The 7 design tools I actually use in 2026 (and the ones I dropped)

Once a year I do a tool audit. Open every app on my dock, every browser bookmark, every recurring subscription. Ask myself "would I pay for this again today?" Most of the time the answer is no, and I cancel.

After ten years of running this audit, I'm down to seven tools that survive year over year. Everything else is rotation — try it for a quarter, drop it if it doesn't earn its place.

Here's the current 2026 stack, why each one stays, and a few notable drops.

The seven that stay

1. Figma

Inevitable. The reason it stays isn't because it's the best at any specific task — Sketch was better at drawing for years, FigJam was nominally better at whiteboarding. It stays because it's the lingua franca. Every founder, engineer, and designer I work with already has it. Switching cost would only be justified by a 3x productivity gain elsewhere, and nothing else delivers that.

I use Figma roughly 70% of my desktop hours. It's the surface where decisions get made, not just where files live.

2. Notion

For everything that's not in Figma. Project briefs, client notes, sprint plans, the personal database of decisions I'd make differently. It's a worse Word, a worse Roam, a worse Airtable — but it's tolerable at all of them, which is rare.

The killer feature isn't the database or the AI or the templates. It's that I can paste a Figma frame, a Loom video, a screenshot, and a YouTube embed into the same doc and have them all render. Nothing else handles that mix as well.

3. Linear

The only tool I use that I actively enjoy. Tasks, sprints, roadmaps. The interface respects you. The keyboard shortcuts are sensible. It loads instantly.

Most importantly: clients I introduce it to convert within a week. That signals it's not just my preference — it's actually well-designed. Jira's not even in the conversation.

4. Loom

Video walkthroughs of design files. The best 30 seconds of asynchronous communication available. I record one for almost every design handoff: a 90-second voiceover walking through the rationale of a flow, with my cursor highlighting the key decisions.

This single habit has done more for client trust than any deck I've ever written.

5. Raycast

Spotlight, but for power users. I trigger Figma files, search Linear tickets, run Calendar checks, switch wifi networks, and do unit conversions all from one keyboard shortcut.

The investment to learn the keyboard shortcuts is real. So is the payoff. Once I had it muscle-memorized, my time-to-context-switch dropped by maybe 30 seconds per switch. That compounds in a day of pairing across clients.

6. Things 3

Personal task manager. Not Linear's job. Things keeps everything that isn't pinned to a project — gym schedule, personal admin, articles to read, ideas in progress.

The reason it stays is that the friction of capture is near zero. I open Things, type, close. No project assignment, no tag taxonomy, no priority field. Capture first, organize later — most tasks die before they need organizing, which is a feature, not a bug.

7. iA Writer

Long-form writing, including this post. Markdown. No formatting. No distractions. Files saved as .md to iCloud, openable in any other tool.

I wrote a draft of this article in iA Writer, in a blank window, in 45 minutes. The same draft in Notion would have taken 90 minutes because Notion's interface invites futzing.

Notable drops

Sketch — dropped 2018

Was better than Figma at vector work for years. Lost on collaboration, never recovered. Adobe XD was a worse pivot.

Webflow — dropped 2024

Used it for marketing pages. Powerful, but the ceiling is lower than the floor of code-built sites. When my Cursor + Tailwind + Next.js workflow got reasonable, Webflow's value proposition collapsed.

Slack — replaced with Discord/iMessage in 2025

Most of my client work moved to Discord (community-style) or iMessage (1:1 founder convos). Slack still runs for two clients. I don't miss it. It's a tool that scaled itself out of useful.

Roam Research — dropped 2022

Every two years I try a "second brain" tool. Roam was the most ambitious. The bidirectional links are seductive in theory, useless in practice for how I actually retrieve information. I retrieve by fuzzy search 99% of the time. Roam optimized for the 1%.

Twitter / X — dropped 2024

Too noisy. Was a great network for design 2014-2018. Now it's a feed I can't read sober. LinkedIn took its place for design discourse, which says something I'd rather not interrogate.

Airtable — dropped 2023

Fine for ops, never fit my workflow. Notion's database covers 90% of my use case at lower friction.

Pitch — dropped 2024

Was meant to be the modern Keynote. The team was great, the design was great, the product was incomplete. I went back to Keynote for client decks. Reliability beats novelty.

What's not on this list (and why)

People ask me why I don't use:

• Cursor / AI editors — I do, but I don't classify them as a design tool. They're an engineering tool I use because I write code.
• Framer — tried it three times. The Figma-to-Framer friction never went down. Use Framer if you don't already use Figma; otherwise skip.
• Designer-specific AI tools (Galileo, etc.) — they output things that aren't designs. They're vibe machines. Some are useful for ideation; none have replaced any tool on my main list.
• Whiteboarding tools (Miro, Mural) — I use FigJam inside Figma when I need a whiteboard. Avoiding tool sprawl is worth more than feature parity.

The principle

The list shrinks every year because I optimize for one thing: how fast can I get from "I have a thought" to "the thought is in the right tool, in the right context, where I can act on it next time."

Every additional tool adds switching cost. Every tool that requires its own taxonomy adds organizational tax. The seven above survive because the per-tool tax is low, the value is high, and I'd be slower without them.

If you're auditing your own stack, the question to ask isn't "is this tool useful?" — almost any tool is. It's "would I notice if it disappeared this week?" If the answer is no, cancel the subscription.

If you're trying to figure out the broader process around how I work with founders, the tools are mostly downstream of the process. The other way around almost never works.