Presenting slides on camera when nobody is clicking for you.

The usual explanation is nerves or lack of rehearsal. Usually it is neither. It is that one person is doing three jobs that, on a stage, are split between a speaker, a clicker and a camera operator, and the clicker is the one that gets dropped. It is one of the reasons a short video takes all evening.

Three things in two hands

Presenting slides on camera alone, you are juggling:

• The script. Reading it, or at least following it, without sounding like you are.
• The lens. Keeping your eyes close to it, which already competes with the script.
• The clicker. Remembering, at the right word, to press a key or a remote that changes the slide.

The first two pull on your eyes. The third pulls on your attention, which is worse, because it is a background task you have to keep running for the whole take. Somewhere in your head a small voice is waiting for the cue: "after this sentence, click". While it waits, your delivery thins out. And when you do click, it is often visible: a glance down at the keyboard, a hand moving into frame, a small pause while you check the slide actually changed.

Then there are the two failures that cost the take.

The late slide. You make the point, then remember to click. For a second or two the viewer is looking at a slide about the previous point while you talk about the next one. In the edit you can sometimes cover it. Often you cannot, because the slide is the shot.

The early slide. You click on the transition, a sentence before the payoff. The new slide gives away the answer, the list, the number, and your next sentence becomes a recap of something the viewer has already read.

Both happen for the same reason: the slide change lives in your head, as an intention, instead of on the page, as an instruction.

A slide change belongs to a word

The fix starts before any tool. Go through the script with the deck open and, for every slide, decide the exact word it should appear on. Not the sentence, not the paragraph: the word.

This sounds fussy. It is the whole job. Some practical rules for choosing:

• Reveal on the noun, not before it. If the slide shows a number, change it as you say the number, not as you say "here is what we found". Otherwise the slide does the reveal and you are left narrating.
• Change on the first word of a new idea. When a slide introduces a section, put the change on the word that starts it, so the picture and the voice turn the corner together.
• Leave a breath after a dense slide. A chart or a list needs a second of reading time. Put the change a little ahead of your explanation, and write a short pause into the script after it, rather than talking straight over it.
• Fewer slides than you think. On camera your face is doing half the work. A slide per idea, not per sentence, means fewer cues to land and fewer chances to miss one.

Once every slide has a word, you have something you can rehearse, check and fix. Before that, you had a vague plan to click "around here".

Mark it in the script

Now write the decision down where you will see it while you read: inline, right before the word. Not in a separate list of slide timings, not as a note in the margin. A marker in the line itself means you meet the cue at exactly the moment it matters.

Keep the marker short and distinct from the prose, so your eye recognises it as an instruction and skips it instead of trying to read it out loud. Square brackets work well because nothing else in a spoken script uses them. Something like this:

So where does the first month actually go? [[slide:4]] Into the tool, before anyone has looked at the process.

Even if you still click by hand, this alone removes most of the late and early slides, because you no longer have to remember the cue. You read it. It also does something for the rest of your preparation: you can see at a glance which parts of the script are carrying a slide and which are just you talking, which is useful when you plan your takes in blocks and need to know where a section cleanly starts.

Slide numbers beat "next"

There are two ways to write a cue: "go to the next slide", or "go to slide 4". Next feels simpler. Use numbers anyway, at least at the start of each section.

The reason is retakes. Recording alone, you will restart a section several times. Say you get two slides into a section, fluff a line and go back to its beginning. The deck is still two slides ahead. With "next" cues, the first cue you read on the retake moves it one further, and you are now three slides off for the rest of the take. With a number, the first cue puts the deck back where it belongs, whatever happened before.

The cost of numbers is maintenance: insert a slide in the middle of the deck and every number after it shifts. So the rule is simple. Numbers on the first slide of every section, "next" inside a section if you like, and check the numbers against the deck every time you edit it.

Letting the script press the button

Once the cue is in the line, the obvious next step is to stop pressing the button at all. If the teleprompter knows where you are in the script, it can change the slide when you reach the marker.

That is what slide sync does in Mirrento, a teleprompter for the Mac. You write cues as [[slide:4]], [[slide:next]], [[slide:prev]] or a relative step like [[slide:+1]], and on the prompter each one shows as a small pill in the line. When a pill crosses the reading line (or, when the text follows your voice, when the word you are reading passes it), Mirrento tells Keynote or PowerPoint to change the slide, through AppleScript. Without voice follow, autoscroll and scrolling by hand trigger it too. If you scroll back so a pill sits below the line again, it re-arms and fires again the next time you read past it. Jumping straight to a chapter does not fire the cues you skipped, which is one more reason to start each section with a slide number.

A few things worth knowing before you rely on it:

• It is off until you pick an app. Choose Keynote or PowerPoint from the Slide sync button, then use the test action, which jumps your open deck to slide 1. If macOS asks for Automation access, that is the permission it needs.
• Open the deck first. With no presentation open, the cue has nothing to drive, and the control panel shows the error instead.
• PowerPoint slideshows step one slide at a time. PowerPoint's scripting has no "go to slide 9" while a slideshow is running, so a numbered cue gets there by stepping through the slides in between, and each step plays its transition. Keynote jumps directly. Keep big numbered jumps for moments when the deck is off camera.

If you already record with a deck, the test is simple: take one section, put the cues in, and record it twice, once clicking by hand and once letting the script do it. The 7-day trial is enough for that comparison.