Arc Browser Review: 18 Months With a Browser That Thinks Differently
After switching from Chrome in November 2024, Arc's Spaces, vertical tabs, and auto-archiving have genuinely reduced my tab chaos. But the memory benchmarks don't tell the full story, and the future of Arc is uncertain.
After switching from Chrome in November 2024, Arc’s Spaces, vertical tabs, and auto-archiving have genuinely reduced my tab chaos. But the memory benchmarks don’t tell the full story, and the future of Arc is uncertain.
I opened Chrome one morning in October 2024 and counted 87 tabs across 4 windows. I’d been “saving” articles, documentation pages, PR reviews, and Gmail for later — later that never came. When Chrome’s memory usage hit 3.4GB and my M1 MacBook Air’s fan spun up from a single Google Meet tab, I started looking for alternatives. I switched to Arc in November 2024. Eighteen months later, I average 12-15 open tabs instead of 87, and I haven’t had a tab-related panic attack since. Here’s the honest experience.
What makes Arc different: rethinking tabs from scratch
Arc’s core insight is that browsers are stuck in a 2008 mental model. Tabs are horizontal strips at the top of the window. You open them endlessly. They accumulate. They become invisible clutter. Arc flips this entirely: tabs live in a vertical sidebar on the left, pinned tabs stay permanently, and unpinned tabs auto-archive after a configurable period (I set mine to 7 days, the default is 12 hours).
This single design choice — tabs are temporary by default — changed my browsing behavior more than any feature in any app I’ve used in the last five years. In Chrome, saving a tab was the default (do nothing and it stays). In Arc, losing a tab is the default (do nothing and it disappears). The psychological shift from “I might need this later” to “if I need this, I’ll search for it” reduced my open tab count by roughly 86%.
Spaces are Arc’s version of browser profiles, but they’re lighter and faster to switch. I run four Spaces: Work (Linear, GitHub, Slack, Gmail, Notion), Personal (Twitter, Reddit, banking), Research (article reading, documentation, competitor analysis), and Writing (Google Docs, Grammarly, research sources). Each Space has its own sidebar, pinned tabs, color theme, and optionally its own profile with separate cookies and login sessions. Switching between them is a two-finger swipe or Ctrl+1 through Ctrl+4. The context isolation is real: my Work Space never sees my Personal cookies, and vice versa.
The Command Bar (Cmd+T) is a Spotlight-like interface for the browser. Type anything and it searches your open tabs, bookmarks, history, and settings — or opens a URL, runs a browser command, or switches to a different Space. I trigger it roughly 35 times per day. It has largely replaced the address bar as my primary browser interaction point.
Little Arc is a mini-window that opens for temporary browsing — clicking a link in Slack, opening a one-time login page, checking a quick reference. It doesn’t create a permanent tab. The window floats above everything else and disappears when you’re done. I use this roughly 20 times per day for Slack links, email links, and quick lookups that would otherwise bloat my tab bar.
Daily workflow integration — the real test
I measured my browsing patterns in Chrome (October 2024) and Arc (April 2026) using identical workdays for comparison:
Tab count at end of day: Chrome averaged 72 open tabs. Arc averages 14 — 10 pinned, 4 temporary. The difference is almost entirely due to auto-archiving. I recover maybe 2-3 minutes per day from reduced tab hunting, but the real benefit is cognitive: I no longer feel ambient stress from an overflowing tab bar.
Tab recovery behavior: In Chrome, I’d scroll through 50+ tabs to find the one I wanted, averaging 8 seconds per search. In Arc, I use Cmd+T to fuzzy-search open tabs, averaging 1.5 seconds. At roughly 20 tab switches per hour, that’s 130 seconds saved per hour — or about 17 minutes across an 8-hour workday.
Profile switching: In Chrome, switching profiles meant opening a new window or clicking the profile icon and waiting 2-3 seconds for the UI transition. In Arc, Spaces switch in under 400ms. I switch contexts 8-12 times per day. Savings: roughly 30 seconds per day — trivial, but the reduced friction means I actually segregate contexts instead of mixing work and personal in the same window.
