Obsidian Peaked. Here’s What I’m Watching Instead.
Obsidian is one of the best note-taking tools ever made. I also think it peaked.
Not in the way people mean when they say a product peaked. The app didn’t get worse. The team didn’t stop shipping. The core experience of writing in markdown, linking your ideas, and owning your files is still the most principled stance in the entire category. None of that changed.
What changed is the trajectory. The growth story. The cultural momentum that carried Obsidian from “niche tool for developers” to “the note app smart people use.” That wave is receding, and I don’t think another one is coming.
I’ve been following the note-taking landscape for about four years now, across my own usage and across the teams I’ve worked with at three different companies. I’ve watched tools get adopted and abandoned at real companies. The pattern is always the same: the cultural moment creates the adoption surge, the product sustains (or doesn’t sustain) the retention, and eventually the energy moves somewhere else. Obsidian is in the third phase.
Here’s what I’m watching instead.
TLDR
Obsidian peaked on cultural momentum, not product quality. The “second brain” wave that carried its adoption surge is winding down. Plugin fatigue, vault complexity, and a power-user identity that intimidates newcomers have created a growth ceiling. The product remains excellent for its existing users. But the energy is moving to tools that solve different problems in fundamentally different ways. The four I’m tracking: Tana (structured data through supertags), Heptabase (visual-spatial thinking on whiteboards), Capacities (object-based notes without the folder debate), and SiYuan (open-source, local-first, no plugin dependency). If you’re happy on Obsidian, stay. If you’re feeling the friction and wondering whether something better exists, keep reading.
Note: The opinions here are my own. I’m not paid or incentivised by any tool mentioned below. No affiliate links, no sponsored placements. Product links are for your convenience, not my revenue.
The case for “peaked”
Let me be precise about what I mean, because this is going to annoy people.
Obsidian the product is still excellent. The team ships consistently. The markdown-first, local-first philosophy is still the most honest approach to data ownership in the category. If you’re an Obsidian user who’s happy, nothing I’m about to say should make you switch. Stay where you are.
But “peaked” isn’t about product quality. It’s about adoption trajectory. And the adoption trajectory has three problems that I don’t think the Obsidian team can solve, because two of them aren’t product problems at all.
The “second brain” wave is receding
Obsidian’s biggest growth period coincided with the Zettelkasten and “second brain” cultural moment. YouTube was full of vault tours. Reddit was full of workflow showcases. Productivity Twitter was full of people debating folder structures versus MOCs versus tags.
That moment created enormous awareness for Obsidian. It was the right tool at the right time for a specific cultural energy. But cultural waves don’t last. The “build a second brain” enthusiasm has cooled. The YouTube vault-tour era is winding down. The people who were going to get excited about linked thinking already got excited about it. The next cohort of note-taking adopters isn’t looking for the same thing.
This isn’t Obsidian’s fault. But it is Obsidian’s ceiling.
Plugin fatigue flipped from strength to friction
Here’s the thing that nobody on the Obsidian subreddit wants to hear: the plugin ecosystem, which was Obsidian’s single biggest selling point, has become its single biggest source of friction.
Two years ago, the energy around Obsidian plugins was “look what I discovered.” Someone would share a plugin that solved a specific problem in a clever way, and people would get excited about the possibilities. That energy was infectious. It pulled people in.
Now, the energy has shifted to “here’s my 47-plugin setup.” The community content has become intimidatingly complex. A new user opens Reddit, sees someone’s elaborate Dataview queries and Templater scripts and custom CSS snippets, and thinks: I just wanted to take notes.
I watched this happen in real time with someone on my team. She downloaded Obsidian after I recommended it, spent a weekend trying to replicate one of those elaborate setups she saw on YouTube, got frustrated, and went back to what she was using before. She’s smart. She’s technical enough. But the gap between “install Obsidian” and “have a working system” has grown, and the community’s escalating expectations of what a vault should look like are widening it.
