Most note apps assume you already know what you want to say. You open a blank document, type, and file it somewhere. Heptabase starts from the opposite assumption: that the hard part of knowledge work is the stretch before you understand something — when ideas are scattered, half-formed, and don't yet connect. It tries to give that stage an actual workspace instead of a folder.
We ran three real projects through it over a few weeks — a literature review, a product decision with competing constraints, and a pile of meeting notes that needed to become a plan — to see where the visual model helps and where it gets in the way.
How Heptabase actually works
The atomic unit is a card: a small markdown note. Instead of living in a folder tree, cards get dropped onto infinite whiteboards, where you arrange them spatially, draw arrows between them, group them into sections, and nest whiteboards inside cards.
A few mechanics matter more than the marketing screenshots suggest:
- One card, many places. A card can sit on several whiteboards at once. Edit it in one spot and it updates everywhere — there's a single source of truth, not copies that drift apart.
- The sidebar is the spine. It holds your Map, a Journal of daily notes, Tags, and a Card Library of everything you've ever written. The whiteboards are the thinking surface; the sidebar is how you get back to anything.
- Tags double as lightweight databases. Add properties like status, date, or rating to tagged cards, then filter them in table or board views. It's not as deep as a real relational database, but it covers most personal tracking.
- PDFs annotate in place. Highlights from a PDF become cards you can pull straight onto a whiteboard next to your own notes — useful when you're reading sources and building an argument at the same time.
- Local-first, cross-platform. Data lives on your device and syncs across Mac, Windows, Linux, iOS, Android, and web. It works offline, which matters if you draft on planes or trains.
The learning curve is real but short. The first hour feels like too much freedom. By the second or third session the spatial layout stops being decorative and starts doing work.
Where the whiteboard model earns its keep
The payoff shows up when you're synthesizing, not just capturing. A linear document forces you to commit to an order before you understand the material. A whiteboard lets you hold twenty loose ideas in view, shove the related ones together, and notice the structure after it emerges rather than guessing it up front.
For the literature review, this was the difference-maker. Each paper became a card with its key claims. Pulling those cards into clusters on a whiteboard made the disagreements between sources visible in a way a bulleted outline never did — two cards pointing at the same conclusion from opposite premises is obvious when they're sitting six inches apart. The product decision worked the same way: options as cards, constraints as cards, arrows showing which constraint killed which option.
The spatial memory is underrated. Weeks later you remember where on the canvas an idea lived — top-left, near the messy cluster — and that location recall pulls the content back faster than searching for a keyword you've half-forgotten.
Don't start on a blank whiteboard. Capture into the Journal first — fast, linear, no layout decisions. Once you have a pile of raw cards, then open a whiteboard and promote the ones worth arranging. Treating the canvas as a synthesis step rather than a capture step is what keeps it from turning into clutter.
Where it frustrates
Heptabase is opinionated, and the opinions don't fit everyone.
Spatial freedom is also spatial overhead. A big whiteboard can sprawl into a mess that takes as long to tidy as the thinking it was supposed to support. If your work is mostly capture-and-retrieve — clip an article, find it later — the canvas is friction you don't need.
It's a thinking tool, not a team wiki. Sharing exists, but real-time multiplayer collaboration is limited compared to a tool built for teams from day one. If you need ten people editing the same doc, this isn't it.
Markdown without the plugin ecosystem. You get clean markdown and good linking, but not the sprawling community-plugin marketplace that something like Obsidian offers. What ships is what you get, and the team's roadmap is the only path to new capabilities.
Mobile is capture-first. Jotting a card on your phone is fine; arranging a whiteboard on a small screen is awkward. Treat mobile as inbox, desktop as workshop.
Subscription only. There's no permanent free tier — a free trial, then paid. Pricing sits at around $8.99/month billed annually (roughly $11.99 month-to-month). For a single-purpose thinking app, that's a real line item to justify against tools you already pay for.
That last point is the honest tension. If you already live in a structured workspace, the question isn't whether Heptabase is good — it's whether visual synthesis is a big enough part of your work to warrant a second tool.
A workflow that worked for us: think on the whiteboard, then move the conclusion — the cleaned-up decision or outline — into a structured doc tool for sharing and reference. Heptabase is where the argument gets built; the doc tool is where it gets published.
Who it's actually for
Reach for Heptabase if a meaningful slice of your work is open-ended synthesis: researchers, writers structuring long pieces, students mapping a field, anyone who routinely faces a pile of related-but-tangled material and has to find the shape in it. The whiteboard-plus-card model is genuinely better than a linear editor for that specific job.
Skip it if your notes are mostly atomic capture, if you need team collaboration, or if you want a deep plugin ecosystem. None of those are what it's built to do, and paying for a tool to use it against its grain rarely sticks.
Originally published at pickuma.com. Subscribe to the RSS or follow @pickuma.bsky.social for new reviews.
Top comments (0)