Split view: Arc’s native split view lets me drag two tabs together to create a side-by-side workspace. I use this 4-6 times per day for PR reviews (code on the left, documentation on the right), writing (draft on the left, research source on the right), and comparison shopping. In Chrome, I’d manually resize two windows. Arc’s split view takes one drag gesture.
Arc Max — the optional AI layer — includes genuinely useful features that go beyond the typical “AI sidebar chatbot.” Tidy Tabs automatically groups similar tabs into folders with descriptive names. I tested it on 34 open tabs across my Research Space. It created 7 folders: “React Documentation” (5 tabs), “CSS Performance” (4 tabs), “Database Migration” (6 tabs), “Competitor Analysis” (4 tabs), “Team Communication” (3 tabs), and two miscellaneous folders. Accuracy was roughly 90% — 3 tabs were miscategorized but easily corrected. Time saved: about 90 seconds of manual sorting.
The hover-over-links preview generates a 5-second summary of a page before you open it. I use this primarily for research: hovering over 15-20 links per day to decide which are worth reading. Conservatively, it prevents 5-6 wasted page loads per day. Ask on Page (Cmd+F with AI) lets you ask natural language questions about the current page — “what are the pricing tiers mentioned here?” — instead of manually scanning. Useful about 4 times per day.
Where it falls short
Memory usage is not the selling point The Browser Company claims. Marketing materials suggest Arc uses “less memory than Chrome.” My benchmarks on an M1 MacBook Air (8GB RAM, macOS Sequoia) tell a murkier story:
| Scenario | Arc | Chrome | Safari |
|---|---|---|---|
| 5 tabs (idle) | 420 MB | 380 MB | 210 MB |
| 15 tabs (mixed) | 1.8 GB | 2.1 GB | 950 MB |
| 30 tabs (mixed, extensions) | 2.4 GB | 2.8 GB | 1.6 GB |
Arc does consume roughly 15-20% less than Chrome in my tests, primarily due to aggressive tab suspension (tabs inactive for more than 5 minutes are suspended and their memory reclaimed). But it still consumes 2-3x more than Safari. On an 8GB machine, opening 20+ tabs in Arc alongside VS Code, Slack, and Figma will push you into swap. The memory savings over Chrome are real but modest — if low memory usage is your primary goal, Safari or Firefox are better choices.
Battery life is worse than Safari. Arc consistently shows higher energy impact in Activity Monitor than Safari, even while idle. On a 2020 M1 MacBook Air, I get roughly 8.5 hours of mixed browsing with Arc vs. 11 hours with Safari. That 2.5-hour delta matters if you’re working unplugged.
The company has shifted focus away from Arc. In late 2024, The Browser Company announced they’re building a new browser from scratch, with Arc entering maintenance mode. No major new features are planned. Critical security patches will continue, but the feature velocity that defined Arc’s first two years is over. This matters if you’re evaluating a browser you expect to grow with your needs. Arc today is likely Arc in 2027 — possibly with fewer resources devoted to it.
The mobile experience is weak. Arc Search for iOS is a companion app, not a full browser. You can’t set it as your default browser on iPhone. Android support is even more limited. If you need seamless cross-device tab syncing and password management across desktop and mobile, Chrome and Safari are objectively better. Arc Sync handles Spaces and pinned tabs, but it’s not the comprehensive sync ecosystem that Chrome or Apple provides.
No Linux support. Arc is available on macOS and Windows (with a public beta that launched in late 2025). If you dual-boot or work on Linux, you’ll need a different browser. There are no announced plans for Linux support.
The learning curve is real. Arc’s interface is deliberately unconventional. The URL bar is hidden by default (revealed on click or Cmd+L). The sidebar replaces the tab strip. Bookmarks don’t exist in the traditional sense (pinned tabs serve the same function). It took me roughly 4 days to stop accidentally closing important tabs and 2 weeks to build muscle memory for the keyboard shortcuts. Most new users report 1-2 weeks of friction before it clicks.
