When we started Napkin, the idea was simple: your words contain ideas, and those ideas deserve to be seen – not just read. We built a tool that could look at what you’d written and turn it into a visual that actually communicated something.
Over the past two years, millions of people have used Napkin to do exactly that – generating over a hundred million visuals in the process. We read their feedback. We talked to them. And through all of it, a pattern kept emerging, quietly and consistently: people were taking the visuals they’d made in Napkin and putting them into presentations.
And alongside that, the requests kept coming in. Can Napkin do the whole presentation? Not just the visual for one slide, but the entire thing – the structure, the narrative, the slides.
We sat with that question for a while. Because the right answer wasn’t just “yes, we can build that.” It was “yes – and we have something specific to say about how it should be done.”
The Beliefs That Shaped Napkin Slides
A great presentation is born out of clarity – and clarity comes from thinking, not just great design. The best decks we’ve ever seen are built on a clear argument shaped for a specific audience. They have a point of view that only that person or team could have. Design amplifies that clarity. But it can’t replace it.
This is where we think the current generation of AI tools gets it fundamentally wrong. They optimize for a beautiful looking output – a finished deck, quickly. But in doing so, they skip the thinking entirely. They take a topic and generate. They don’t try to understand what you’re trying to achieve, who the audience is, or what you need them to walk away believing. They can’t, because the whole model is built around speed. One prompt in, one deck out. The result looks great. But it reads generic and misses your thinking. And that’s AI slop.
We made a deliberate choice to go a different way.
Before building anything, Napkin collaborates with you to understand your goal, your audience, and the one thing you need them to walk away thinking. The answers to those questions shape everything that follows: the outline, the structure, the content, the visuals. Because the same topic, presented to different people for different reasons, should produce a completely different deck. Napkin is built around that belief.
But asking the right questions is only half of it. The other half is what happens next, and this is the part we care about most.
For every slide, Napkin generates multiple ways to present your ideas – from detailed to brief, text-driven to image-led. So you can choose how involved you want to be – go through every slide one by one, or pick a few and let Napkin handle the rest. Once it’s built, everything is editable – directly on the slide or through the agent. Nothing is locked. Nothing is final until you say it is.
We think about this as authorship. You are the author of your presentation. Napkin is your collaborator – it handles the mechanical, time-consuming work. The thinking always stays with you. The decisions always stay with you. The presentation that comes out at the end feels like you.
Why Napkin, and Why Now
We didn’t come to presentations from the outside. We came from years of building something that sits at the center of how people communicate visually. And through millions of users, we learned what it actually takes to make a visual communicate – rather than just decorate.
That learning is what shapes every visual inside a Napkin presentation. They are not stock photos or AI-generated images dropped in to fill space, but the same purposeful, meaning-carrying visuals Napkin has been building since the beginning – diagrams that explain relationships, charts that make data legible, frameworks that give ideas structure. And now, images that support the message. All of them are there because they make the point clearer. Not because the slide needed something in the corner.
Those years also gave us a clear vision for how a great presentation should be built. For a long time, getting it right wasn’t fully possible – the underlying technology wasn’t ready. That’s changed. LLMs have become genuinely strong at understanding context and reasoning through it. Agents can now handle complex, multi-step workflows intelligently. And image models have improved dramatically, opening up new ways to transform the visual quality of a presentation. These three things together are what made Napkin Slides possible – and what made now the right time to build it.
Come Build This With Us
Napkin Slides is live in beta today. I want to be honest about what that means.
It means the product works. You can create a real presentation, shaped around your thinking, with visuals that actually communicate your message. You can edit everything. You can download it and walk into a room with it. That part is ready.
It also means we’re not done. We know where the rough edges are. We’re working on them. And we’d rather work on them with you than without you.
If you’re already on Napkin, Slides is part of your subscription. No extra cost. Try it at napkin.ai/slides and tell us what you think.
If you’re new here – welcome. We’ve spent years helping people communicate their ideas visually. We’re just going deeper.
We’d love to hear what you make.
Here are some examples of slides generated in Napkin to give you an idea of the possibilities.
