Tana and Roam Research are both personal knowledge management tools designed for researchers and knowledge workers who want to build a connected note system over time. Both use a block-based approach, but they are built around different core models: Roam Research is built on daily notes and bi-directional block references, creating a journal-first knowledge graph. Tana is built on supertags — typed node definitions that give every piece of information a schema — creating a structured database of knowledge nodes rather than a free-form networked journal.
Tana vs Roam Research: Core Comparison
| Feature | Tana | Roam Research |
|---|---|---|
| Core model | Supertag-typed nodes — structured knowledge graph | Daily notes + block references — journal-first hyptertext |
| Storage | Cloud (Tana servers) | Cloud (Roam servers) or self-hosted via Roam Depot |
| Data format | Proprietary (JSON export available) | EDN / Markdown export |
| Pricing | Free beta (limits); Tana+ $10/mo | $15/mo or $165/yr — no free tier |
| Learning curve | Moderate — supertag system takes time to design | Steep — block references and Datalog queries for power users |
| Query / filter | Supertag-based live searches and views | Datalog queries (powerful but complex) |
| AI features | Tana AI — inline commands, summaries, field suggestions | No native AI |
| Templates | Supertag schemas as templates — reusable node types | Templates via block embedding |
| Mobile | iOS/Android apps available | Mobile web — limited experience |
| Collaboration | Workspace sharing in beta | Multiplayer (shared graphs) |
| Plugin ecosystem | Limited (growing) | Moderate (Roam Depot) |
| Graph / visualisation | No graph view | Graph overview — visualise connections |
Why Researchers Choose Tana
Tana's supertag system is its central innovation for researchers. A supertag is a typed node definition — you define a "Paper" supertag with fields for author, year, title, DOI, research question, methodology, and key finding, then every note tagged as a Paper inherits those fields. This creates a personal research database where every paper you read becomes a typed object with consistent metadata, queryable and viewable in filtered lists, tables, or boards. For researchers who have previously maintained a separate reference database (Zotero, Notion database, spreadsheet), Tana unifies structured metadata with free-form synthesis notes in a single node.
Tana AI is built into the supertag architecture in a way other tools do not match: you can ask Tana to extract fields from a pasted abstract, suggest supertag types for a new note, or generate summaries that respect the schema of your node types. While the AI operates on your notes rather than source PDFs, the integration with your personal knowledge structure makes it more contextually relevant than a generic AI assistant overlay.
Researchers tackling multi-project knowledge management — multiple ongoing research threads, teaching preparation, grant writing, committee work — find Tana's typed node system more scalable than purely chronological systems. Each project, concept, and source becomes a node with defined relationships rather than a daily note entry.
Why Researchers Choose Roam Research
Roam Research's daily notes model aligns naturally with a researcher's actual work rhythm. Every day, you open a dated page and write your thoughts, reading notes, and ideas. Block references let you pull any block from any previous page into today's note — connecting yesterday's reading note about a theory to today's idea about an application — building a personal hypertext without conscious filing decisions. For researchers who find categorisation overhead a barrier to capturing ideas, Roam's friction-free daily writing model keeps the capture habit lower-cost.
Roam's bidirectional links are its most celebrated feature for knowledge work. Every time you mention a concept in double brackets, a backlink is created automatically — over time, clicking any concept shows every note that has ever mentioned it. Researchers building conceptual frameworks find this valuable: you discover unexpected connections between papers and ideas without having structured them in advance. The serendipitous association is a feature of Roam's design that Tana's structured typing does not fully replicate.
Roam's Datalog query language is the most powerful personal query system in any note-taking tool. For researchers comfortable with query syntax, Datalog allows arbitrarily complex retrieval: find all notes tagged with a specific author that also contain a specific concept, written in a date range, not yet reviewed. This power is available without the upfront supertag schema design that Tana requires.
The Synthesis Problem Neither Solves
Tana and Roam Research are both note systems that operate on what you have written about papers — your own processing and reflection. Neither works with the full text of papers you have not yet fully read and annotated. For researchers managing 40-100 papers for a literature review, the bottleneck is often not note organisation but initial synthesis: understanding what the papers collectively argue before deciding which are worth deep annotation.
Ponder addresses this gap: upload PDFs directly and ask synthesis questions across the full text without prior note-taking — "what are the main disagreements in this paper set?", "which papers address the population I'm studying?", "what methodological variations exist across these studies?" The answers come with citations to the original papers. The synthesis Ponder generates can then be imported into your Tana or Roam system as the starting point for your own structured notes.
Try Ponder for academic research →
Frequently asked questions
Is Tana better than Roam Research?
They serve different research styles. Tana is better for researchers who want structured, typed knowledge — consistent metadata across all papers and concepts, queryable as a personal database. Roam Research is better for researchers who want maximum capture speed and emergent connection — daily journaling where links grow organically without schema design overhead. Tana suits researchers with clear ontologies and repeatable workflows; Roam suits researchers who think by writing and discover structure retrospectively. Both are well-designed tools; the match depends on how you work rather than objective superiority.
Does Tana have a free plan?
Tana has a free tier (currently in beta) with some usage limits. Tana+ is $10/month for unlimited use. As of mid-2026, Tana is still in a relatively early stage compared to Roam's longer track record; the team has been gradually expanding access and features. Roam Research has no free tier — it costs $15/month or $165/year from day one. If cost is a constraint, Tana's free tier and a willingness to work within limits is a lower-risk starting point.
Can I migrate from Roam Research to Tana?
Migration between the two is manual and imperfect. Roam exports to Markdown and EDN formats. Tana can import Markdown, but the bidirectional link structure and block reference system of Roam does not directly translate to Tana's supertag model — links become mentions but lose their typed structure. In practice, most researchers who migrate choose to start fresh in Tana with new supertag schemas rather than attempting to port their full Roam graph. If you have years of Roam notes, treat the migration as redesigning your system rather than copying it.
What happened to Roam Research?
Roam Research is still active as of 2026, though its development pace has slowed compared to its 2020-2021 peak. The tool remains widely used among academics and researchers who built their knowledge systems around it; Roam Depot (the plugin ecosystem) continues to be maintained. Some users have moved to tools like Obsidian (more offline-friendly, larger plugin ecosystem), Logseq (open-source Roam alternative), or Tana (more structured). Roam has not shut down and continues to function as described.
See also: Tana Alternatives | Roam Research Alternatives | Obsidian vs Logseq | Best Note-Taking Apps for Researchers