Executive presentations, five years later: the rules now live in my AI

Looks back at How You Can Create Excellent Executive Presentations, from 2021.

In April 2021, I wrote a post about building executive presentations. It had five rules, a book recommendation and, in the third paragraph, an admission: this was a topic where I felt “huge imposter syndrome.”

I gave the rules anyway. Start with the why. Write the executive summary last. Use sections. Give every slide a title that says what it means. Keep it to four to eight slides.

Since then, I’ve built decks for our CEO and for senior vice presidents. Most of those rules held up. What changed is where they live.

Slides come last

I don’t start in PowerPoint anymore. I start with a document: a summary of what I think, in full sentences, before a single slide exists. Then an outline. The slides come last.

That flips my 2021 advice. Back then, I built the slides first and wrote the executive summary last, from the section titles. Now the thinking comes first, on a page, so by the time I open the slides, I already know what the executive summary says.

That first page still does most of the work. If the person reading it stops there, they should know what I’m asking for and why it matters to the business. A strong executive summary is still the best part of a strong deck.

The pre-read is the deck

In 2021, I wrote that the material travels further than the meeting. That’s truer now. I like pre-reads, and in practice the pre-read is usually the deck itself.

So the deck has to work without me in the room. That’s why the slide titles still matter. Each one has to say what the slide means, because nobody is there to explain it.

The mistake I see most

The decks people send me have one problem more than any other. They’re too long, and the main point is buried.

In 2021, I said four to eight slides. I’d still say it. Most long decks aren’t long because the subject is complicated. They’re long because the writer hasn’t decided yet what they’re asking for.

The rules moved

In 2021, my rules lived in a blog post and in my head. Now they live in my AI.

I’ve written skills for Claude and ChatGPT. One holds my voice, so a draft sounds like me before I touch it. Another holds my rules for writing to senior leaders. It puts the bottom line first, works out what kind of message it is (a decision, an approval, a risk or an update), runs the draft through a set of tests, and never invents a number. A third builds decks in our company template.

So the AI starts where the rules end. It drafts. I edit. The decision about what to recommend, and how hard to push, is still mine.

Writing the rules down for a machine also showed me which ones were rules and which were habits. A rule I couldn’t explain to the AI usually turned out to be something I did because I’d always done it.

The decision

In 2021, I wrote that “the way you win is if a decision is made, even if it’s declining your recommendation.” I still believe it, mostly. The decks I’ve taken to senior leaders have mostly ended in a decision. None of them has ended without feedback.

That’s the part no rule covers. A good deck gets you to the decision faster. It doesn’t get you there untouched.

In 2021, I called this my imposter-syndrome topic. I feel it less now. Part of that is five years of decks. Part of it is that the rules no longer depend on my memory. The judgment still does.

I first wrote about this in 2021, in a post called How You Can Create Excellent Executive Presentations.