The plugin ecosystem also creates a maintenance burden that compounds over time. Plugins break with updates. Plugins conflict with each other. Plugins get abandoned by solo developers who moved on. My own vault has had three plugin-related issues in the past six months that required me to troubleshoot things I shouldn’t have to troubleshoot in a note-taking app. Each one was minor. Collectively, they add up to a feeling I associate with tools that are starting to lose me.
The power-user identity became the growth ceiling
Obsidian’s most passionate users are also its adoption ceiling. This is the part that’s genuinely hard to fix.
The community is so enthusiastic, so technically minded, and so specific about workflows that Obsidian’s public identity has calcified around the power user. The tool’s reputation is “the complicated one.” Not because it has to be complicated. Because the loudest voices in the community are the ones with the most elaborate setups, and that’s what new users encounter first.
Compare this to how Notion grew. Notion’s community created templates that made the tool feel accessible. You could download a template and have a working system in ten minutes. Obsidian’s community creates showcases that make the tool feel aspirational but intimidating. You can watch a vault tour and have no idea how to build something like that yourself.
This isn’t a criticism of Obsidian’s community. They’re genuine enthusiasts sharing what they love. But the effect on adoption is real. The people who would benefit most from Obsidian’s core experience (simple markdown notes with links) never get to that core experience because the on-ramp is buried under complexity they didn’t ask for.
The product moves that worry me
Beyond the cultural dynamics, a few product decisions signal where Obsidian is heading, and I’m not sure it’s the right direction.
Obsidian Canvas launched as a visual thinking surface. It’s fine. But it feels like a feature that was built because other tools have whiteboards, not because the Obsidian user base was begging for one. Heptabase does visual thinking natively and better. Canvas feels bolted on.
Obsidian Publish at $8/mo ($10/mo monthly) is priced for a feature most users will never need. The few who do need web publishing have better options.
Obsidian Sync at $4/mo (annual) is honestly reasonable. But it creates a weird incentive split. The core pitch is “it’s free and your files are local.” The reality for anyone who uses more than one device is “it’s free plus $48/year, or you can wrestle with iCloud sync and hope nothing breaks.” That’s not a dealbreaker, but it softens the “free and local” narrative that drives adoption.
The new Bases feature (database views inside Obsidian) feels like the team noticed that Notion’s databases are popular and decided Obsidian should have them too. But databases inside a markdown-first tool feel like a philosophical compromise. Obsidian was the tool that said “you don’t need databases, you need links.” Now it’s saying “actually, maybe you need databases too.” That’s not confidence. That’s hedging.
The counter-argument (and where it’s right)
Someone reading this is already composing a comment about data ownership. And they’re right.
Obsidian’s local-first, you-own-your-markdown philosophy is still the most principled stance in the note-taking category. Your notes are plain text files on your hard drive. No proprietary format. No lock-in. No “what happens to my data if this company shuts down” anxiety. Every tool I’m about to recommend fails this test to some degree.
That matters. For some people, it matters more than anything else I’m about to say, and those people should stay on Obsidian. Full stop.
But data ownership, by itself, isn’t enough to sustain a growth story. Most people don’t choose note-taking tools based on file format philosophy. They choose based on how the tool feels when they use it every day. And the daily experience of Obsidian, for a growing number of people, involves more friction than it used to.
What I’m watching instead
These aren’t “Obsidian alternatives” in the way that phrase usually gets used. They’re not trying to be Obsidian with a different coat of paint. They represent genuinely different ideas about what a note-taking tool should do. That’s what makes them interesting.
Tana: the most ambitious rethink of what a “note” is
Tana is hard to explain without sounding like you’re overselling it. The core idea is supertags: you tag a node with something like #meeting or #project, and that node gains structured fields. A meeting has an agenda, attendees, action items. A project has a status, an owner, a deadline. Your daily notes become a living database without any extra work.
