Put on any Apple product film and watch the bottom of the frame. Sooner or later, in type small enough that it clearly isn’t trying to persuade you of anything, it says sequences shortened.
That’s most of the article, honestly. The most confident marketing operation on earth quietly tells you that what you just watched is not how long any of it takes.
Every few months a founder sends me an Apple film and says “like this.” They mean the feeling. Calm, expensive, inevitable. Then whatever we can actually make gets held next to a film with a motion control rig in it and everybody is a bit deflated. So I’ve spent a lot of time pulling those films apart to work out which parts are money and which parts are just decisions.
More of it is decisions than people expect. Not all of it, and I’ll be specific about which is which, because the version of this article where a startup can have all of it is a lie.
Count the claims
The most copyable thing about an Apple film costs nothing at all: they say almost nothing.
Watch one with a pen and count the separate claims. It’s usually one per sequence and a handful across the whole film, and each one gets an entire visual idea to itself. Then the specifications arrive right at the end, fast, in a block, for the people who came for specifications.
Now count a startup launch film. I’ve seen ninety second films carrying forty distinct claims. That isn’t a taste failure, it’s an org chart failure. Every feature has somebody internally who fought for it, and cutting it feels like telling that person their quarter didn’t count. The claim count is the first thing I’d change on almost any film, and it costs nothing except an argument.
Watch what they refuse to show
No ports. No settings screens. No comparison chart until the very end. Nobody fumbling, no hand hesitating over a control, no menu opened by mistake.
And almost no real time. That fine print is doing real work: every interaction you see has had its dead air removed. The thing loads instantly. The panel is already open. This is why honest footage of your own product, at real speed, feels so much worse than you expect when you cut it together.
Trimming the dead air out of your own demo is free, and it closes more of that gap than any camera will.
Three typefaces in forty years
Apple Garamond, then Myriad from around 2002, then San Francisco from 2015. That’s the list. One typeface at a time, held for a decade or more. Inside the films it’s nearly always a single line, centred, held long enough to read twice, then gone.
Compare that to the default startup film, where type animates in with a bounce, three lines at once, over the product, in the brand colour, with a gradient on it. Every one of those is a decision to add. The restrained version is a decision not to, and it’s the same price.
Sound works the same way. Apple films regularly open on the sound of the product itself, or on nothing. Silence is free. It’s also the most underused tool in every startup film I get sent, because by the fortieth time you watch your own edit a quiet moment feels like something has broken.
The most famous ad they ever made barely shows the product
1984 is a one minute film, directed by Ridley Scott, about a computer that is for practical purposes not in it. There’s a drawing of a Macintosh on the runner’s vest. At the end, a line of text says the product exists and gives a date. That’s the entire appearance.
It aired nationally exactly once, in the third quarter of the Super Bowl on 22 January 1984. The only other broadcast was a single late night run in Twin Falls, Idaho on New Year’s Eve, so it would qualify for that year’s awards. One national airing, for the ad people still argue about forty years later.
I’m not suggesting anybody copy that. You are not going to launch your product by refusing to show it. The transferable part is the ratio: how little of a film Apple has historically been willing to spend on saying things, and how much of it they’ll spend making one thing felt.
The cheapest thing Apple ever ran, ran for three years
Get a Mac ran from 2006 to 2009. White background. Two people standing on it. Thirty seconds. One idea per spot, played as a small scene, then over. They made dozens of them and shot localised versions across seven countries.
There’s no rig in that. No location, no crane, barely a grade. Two actors, a white cyc and a script disciplined enough to compare exactly one attribute per film.
If you have a small budget and you want to steal from Apple, steal from that campaign rather than from the ones with a product exploding gently in a black void. The format was nearly free. The writing was the expensive part, and writing is the part you can actually afford.
What genuinely costs money
Motion control rigs. Macro probe lenses. Weeks of shoot time for sixty seconds of finished film. Prototypes flown in for photography. A colourist who normally works on features. A sound team on a product film. That’s real money and no amount of cleverness substitutes for it.
There’s also one thing money can’t buy quickly, and it’s the reason the imitations feel hollow. Apple can open on a black screen and one word because everybody already knows what the product is. That’s decades of accumulated recognition doing the work. A startup opening the same way is asking for reverence before it has said what it does.
You have to explain before you get to be admired. That’s the real difference between those films and yours, and it isn’t a line in the budget.
The camera was the cheapest thing on set
Apple ran an entire product event in October 2023 under the banner that the whole thing was shot on iPhone. Then they released a behind the scenes reel of the shoot, which shows dollies, cranes, a full lighting package and a proper crew.
I love that they published it, because it settles an argument I have constantly. The camera was the cheapest thing on that set. What made those shots look like that was the lighting, the movement, the blocking and the people operating all of it. If you’re a founder who thinks the gap between your film and theirs is a camera body, that reel is your answer.
Even Apple gets it wrong, and not for the reason you’d guess
On 7 May 2024 Apple put out a film for the iPad Pro in which a hydraulic press slowly crushes a piano, a trumpet, paint cans, sculptures and an arcade machine down into a tablet. The craft is immaculate. The idea, landing on an audience of the exact creative people who buy iPads, read as a machine destroying every instrument they love.
Two days later Apple apologised. Tor Myhren, who runs their marketing communications, said they’d missed the mark. It was pulled from television.
Nothing about that failure was technical. Every dial that costs money was set correctly. What went wrong was a judgement about how a metaphor would feel to one specific person, which is a decision fully available to a team of two with no budget at all. It’s also the decision that actually determines whether a film works.
What I’d take
One claim per sequence, and fewer of them than feels comfortable. Cut the dead air out of your own footage before you compare it to theirs. One line of type, held, no bounce. Let it be quiet somewhere. And explain before you ask to be admired, because you haven’t earned the black screen yet and pretending you have is the tell.
None of that is a budget. All of it is somebody in the room being willing to say we’re cutting this, which is the only Apple habit that really transfers.