BetterTouchTool

The secret to loving BetterTouchTool is restraint. The app can bind everything, but the best configs are small, documented, and exported. Treat presets like code: version them, comment them in notes, and avoid copying giant community packs you do not understand. In 2026, Raycast owns much of the launcher mindshare and AI model routing. BTT's launcher/AI chat improvements show folivora knows that battlefield, but BTT's durable advantage remains hardware-proximate control, gestures, buttons, decks, not cloud model catalogs. Use Raycast for knowledge/commands; use BTT for body-level Mac control. Licensing that still sells 2-year update windows instead of rent-forever is a gift. Budget a calendar reminder before your update window ends if you rely on bleeding-edge macOS betas. Accessibility permissions are non-negotiable. Corporate Macs with MDM may block BTT; verify before you standardize a whole team on it. Performance-wise, BTT is native and light relative to Electron launchers. The risk is not CPU, it is human complexity. Schedule a quarterly config cleanup: delete unused triggers, rename cryptic actions, and re-export. Future you is a different coworker. On multi-user Macs, keep BTT configs per login account. Shared machines with divergent gesture expectations create support nightmares. For teams, document a minimal standard preset rather than allowing every engineer to invent a personal dialect of trackpad. ### Migration patterns Coming from Magnet or Rectangle: recreate left/right/quarter snaps first, bind them to the same shortcuts, and only then delete the old app. Coming from Karabiner-only setups: keep Karabiner for key rewrites, add BTT for gestures and window actions. Coming from Hammerspoon: you can replace many scripts with BTT actions, but keep Hammerspoon if you have deep Lua libraries already. ### Stability practices When a macOS update lands, temporarily disable brand-new experimental triggers. Update BTT promptly from official folivora channels. If gestures double-fire, check for duplicate triggers across global and app-specific sections, the UI makes it easy to accumulate copies. For trackpads, prefer slightly higher threshold settings to reduce accidental activations while typing. ### AI chat / launcher role Use BTT's launcher features for personal snippets and machine-local actions. Use Raycast or system tools when you need cloud model catalogs (GPT-5.6 family, Claude Opus 5 / Sonnet 5, Gemini 3.x, etc.). The tools overlap at the "type to run" layer but diverge on ecosystem. Running both is normal; fighting about which is "the one true launcher" is not productive. ### Extended workflow cookbook Window management: map Hyper+H/J/K/L to left/bottom/top/right halves, Hyper+U/I/O/P to quarters, and Hyper+N to next display. Trackpad: three-finger swipe up for Mission Control remains system-owned, do not fight it; use four-finger or tip-tap for BTT-only actions. Mouse: bind side button to "middle click" in browsers and to "razor tool" in your NLE via per-app overrides. Clipboard: create a floating clipboard history HUD if you do not already use a dedicated manager. Text transformations: uppercase, slugify, and JSON-pretty actions save surprising amounts of time for developers pasting between terminals and docs. Notch Bar / floating menus: put the ten actions you actually use daily there instead of memorizing fifty shortcuts. Scripting bridge examples: a shell action that toggles dark mode, a JavaScript action that resizes the front window to a golden-ratio frame, and an AppleScript that talks to Music for focus playlists. Conditional activation: only enable "force quit front app" when the frontmost app is not Finder. Named triggers: expose a few BTT triggers to Shortcuts so Siri/automation can call into your gesture engine. ### Troubleshooting checklist If BTT stops seeing gestures after a macOS update, re-check Accessibility and Input Monitoring toggles, OS upgrades sometimes reset them. If CPU spikes, look for looping scripts or overly sensitive trackpad continuous gestures. If two actions fire, search the config for duplicate shortcuts. If licenses fail to validate, confirm system clock and network access to the license provider. Export a backup before major BTT version jumps within your two-year update window. ### Who should not buy People who want a pretty curated launcher storefront should start with Raycast. People who only need Caps Lock remapping should use Karabiner. People who only need snap zones should use Rectangle free. BTT is for the remainder, the users who feel constrained by those simpler tools and are willing to configure deeply. Across real Mac workflows the difference between a tool you open weekly and a tool you live inside is configuration depth plus reliability after OS upgrades. Tahoe 26.x users should treat each point release as a chance to re-validate permissions, favorite presets, and any automation that hooks Accessibility APIs. Apple Silicon performance is rarely the bottleneck; human-created complexity is. Documentation literacy beats forum folklore. Official release notes, pricing pages, and requirement matrices change faster than blog posts. Before you standardize a team on any utility, assign one person to re-check vendor pages quarterly and paste findings into an internal runbook. That habit prevents surprise renewals and unsupported OS combinations. Security hygiene still applies to productivity apps. Download only from official domains, verify code signatures when macOS warns you, revoke permissions for abandoned tools, and keep FileVault on. Menu bar managers, display utilities, automation tools, and password managers all request powerful rights; treat those rights as credentials. Competitive landscapes shift. Free open-source alternatives keep paid apps honest, while subscription tiers fund continuous macOS compatibility. The rational buyer trials two options for a week with real documents, real monitors, and real gestures, not synthetic checklists. Keep the tool that reduces friction you can measure. For bundl readers assembling a kit: prefer a small set of deep tools over twenty shallow ones. Pair a password manager, a window/input automator, a display helper if needed, a notes app, a text editor, a SQL client, and a creative suite. Remove anything you have not launched in thirty days. Disk space is cheap; attention is not. Accessibility matters. Keyboard users should confirm full feature reach without a mouse where possible. Color and type size choices in editors and notes apps affect multi-hour comfort more than benchmark screenshots. If a tool fights Dynamic Type or reduced-motion preferences, file feedback or choose another app. Backup discipline: export configs, vault emergency kits, Blender projects, and SQL connection lists on a schedule. Cloud sync is not a backup. Versioned zip files in a different account have saved more freelancers than any premium support plan. Finally, pay indie developers when a tool becomes infrastructure. One-time licenses and cheap yearly plans are how excellent Mac software survives. If Free Mode or Community Edition covers you forever, contribute code, docs, or donations so the next OS release still has a maintainer awake. In real use the apps that stick are the ones with configuration depth that holds up after an OS upgrade, not the ones you launch once a week. If you are on Tahoe 26.x, use each point update as a reason to recheck permissions, presets, and any Accessibility automation you depend on. Raw Apple Silicon speed is seldom what slows you down; tangled presets and duplicate triggers are. Learning to read release notes properly matters more than forum threads. Vendor pages for pricing and requirements change quickly, faster than most blog roundups. If your team will standardize on a utility, pick one person to verify those pages each quarter and log notes internally. That routine avoids renewal surprises and mismatched OS requirements. Productivity tools still need hygiene. Grab downloads from official domains, heed Gatekeeper warnings, revoke access for tools you abandoned, and keep FileVault enabled. Inputs, menu bars, and password managers ask for broad access; handle that access like a credential. The market moves. Open-source options keep commercial pricing in check, and paid tiers often fund the steady macOS work behind the scenes. Try two candidates for a week with actual files, displays, and gestures and keep whichever measurably lowers your friction.