Section 1
Why "worth including" is the wrong filter
Everything is worth including. That is the problem. Your best case study is worth a slide. Your methodology is worth a slide. Your team's credentials are worth a slide. Applied honestly, "is this worth including?" approves nearly every slide you could imagine, which is why decks only ever grow. The filter has no teeth because it measures the wrong thing: the value of the content in isolation, not its cost to the buyer's attention. And attention has a hard, measured ceiling. Cognitive load theory, one of the most replicated findings in learning science, holds that working memory can process only a small number of chunks of new information at any moment, and that when you exceed that limit, comprehension does not degrade gracefully, it falls off, because the brain has nowhere to put the overflow . A buyer watching your pitch is not a hard drive accumulating everything you show. They are a narrow channel that drops most of what arrives. Every slide you add does not increase what they retain. Past the limit, it decreases it, because the new slide evicts the previous one before it was encoded. This is why the forty-slide deck loses to the twelve-slide one. Not because thoroughness is bad, but because thoroughness, delivered to a limited channel, is indistinguishable from noise. The buyer cannot tell your critical slide from your filler slide. They are all just arriving, faster than memory can catch them.
Section 2
The coherence principle: less is literally more retained
The fix is not a matter of taste. Richard Mayer's research on multimedia learning produced the coherence principle, a specific, tested finding: people understand and remember a message better when extraneous material, anything that does not directly serve the core point, is removed . Not minimized. Removed. The interesting-but-tangential fact, the decorative image, the bonus data point, the third bullet, each of these does measurable harm, because the limited channel spends capacity processing it and has less left for the thing that mattered. Sit with the direction of that finding, because it is counterintuitive to a founder who equates more with better. Adding a relevant-but-secondary point to a slide reduces retention of the primary point. The slide with one idea outperforms the slide with three, not marginally, but because the three-idea slide forces the buyer to split a channel that cannot be split, so all three ideas land at partial strength and none encodes. Cognitive load research on slideshow design reaches the same practical conclusion: slides work when each one carries a single, clear focus and offloads rather than overloads the viewer's working memory . So the goal of a slide is not to say as much as it can. It is to make one thing land completely. Every additional element on that slide is not free enrichment. It is a tax on the one thing.
Section 3
The Purpose Test: one question, applied ruthlessly
Here is the test that kills the filler. For every slide, ask: "What is the single thing the buyer should remember from this, and could they state it in one sentence after seeing it once?" Then apply three verdicts. Run this honestly on a typical founder deck and the result is brutal: a large share of slides fail. Some fail because they carry no single memorable idea, they are context, throat-clearing, or "we should probably mention this." Others fail because they carry three ideas fighting for one channel. The "80%" in the title is not a literal law; it is what happens in practice when founders who have never applied the test finally do. The deck that survives is smaller, and it is remembered. Note the middle row, because it is where most of the work is. The problem is usually not whole junk slides. It is good slides carrying one great idea plus two competing elements, a strong headline undercut by a busy chart, three bullet points where one would land. You are not just deleting slides. You are removing the extraneous elements the coherence principle warns against , so the one thing on each surviving slide can actually land.
Section 4
How to run the cut without losing the substance
Cutting terrifies founders because it feels like deleting proof. It isn't, if you do it in this order. Four steps. 1. Name the one thing for every slide, in writing. Before cutting anything, force yourself to write the single sentence the buyer should remember from each slide. Slides where you cannot write that sentence in under a minute are your first casualties, they never had a purpose, you were including them because they felt owed. 2. Move deleted substance to an appendix or leave-behind, don't destroy it. The methodology detail, the extra case study, the deep data, these are not worthless. They are just not for the live pitch, where they overload the channel . Put them in a leave-behind document the buyer can study alone, at their own pace, without a limited working-memory channel racing against your delivery. You are not losing the proof. You are moving it to where it can actually be absorbed. 3. Split any slide that fails on two-or-more. A slide with three ideas is three slides pretending to be efficient. If all three matter, give each its own slide and its own single purpose. Three focused slides outperform one crowded slide, because each gets a clean shot at the channel instead of all three arriving at once. 4. Test the survivor deck on one-viewing recall. Show the cut deck to someone once, then ask them a day later what they remember. If the things they recall are the things you intended each slide to plant, the Purpose Test worked. If they remember the decoration and forget the point, you cut the wrong elements. A checklist version of this pass sits in the free LeverageOS starter guide.
Section 5
You're passing the Purpose Test right when…
You're passing it right when you can state, in one sentence, the single thing every surviving slide is supposed to plant in the buyer's mind, and nothing on that slide is fighting it. You're passing it right when your deck got dramatically shorter and your buyers started remembering more, not less, because you stopped mistaking volume for communication. You're passing it right when your deleted material lives in a leave-behind the buyer can absorb alone rather than in the live pitch where it overloaded a channel that cannot be overloaded . And you're passing it right when adding a slide has started to feel like a cost rather than a courtesy, because you finally internalized that every extra thing you show does not add to what the buyer keeps. Past the limit, it subtracts.