Most slides follow a default structure nobody consciously chose: a short topic phrase at the top, a bulleted list underneath. “Q3 Sales Overview.” Four bullets summarising the quarter. Assertion-evidence slides replace both halves of that default at once, and the case for doing so is backed by actual published research, not just a stylistic preference.
Assertion-evidence slides: the structure, stated plainly
An assertion-evidence slide pairs two elements and nothing more. The headline is a complete sentence stating the specific finding, not a topic label pointing vaguely at a subject: “Q3 revenue hit €108m, up 12.5% on Q2” rather than “Q3 Sales Overview.” Below it sits one piece of visual evidence, usually a single chart, diagram or image, chosen specifically because it supports that exact sentence rather than the general topic. Everything else that would normally compete for attention on a slide, decorative badges, a second chart making an unrelated point, a bulleted list of caveats, gets removed rather than added underneath.
The bulleted list is the harder habit to break, because it feels like thoroughness rather than clutter while it’s being built. A list of four sub-points under a topic phrase asks the audience to read and synthesise the actual finding themselves, in real time, while also listening to whatever the presenter is saying out loud. A single sentence headline does that synthesis for them, once, before the audience has to do any work at all.
Where the research comes from
The assertion-evidence approach was developed and researched by Michael Alley at Penn State’s Leonhard Center, with funding that has included a National Science Foundation grant. This is not merely one professor’s stylistic opinion dressed up as a methodology; it is an actively researched claim with published, testable results behind it.
A study led by Garner and colleagues in 2013 compared audiences who watched identical spoken content delivered with two different slide formats, conventional topic-and-bullets slides against assertion-evidence slides. The audiences who saw assertion-evidence slides understood and remembered the material better, a difference the researchers reported as statistically significant. A separate 2016 study by the same research group looked at the effect on presenters rather than audiences, and found that people who built their own slides using the assertion-evidence structure developed a deeper understanding of their material in the process of building it, compared to presenters who built a conventional deck covering the same content. A more recent 2023 study, led by Wolfe and colleagues, extended the research further, finding that the same underlying multimedia-learning principles hold even when the audience is made up of experts hearing advanced material, not just students in a classroom.
Building one, step by step
Start from the sentence, not the slide. Write down the single finding a slide needs to communicate, in full-sentence form, before opening any design software at all. If that sentence can’t be written cleanly, the slide isn’t ready to be built yet, regardless of what chart or image might eventually sit underneath it.
Next, choose exactly one piece of visual evidence that supports that sentence directly. A sorted bar chart showing regional sales supports “Region three grew fastest this quarter.” A line chart with a clear inflection point supports “Churn fell after the March onboarding change.” The evidence should be chosen the same way a lawyer picks an exhibit: because it proves the specific claim being made, not because it happens to be the most interesting chart available from the same dataset.
Finally, remove anything on the slide that isn’t the headline or the evidence. That includes decorative shapes, secondary charts, and bulleted context that belongs in speech rather than on the screen. A slide with one sentence and one chart, and genuinely nothing else competing with them, is the target shape.
A simple template, and a worked example
The template underneath all of this is short enough to write on an index card: one sentence stating the finding, in the header position; one chart, diagram or image beneath it that proves that sentence; nothing else on the slide competing with either. Filling that template in is where most of the actual work happens, and it helps to see it applied to a real document rather than a hypothetical one.
EY’s 2015 report on the sharing economy sorts the field into four categories, one of which, recirculation of goods, it traces back to platforms like eBay and Craigslist emerging in the 1990s. A conventional slide built from that report would likely carry the topic phrase “Categories of the Sharing Economy” over four bullets, one per category. An assertion-evidence version of the same content picks one category to make the point with rather than listing all four at once: “Recirculation of goods is the sharing economy’s oldest category, dating to 1990s platforms like eBay,” paired with a single timeline graphic showing that category’s platforms against the newer ones EY documents elsewhere in the same report. The full four-category breakdown still exists, just spread across four slides instead of stacked into one, each one making its own specific, evidenced claim instead of all four sharing one crowded topic slide between them.
Testing a finished slide: the squint test
A useful check for whether an assertion-evidence slide is actually working, rather than merely following the format on paper, is to blur it from across a room, or squint at it from a few feet back, and ask what the eye lands on first. On a working slide, the answer is the headline, then the one piece of evidence beneath it, because together they are the largest and most visually dominant elements on the page. On a slide that has drifted back toward the old default, a logo, a decorative shape or an oversized brand mark often wins that contest instead, even with a proper sentence headline technically sitting above it.
The squint test below runs exactly this experiment on three different slide designs covering the same underlying finding, a churn number that fell over six months, with a slider that blurs each one progressively. It is a fast, visual way to catch the gap between a slide that follows the assertion-evidence rules on paper and one that actually earns the reader’s attention in practice, which is ultimately the only test that matters.
Where this fits with the rest of slide design
Assertion-evidence slides solve a structural problem, one assertion, one piece of evidence, nothing competing with either. They don’t automatically solve chart selection or chart honesty on their own: a beautifully structured assertion-evidence slide can still carry a chart with a cut axis, or the wrong chart type for what the data actually needs to show. The which chart should I use guide and the misleading graph lesson cover exactly that layer, and a genuinely strong assertion-evidence slide needs both: a structure built around one message, and a piece of evidence that is itself chosen and drawn honestly.