Every tutorial about animating a zoom says the same thing: set a keyframe here, move the playhead, set another keyframe there. And every beginner hits the same wall, which is that nothing about that sentence tells you what a keyframe is or why the first one has to exist before the second one means anything.

The concept is genuinely simple. The interface for it is genuinely bad.

What a keyframe is, in one paragraph

A keyframe is a value pinned to a moment. On its own it does nothing interesting. Two of them make an animation, because the editor fills in everything between: at two seconds the zoom is 1.0, at four seconds it is 1.4, so at three seconds it is somewhere in the middle and the picture appears to move.

An editing timeline: video clips, a zoom keyframe and an audio track below.

That is the whole idea, and it is the same idea in every editor: a keyframe records a value at a specific point in time, and two of them define a change between those points. The problem is that the first keyframe is invisible work. You set a value that is already the current value, at a moment where nothing is meant to change yet, and nothing on screen looks different. Plenty of people conclude it did not work and stop.

The shortcut: mark the start, then just make the change

The alternative flow is to tell the editor when the move should begin, and then edit normally.

In Zella on iPhone it is in the Motion panel. Tap Animate: set start point, which captures the current moment and the clip's current transform. Move the playhead to where you want the move to finish. Then change Scale in Effects the way you would if you were not animating at all.

Both keyframes are written for you: one at the moment you armed, holding the value it had then, and one at the playhead, holding the value you just set. The session disarms itself, so the next change you make is an ordinary edit again.

The mental model has changed in a useful way. You are no longer thinking "pin a value, pin another value". You are thinking "start here, end like this", which is what you wanted in the first place.

The safeguards that make it usable

Three behaviours stop the shortcut from producing surprises:

  • If you do not move the playhead, nothing is written. A start point and an end point at the same moment is not an animation, it is a contradiction, so the editor leaves the session armed and waits.
  • Tapping the button again cancels. Armed is a visible state, not a mode you get stuck in.
  • The value you land on is the value you keep. After the animation is written, the clip still reads at the number you dialled, so what you see while adjusting is what you get.

That third one is the fiddly part under the hood, because the animation composes on top of the clip's static transform rather than replacing it. If that is handled carelessly you get double application, where the effect is applied twice and the zoom overshoots badly. Worth knowing mostly so you can recognise the symptom if you meet it in another tool: an animation that ends up far stronger than the number you typed.

It is not only for zoom

The same flow covers the four properties you would actually want to animate:

  • Scale for push-ins and pull-backs.
  • Position for slow pans across a still or a screenshot.
  • Rotation for tilts, used sparingly.
  • Opacity for fades that are not tied to a transition.

A slow push on a static talking head is the highest-value one, because a frame that is changing holds attention longer than a frame that is not. Keep it small. Something around 1.05x over a few seconds is felt rather than seen, and that is the intent.

When you still want real keyframes

The shortcut writes two keyframes. Anything with three or more moments still needs the manual tools, and they are all still there: the keyframe graph, the add-keyframe button, and dragging keyframes to retime them.

Reach for those when you want a move that goes out and comes back, a hold in the middle, or easing you want to shape yourself. Reach for the shortcut for the ordinary case, which is most cases.

One caution: a manual zoom you placed yourself and an automatic polish pass can end up arguing over the same seconds, each adding motion the other does not know about. If you have animated a clip deliberately, check what any one-tap polish is doing to it before you export, and read how to colour grade phone video if the picture itself is what needs work rather than the movement.

For the wider question of how much motion a video needs and where it belongs, the three edits that fix a weak opening covers the pacing side.

Frequently asked questions

Do I still need to understand keyframes to use this? No. You mark where the move starts, move the playhead, and make the change. The keyframes are written for you.

What happens if I change the value without moving the playhead? Nothing is written and the session stays armed, because a start and an end at the same moment is not an animation.

Can I animate more than the zoom? Yes. Scale, position, rotation and opacity all work the same way.

Can I edit the animation afterwards? Yes. It produces ordinary keyframes, so the graph and the drag editor can retime or reshape them.

How do I cancel once I have armed it? Tap the button again. It is a visible armed state, not a mode you are stuck in.

What if I need three or more points in the move? Use the manual keyframe tools. The shortcut is built for the two-point case, which covers most real moves.