This sounds like Notion’s databases. It isn’t. Notion makes you choose: is this a page or a database entry? Tana makes everything both at once. Every bullet in your outline is simultaneously a piece of free-form text and a structured data object. You don’t switch contexts between “writing mode” and “database mode.” They’re the same mode.
I’ve been watching Tana closely for about a year. The learning curve is steep. Genuinely steep, not “spend an afternoon” steep. The mobile app needs work. The free tier is limited (500 AI credits, only 5 supertags). Plus runs $10/mo. The community is small but intense in the way that early Obsidian’s community was intense.
What I keep coming back to is that Tana is the only tool I’ve seen that makes me rethink the unit of note-taking itself. Not “where do I put this note” or “how do I find this note later.” Instead: “what is this note, structurally, and how does it connect to everything else I know?” That’s a different question. And it’s the right question.
Best for: people who think in systems. Researchers, PMs, anyone who needs their notes to be queryable without building elaborate workarounds.
Think twice if: you want something simple. Tana’s power is its complexity, and that complexity isn’t optional.
Cost: Free (limited), Plus $10/mo, Pro $14/mo.
Heptabase: the tool that solved “I have 2,000 notes and can’t find the insight”
Every note-taking tool eventually runs into the same problem: at scale, notes become a graveyard. You wrote something six months ago. It was useful when you wrote it. Now it’s buried in a vault of 2,000 other notes and you’ll never find it again unless you remember the exact words you used.
Heptabase is the only tool I’ve seen that genuinely addresses this. Notes live on infinite whiteboards. You arrange them spatially. You cluster related ideas. You draw connections visually. The spatial layout means your brain has a second way to find things: not just “what was it called” but “where was it on the board.”
This sounds gimmicky until you use it. I spent a few weeks with Heptabase earlier this year and the thing that surprised me is how much the visual layout changes the way you think about your notes. Seeing three related cards next to each other triggers connections that a text search never would. It’s the difference between reading a list of ingredients and seeing them laid out on a counter.
No free tier. That’s the catch. It’s $8.99/mo (annual) with a 7-day trial, and if you’re used to Obsidian’s “free forever” ethos, paying nine bucks a month for a note app feels like a lot. There’s a lifetime option at $659 for committed users. The learning curve is real but different from Tana’s. It’s not conceptually difficult. It’s just a different paradigm, and the adjustment takes a couple of weeks.
Best for: researchers, writers, anyone who processes complex information and needs to see relationships between ideas, not just link them.
Get Justine’s stories in your inbox
Join Medium for free to get updates from this writer.
Think twice if: your notes are mostly quick captures and to-do lists. Heptabase’s spatial model is built for thinking, not task management.
Cost: $8.99/mo (annual), $11.99/mo (monthly), $659 lifetime. 7-day free trial. No free tier.
Capacities: the one quietly converting people who left Obsidian
Capacities is the tool I keep hearing about from people who tried Obsidian and bounced off it. The premise: everything is a typed object. A person is a person. A book is a book. A meeting is a meeting. Each type has its own properties and relationships, and the tool builds a knowledge graph from those objects automatically.
This sidesteps the folder-versus-tags-versus-links debate entirely, which is the debate that consumes an unreasonable amount of Obsidian community energy. In Capacities, you don’t organise notes. You define what things are, and the organisation emerges from the structure.
The free tier is genuinely generous. Unlimited notes, unlimited objects, sync across devices, 5GB of media storage. Pro runs $9.99/mo (annual) and adds AI features and unlimited storage. A guy on my team started using it a few months ago after getting frustrated with his Obsidian vault, and his exact words were “I stopped thinking about where to put things and started just putting them.” That’s the pitch, and it actually delivers.
The catch: no plugin ecosystem, which means if Capacities doesn’t do what you need out of the box, you’re stuck. The mobile app is still catching up. And there’s no collaboration, so this is a solo-thinker tool only. But for individual knowledge work, the object-based model is the most genuinely different approach to personal notes I’ve seen since Obsidian itself.
