Roam Research Review 2026: The Influence vs. The Reality

Roam Research matters. Without Roam, there is no Obsidian, no Logseq, no Tana, no Reflect. The core insight — that notes are a graph, not a hierarchy — is Roam's contribution to the field. That insight has been taken up, extended, and in many cases improved upon by the products Roam inspired. Roam itself has not kept pace.

I want to be careful in this review. Roam is a product that has real, passionate users who love it, and I am not interested in dismissing their love. I am interested in telling someone who is considering Roam in 2026 whether they should.

What Roam is

An outliner-based, web-only note-taking tool built around bidirectional links and block references. The daily-notes workflow — you open Roam, today's date is the default page, you type your day into bullets — is the canonical Roam experience. Every bullet is a block with its own unique ID; you can reference blocks in other pages; you can query blocks by tag, keyword, or structure.

What Roam does well

Daily notes flow. Still the cleanest daily-notes experience in any PKM tool. Opening Roam puts you on today's page. Typing happens instantly. Backlinks appear live as you write [[page-name]] references.

Block references. Roam's block-ref model is the version everyone else has tried to copy. Every bullet has a unique ID; you can embed that bullet anywhere else, and changes propagate bidirectionally. It is a powerful model for people who really think in blocks.

Queries. Roam's query syntax is genuinely powerful. You can write a query like {{[[query]]: {and: [[book]] [[2024]]}}} and get a live view of all blocks tagged both "book" and "2024." Obsidian's Dataview covers similar ground; Roam's is tighter inside its niche.

The community. The Roam community remains vibrant, creative, and generous — more so than the product perhaps deserves at this point.

Where Roam falls short

Pace of shipping. Roam has shipped little of substance in the last three years. The founder's focus has drifted. Features that were promised in 2021 have not arrived. Competitors have closed the gap and, in most cases, passed Roam on the features users care about.

Platform coverage. Web-only. Desktop is an Electron wrapper around the web app. Mobile is a subset of features, often laggy, sometimes broken. For a tool meant to hold a decade of notes, this is a serious gap.

Performance. Roam slows down measurably on large graphs (5,000+ pages). Search is not fast. Load times are not fast. Competitors like Obsidian are dramatically snappier.

Pricing. $15/month or $165/year ($500 for the "Believer" five-year plan). For a product with a slow shipping pace, this has aged worse every year. Obsidian is free. Logseq is free. Tana is $14/month for a product that feels materially more alive.

Lock-in. Roam's export is Markdown-ish but block references and queries do not survive the round-trip. Leaving Roam is technically possible and practically painful.

Who should use Roam

• Existing Roam users who are happy and productive — stay
• Researchers and academics whose daily-notes + block-ref workflow is specifically served by Roam
• Users who want the canonical outliner experience and are willing to pay for it

Who should not use Roam

• New users starting PKM in 2026 — try Obsidian or Logseq first
• Users who want local-first data ownership
• Users with large graphs (performance issues)
• Users who value a shipping roadmap

Bottom line

Roam Research changed the category and then stopped changing. The influence is outsized; the product has not aged well. If you are on Roam and happy, stay — the switch cost is real and the alternatives are not categorically better for your specific workflow. If you are new to PKM in 2026, start with Obsidian or Logseq. You will get most of what Roam does, for less money, with more momentum.