the lowest one that matters.

Check yourself

Answer from memory first — the recall attempt is what makes it stick. Then reveal.

A client's product demo has no voiceover and phone-quality footage. Can it still work?

Yes, if the visuals alone transfer certainty - Paul's silent-demo case; score the message corner, not the camera.

Tools referenced

ToolCoverageMomentContext
Meta AdsdemonstratedLive Facebook ad-account walkthrough with spend and watch-time funnel
YouTubementionedIn-stream skippable ad rules
Microsoft PowerPointmentionedScreen-recorded slides as the $2M video's visuals

Session materials

Archived locally on V: — click to open. Companion pages link to the LMS.

Action items

    Resources mentioned

    Resources
    • docChannel video-ad requirements list
    • docNext Level Creators trainings
    • docFilm clips used as examples

    Extraction notes

    This page was built from an auto-generated transcript, which garbles product and people's names. Those were corrected silently in everything above and logged here for transparency. The warnings flag claims that were true on the recording day but change fast.

    Transcript corrections applied

    The transcript saysThe trainer actually means
    Albert Meribian / Merben 73855 ruleAlbert Mehrabian 7-38-55 rule
    Belair Witch ProjectThe Blair Witch Project
    full on you to browsefull unibrow
    $160 cheaper$140 less than the $200 ceiling (speaker misstatement)
    fincon in New OrleansFinCon (conference)

    True on recording day — verify before relying