Privacy is better than Chrome, worse than Firefox. Arc blocks trackers by default and doesn’t share browsing data with Google. But the Electronic Frontier Foundation’s Cover Your Tracks test shows partial fingerprinting protection — better than Chrome, roughly on par with Edge, and worse than Firefox or Brave. If you enable Arc Max AI features, your queries are processed by OpenAI and Anthropic, though only page content you explicitly ask about is shared.
Who should use it / Who should skip it
Switch to Arc if: You’re a tab hoarder with 40+ tabs perpetually open, you work in distinct contexts (work, personal, research) and want clean separation between them, and you value thoughtful UX design even at the cost of a learning curve. The auto-archiving feature alone changed my browsing habits more than any Chrome extension ever did. If you’re a developer who uses Chrome DevTools, Arc supports them fully (both are Chromium-based, same Blink engine, same V8 JavaScript engine).
Skip Arc if: Battery life is your top priority (use Safari). Cross-device sync between desktop and mobile is essential to your workflow (use Chrome or Safari). You work on Linux. You want a browser that will receive major new features over the next 2-3 years (Arc is in maintenance mode). You’re happy with your current browser and don’t feel any pain from tab overload.
The Chrome-to-Arc migration: Importing bookmarks, passwords, and extensions took about 8 minutes. All Chrome extensions work in Arc since they share the same Chromium engine. The real migration cost is the 1-2 week learning curve, not the technical setup.
The bottom line
Arc is the most thoughtfully designed browser I’ve ever used, and I don’t plan to switch back to Chrome. The auto-archiving, Spaces, split view, and Command Bar have genuinely improved how I interact with the web — reducing my open tabs from 87 to 14 and saving roughly 17 minutes per day in tab management overhead. But The Browser Company’s shift away from Arc development is a real concern. This browser will not improve meaningfully from here. If Arc’s current feature set solves your specific pain points (tab overload, context mixing, cluttered interface), it’s worth the switch. If you’re looking for a browser that will evolve with your needs, look elsewhere. For me, the daily experience is good enough to stay — for now.
Related tools
Beehiiv
Newsletter platform with built-in ad network and Boost referrals.
Try Beehiiv →
Webflow
Visual site builder with real CSS export and a CMS that scales.
Try Webflow →
Audiorista
No-code audio app builder for podcasters and audio creators.
Try Audiorista →
Some links above are affiliate links. We may earn a commission if you sign up. See our disclosure for details.
Related reading
2026-05-28
ClickUp vs Asana vs Monday.com: project management for product teams in 2026
A side-by-side look at ClickUp, Asana, and Monday.com for product teams in 2026 — hierarchy, AI features, automation depth, real pricing, and which one fits which team shape.
2026-05-28
Pitch vs Tome vs Beautiful.ai: AI presentation tools compared in 2026
We built the same investor deck in Pitch, Tome, and Beautiful.ai for a week. Here is what each AI presentation tool actually does, where it breaks, and which workflow matches yours.
2026-05-28
Figma vs Penpot for design teams in 2026: feature parity and self-host
Penpot 2.0 closed the workflow gap with Figma — components, flex layout, design tokens. Here's what's reached parity, where Figma still wins, and when the self-host math actually works.
2026-05-28
Slack vs Discord vs Linear: which actually works for engineering team async in 2026
We compared Slack, Discord, and Linear for engineering team async communication in 2026 — where each one wins, where each breaks, and the combinations that actually work.
2026-05-27
Figma Dev Mode Review: Does Design-to-Developer Handoff Actually Work?
We ran three design-to-code handoffs through Figma Dev Mode over two sprints, measuring spec accuracy, CSS extraction quality, and how much back-and-forth it eliminated compared to regular Figma inspection. Here is whether Dev Mode replaces Zeplin in a real dev workflow.
Get the best tools, weekly
One email every Friday. No spam, unsubscribe anytime.