There is a particular form of procrastination that feels productive: researching tools. You spend hours comparing Obsidian with Notion, Notion with Logseq, Logseq with Roam, Roam with something new that launched last Tuesday. You watch comparison videos, read migration guides, set up trial accounts. At the end of the week, you have not captured a single idea, but you feel like you have made progress because you are “choosing the right tool.”
This is avoidance dressed as preparation. And it is worth addressing directly, because the choice of tool matters far less than most people think — but the criteria for that choice matter far more than most people realise.
The tool trap
The tool trap works like this. You discover a new app and it looks perfect. The interface is clean, the features are exactly what you need, the philosophy aligns with your approach. You migrate your notes, set up your structure, spend a weekend customising everything. For a month, you are productive and happy.
Then the app changes its pricing. Or it adds features that clutter the interface. Or a shinier alternative appears. Or the company gets acquired and the product direction shifts. And you face a painful choice: stay with something that no longer excites you, or migrate again and lose another weekend — plus the mental energy of learning a new system.
People who change tools every year never build a mature system. They are perpetually in the setup phase, optimising structure instead of using it. They have beautifully configured empty systems rather than messy, living, productive ones.
The solution is not to pick the perfect tool. The perfect tool does not exist. The solution is to pick a tool that meets specific criteria and then commit to it long enough for compound value to accumulate. Two years with a good tool will produce more value than six months each with four excellent ones.
But not all tools are equally worthy of that commitment. Some are designed in ways that respect your long-term interests. Others are designed in ways that prioritise the company’s interests at your expense. Knowing the difference is crucial.
Five criteria that actually matter
When evaluating any tool for your second brain, these five criteria separate the durable choices from the risky ones.
Open or exportable formats. This is the single most important criterion. Your notes should exist in a format you can read and use independently of the tool. Markdown files stored locally on your computer are the gold standard — they are plain text, readable by any editor, and will still work in fifty years. Proprietary formats stored only in the cloud are the opposite: if the company disappears, your notes go with it.
Ask yourself: if this company shut down tomorrow, could I access all my notes in a usable format within an hour? If the answer is no, you are renting your knowledge system, not owning it.
Data portability. Even if the format is open, export needs to be straightforward. A tool that stores your notes as markdown files in a folder on your machine gets full marks. A tool that requires you to export notes one by one through a web interface fails this test even if the format is technically open. Look for bulk export, easy migration paths and a community that has built tools for moving data in and out.
Long-term viability. Is this tool backed by a sustainable business model? Venture-funded startups offering free tiers are exciting but risky — they need to grow revenue eventually, and that pressure often leads to pricing changes, feature bloat or acquisition. Tools backed by one-time purchases, reasonable subscriptions or open-source communities tend to be more stable over long horizons.
Consider the track record too. A tool that has been around for five years and maintained a consistent philosophy is a safer bet than one that launched six months ago with revolutionary promises. Time is the best filter for durability.
Community and ecosystem. A tool with an active community of users produces plugins, templates, tutorials and workarounds that extend its functionality far beyond what the core team builds. This ecosystem effect is self-reinforcing: more users attract more plugin developers, which attracts more users. It also provides a form of insurance — if the tool’s development slows, the community often picks up the slack.
Simplicity and speed. Your second brain tool should feel fast and lightweight. If opening the app takes five seconds, you will hesitate before capturing a quick idea. If the interface is cluttered with features you do not use, navigating your notes becomes a chore. The best tool is the one that gets out of your way and lets you focus on your thinking, not on managing the tool itself.
The current landscape
Without recommending a single winner — because your needs and preferences matter more than any generic recommendation — here is how the current major options map against these criteria.
Obsidian stores everything as local markdown files. Your notes live on your computer in folders you control. The plugin ecosystem is vast. It has been around since 2020 and has a sustainable business model based on optional paid sync and publish features. It scores exceptionally well on portability and format openness. The learning curve is steeper than simpler tools, but the investment pays off.
Notion is powerful, flexible and beautifully designed. However, your data lives on Notion’s servers in a proprietary format. Export exists but is imperfect — complex databases and relations do not translate cleanly to other formats. If Notion changes its pricing, gets acquired or shuts down, migration will be painful. It is excellent for team collaboration but carries meaningful lock-in risk for personal knowledge.
Logseq is open-source and stores notes as local markdown files, similar to Obsidian. Its outliner-first approach appeals to some users and frustrates others. The community is smaller but passionate. Being open-source provides strong long-term insurance — even if the company fails, the codebase persists.
Apple Notes, Google Keep, Microsoft OneNote are reliable and integrated into their respective ecosystems. They are excellent for quick capture but limited for serious knowledge management. Export options vary but none uses truly open formats. They work well as capture tools feeding into a more robust main system.
The general pattern is clear: tools that store your data locally in open formats give you the most control and the least risk. Cloud-first tools with proprietary formats offer convenience at the cost of dependence.
AI integration: what to look for
As AI becomes integral to knowledge work, how a tool integrates with AI capabilities matters increasingly. But not all AI integrations are created equal.
Local processing versus cloud processing. Some AI features send your notes to external servers for processing. Others run models locally on your device. If your notes contain sensitive personal or professional information, the distinction matters significantly. Understand where your data goes when you use AI features.
Your data for training. Read the privacy policy carefully. Some tools explicitly state they do not use your data to train their models. Others are vague or silent on the topic. Your personal knowledge base should remain private unless you explicitly choose to share it. Any tool that uses your notes as training data without clear consent is not respecting your interests.
Flexibility of AI integration. The AI landscape is evolving rapidly. A tool that locks you into a single AI provider is a tool that cannot adapt as better options emerge. Look for tools that allow you to connect different models, use local AI, or at least offer plugin architectures that let the community build integrations with new AI services as they appear.
AI as a feature, not the product. Be cautious of tools where AI is the primary selling point rather than an enhancement to solid underlying functionality. The knowledge management fundamentals — capture, organisation, linking, search — should work excellently without AI. AI should make a good tool better, not compensate for a weak one.
The safest approach is to choose your knowledge management tool based on the non-AI criteria and then add AI capabilities through plugins or integrations. This way, you can upgrade your AI layer independently as the technology improves, without having to migrate your entire knowledge base.
The best tool is the one you will actually use for years, not the one that impresses you for weeks. Prioritise portability, open formats and simplicity. Commit to your choice long enough for compound value to build. And remember that your notes are more valuable than any tool — the tool should serve them, not the other way round.