Best for: people who got paralysed by Obsidian’s organisational choices. People who want structure without setting it up themselves.
Think twice if: you need team collaboration or a rich plugin ecosystem.
Cost: Free (genuinely usable), Pro $9.99/mo (annual), Believer $12.49/mo (annual).
SiYuan: the open-source alternative almost nobody in the English-speaking world is talking about
This is the discovery pick. SiYuan is an open-source, local-first, privacy-focused note-taking tool with block-level referencing, bidirectional links, WYSIWYG markdown editing, and database views. It does roughly 80% of what Obsidian does without requiring a single plugin.
Read that again. Eighty percent of Obsidian’s functionality, out of the box, with no plugin dependency. Block-level referencing (not just page-level links, which is all Obsidian offers natively). WYSIWYG editing (Obsidian’s editing view is good but still markdown-first). Built-in flashcards. Built-in PDF annotation. Math equation support. All local, all yours.
SiYuan is huge in the Chinese-speaking PKM community and is slowly gaining traction in English-speaking circles. The GitHub repo has over 25,000 stars. The development pace is fast. The pricing model is remarkable: free for all core features, $64 one-time for pro features (S3/WebDAV sync), $148 one-time for official cloud sync with 8GB storage. One-time. Not monthly. Not annually. You pay once.
The catch: the English documentation is thin. The community is predominantly Chinese-speaking. The UI, while functional, lacks the polish of Obsidian or Capacities. Some of the design patterns feel unfamiliar if you’ve spent years in Western note-taking tools. And while it imports Obsidian markdown vaults, the link format requires updating.
But if Obsidian’s value proposition is “local-first markdown notes with powerful linking,” SiYuan offers a version of that proposition that doesn’t require you to become a plugin curator. For a growing number of people, that’s the upgrade.
Best for: Obsidian users who love the philosophy but are tired of the plugin maintenance. Anyone who values one-time pricing over subscriptions.
Think twice if: you need polished English documentation or an active English-speaking community. You’ll be somewhat on your own.
Cost: Free (core features), $64 one-time (Pro), $148 one-time (Subscription with cloud sync).
Tools I looked at and passed on
Logseq. Outliner-first, which limits the audience. The development pace has slowed noticeably. The database rewrite has been in progress for a long time. I’d wait before committing.
Reflect. Clean and focused, but hasn’t differentiated enough from the broader field. Being “simple and nice” is an identity, not a moat.
Anytype. Ambitious local-first architecture with a genuinely interesting approach to data types. But the daily experience still feels rough. Promising for 2027. Not ready for 2026.
Mem. Bet everything on AI-first note-taking. The bet hasn’t paid off yet. The product feels like it’s searching for its identity.
The pricing reality
Obsidian still wins on raw value for the free tier. Nobody beats “every feature, free forever, no sign-up.” But SiYuan’s one-time pricing model is remarkable in a world where everything is a subscription, and Capacities’ free tier is the most usable I’ve tested in the category.
So should you leave Obsidian?
Probably not. And I don’t say that to hedge.
If you have a vault that works, if your plugins are stable, if your workflow is dialled in, switching has a real cost. Weeks of migration. Months of adjustment. The muscle memory alone takes ages to retrain. The gain would need to be substantial to justify it, and for most working Obsidian setups, it isn’t.
Where I’d think about moving:
If your vault has become a maintenance project instead of a thinking tool. If you spend more time configuring Obsidian than using it. If you’ve tried three times to get a system working and it keeps falling apart because a plugin broke or a Dataview query stopped rendering.
That’s the signal. When the tool starts consuming more attention than the work you’re trying to do inside it, the tool has become the problem.
