Toby — Product Compass

Vision#

A browser where nothing you cared about is ever lost in the noise. You open a new tab and your work is right there — visible, recognized, ready to pick up. The calm of a well-set desk, in the surface you already live in.

Axioms — what makes Toby 10× (4)#

1. Visual tangibility. Tabs are real objects on a board — favicons, colors, spatial layout — not text in a hidden menu. Why this is a moat: bookmarks are abstract and invisible; Toby makes saved content recognizable by glance. Brains pattern-match faster than they read. Evidence: paid-user survey language ("I can see everything I care about"); brand soul axiom 1 (see: product/strategy/soul.md).

2. Ambient, not invocational. Your saved world is the first thing you see every time you open the browser. No app to launch, no menu to dig through. Why this is a moat: every productivity tool that requires invocation (Notion, Sunsama, even AI browsers like Atlas/Comet) loses to a surface that's already there. The new tab page is the most-seen screen in any browser user's life. Sharpened axis: calm-by-ambient-surfacing — value in the gap between intent and next action (was: just "calm"). Evidence: research-docs/toby-delta-2026-05-05-v3.md; competitive map shows "calm" is now a shared category claim (Sunsama, Neurosity, Cold Turkey) — ambient is what stays defensible.

3. Persistence is the product. The bet isn't lowering your tab count; it's making the act of closing safe. Save a window in one click, open tomorrow, it's all still there. Why this is a moat: bookmarks ask you to retrieve; Toby asks you to recognize. Visual recognition beats hidden hierarchy by orders of magnitude in working-memory tasks. Evidence: v1.13.0 activation-research finding "the activation moment is persistence — 'I opened a new tab and my organized world was still there', not 'AI organized my tabs'" (see: docs/ai-onboarding-ideas-analysis.md, commit b9bea18c context); paid-user quotes "peace of mind closing a tab", "horribly. it's why I paid the ridiculously high price for annual" (see: product/strategy/soul.md).

4. Instant clarity. Understand Toby in 30 seconds; feel calm in 5 minutes. If a feature needs explanation, it's too complex. Why this is a moat: power tools (Raindrop, Workona) require investment; Toby should feel like opening a well-organized closet — immediately obvious. Evidence: free-user quote "I love the simplicity and don't need a user guide" (see: product/strategy/soul.md).

Anchors — table stakes we must never break (5)#

1. Save and restore tabs reliably. Cost of breaking it: the entire product. Currently at risk — the May 2026 blank-page-on-new-tab regression is live in CWS reviews and on Vivaldi/Orion forums (see: research-docs/toby-delta-2026-05-05-v3.md). Fixing it is upstream of every other strategic move.
2. Cloud-backed account. Cost of breaking it: data follows users across devices is now table stakes — Chrome 133 ships saved-tab-group cross-device sync natively. We must hold this floor without making it our differentiator (it isn't anymore).
3. Chrome extension, ambient new-tab surface. Cost of breaking it: we live and die by the new-tab page. Mobile is a companion, not the primary surface. Cross-platform expansion (Firefox/Safari) is allowed only if it doesn't compromise the new-tab anchor.
4. Free tier viable. Cost of breaking it: ~88% of active users are free; they're the funnel and the social-proof reservoir. The free experience must be genuinely usable, not crippled. Note: free-tier collaboration limits are an open question — we may restrict those without breaking this anchor (see: bets.md role-based-paywall-gating).
5. Invisible performance cost. Cost of breaking it: CWS 2026 algorithm penalizes Core Web Vitals impact directly; "resource intensive" review complaints already showing in listing. Performance is a category requirement, not a feature.

Identity#

• Product type: Chrome / Firefox browser extension (new-tab page replacement) with a companion mobile app. Not an AI customer-support tool; the AI-support investigation under ai-support/ is a separate productization of an internal tool and is not Toby itself.
• Primary job (in the user's voice): "Give me somewhere safe to set down the tabs I'm not done thinking about, so I can close the browser without grief."
• Audience: Two anchor personas from toby/01-personas.md —
  • Solo Pro Power Saver (9.5% of active users, 6,656 users, 94% weekly stickiness) — the revenue anchor. 4.8-year average tenure, still building 2.5 new lists/month. THIS is product-market fit, in the segment-specific sense.
  • Tenured Free Organizer (62.4% of active users, 43,722 users, 86.6% weekly stickiness) — the mass. 3.5-year tenure, 22 lists each, 0.13 new lists/month. The library is "done"; they're not the revenue, but they're the durability and the social proof.
• Anti-customer: someone who is fine with 5 tabs and Chrome bookmarks; someone who wants a Notion-level link-management power tool; someone who will never pay for a browser extension.

Brand promise#

"It's okay, Toby has it now."

Every interaction is evaluable on: did it lower anxiety, or did it raise it? Saving a tab should feel like setting something down — not filing paperwork.

Anti-promise (what we don't promise)#

• We don't promise to lower your tab count. The count is a side-effect of the medium, not a goal.
• We don't promise AI auto-organize. (Q4 2026 internal target; not before. Do not pre-announce.)
• We don't promise enterprise-grade team/admin controls today — we're solo-Pro-shaped, with team as a future, not a feature parity claim against Notion/Workona's team tier.
• We don't promise to be the only way to manage tabs. If five tabs is your number, you don't need us.

Changelog#

• 2026-05-10 — Initial draft in toby/strategy/. Sharpened axis 2 ("calm" → "calm-by-ambient-surfacing") after v3 research showed "calm" is becoming a category claim. Added axiom 3 (Persistence-as-Activation) — codifies the v1.13.0 activation insight that was implicit in soul.md but not surfaced as a 10× claim. Promoted reliability to anchor 1; relegated cloud-account differentiation (Chrome 133 closed that wedge). Preserved soul.md voice and JTBD intact.