My trajectory call: Obsidian will remain a strong tool with a devoted user base for years. It’s not going anywhere. But the growth story is over. The next wave of note-taking adoption will be driven by tools that are either simpler (Capacities), more visually intuitive (Heptabase), more structurally ambitious (Tana), or less dependent on community-maintained plugins (SiYuan). Obsidian’s era as the default recommendation for “what note app should I use” is ending. What replaces it isn’t one tool. It’s four different answers to four different questions that Obsidian tried to answer with one architecture.
Rule of thumb: if you’re spending more time on your note-taking system than on the notes themselves, the system is the problem. The best tool is the one that disappears.
FAQ
1. Is Obsidian still worth starting with in 2026?
For a specific type of user, yes. If you value data ownership above everything else, if you’re comfortable with markdown, and if you’re disciplined enough to keep your setup simple, Obsidian is still one of the best tools in the category. The problem is that most new users aren’t disciplined enough to keep it simple, because the community content pushes them toward complexity they don’t need.
2. What’s the best Obsidian alternative for someone who isn’t technical?
Capacities. The object-based model removes the organisational decisions that trip up non-technical users. You don’t need to decide on a folder structure, a tagging system, or a linking convention. You just define what things are, and the tool handles the rest. The free tier is good enough to evaluate properly.
3. Is Tana ready for daily use?
Getting there. The core experience is solid if you’re willing to invest the setup time. The learning curve is the steepest of any tool in this piece. Budget two to three weeks before it clicks. The mobile app is functional but limited compared to the desktop experience. I’d recommend it for people who enjoy building systems, not people who want something that works out of the box.
4. Is Heptabase worth $9/mo when Obsidian is free?
If you do research-heavy or synthesis-heavy work, yes. The visual whiteboard approach solves a real problem that text-based tools don’t address: seeing relationships between ideas spatially. If your notes are mostly quick captures and task lists, no. You’d be paying for a thinking tool and using it as a notepad.
5. Should I worry about SiYuan being a Chinese-developed tool?
The code is open source and auditable on GitHub. The data is local-first. If you self-host sync, nothing touches a third-party server. The question is really about community support and documentation in English, which is currently thin. If you’re comfortable navigating a smaller English-speaking community and occasionally translating documentation, SiYuan is solid. If you want hand-holding, look elsewhere.
6. Can I migrate my Obsidian vault to any of these tools?
Capacities and SiYuan both import markdown files, though link formats will need updating. Heptabase imports markdown but the experience is different enough that you’ll likely want to rebuild rather than migrate. Tana is outliner-based, so a direct import from a markdown vault isn’t practical. Budget time for any migration. It’s never as quick as the import tool suggests.
7. What do you mean by “plugin fatigue”?
The pattern where you install plugins to extend a tool’s functionality, then spend increasing amounts of time maintaining, updating, troubleshooting, and replacing those plugins. The tool’s core value gets buried under a layer of community-maintained add-ons that can break, conflict, or get abandoned. It’s a form of technical debt applied to a note-taking app.
8. Isn’t this just the normal lifecycle of any popular tool?
Partially. Every tool has a peak adoption moment. What’s specific to Obsidian is that the peak was driven by a cultural movement (second brain, Zettelkasten, PKM YouTube) rather than by a product feature or a market shift. Cultural movements are harder to sustain or replace than product advantages. When the cultural wave recedes, the tool loses its primary awareness engine.
9. Are any of these tools good for teams?
Honestly, none of them are great for teams. Tana has workspace sharing on paid plans. Capacities has no collaboration features. Heptabase is primarily a solo tool. SiYuan can do shared vaults with its sync service but it’s not built for real-time collaboration. If you need team note-taking, you’re still looking at Notion, Coda, or Slite.
10. What’s your actual daily driver for notes right now?
I’m splitting time between Obsidian (which I’ve been on for years and have too much invested to leave casually) and Capacities (which I’m evaluating seriously as a replacement for my personal notes). I’m watching Tana and SiYuan closely but haven’t committed to either for daily use. That might change by end of year.
