Today’s piece is a collab with the legends over at Figma, and their new product Figma Make. My first-hand look at the frontier of AI-powered design in 2026.

I have to say, I’m feeling more excited than ever about what it means to be building in this era.

It’s not just about the AI models getting better.

It’s the fundamentally new mental models they unlock.

My day-to-day is almost unrecognisable from even 6 months ago.

Vibe-coding/operating is now shockingly overpowered for small teams.

Despite being an AI DAU since Nov ‘22, I’m finding myself starting at my screen in utter disbelief multiple times per day.

Enjoy.

— Tom

Most teams are busy, not productive

Digital leaders everywhere are shipping more tools, yet outcomes keep stalling. Meetings multiply. Decisions slow. Execution suffers.

At Productivity Summit 2026, you’ll learn practical systems used by modern strategy and transformation leaders to simplify work, cut friction, and increase real output - across people, processes, and platforms.

Expect clear frameworks, real examples, and ideas you can apply immediately (#ad).

Don’t let your roadmap die in spreadsheets

Too many product ideas live in spreadsheets, insights get buried, and teams struggle to rally around a shared roadmap.

See how Jira Product Discovery helps you bring ideas and feedback into one place, prioritize with consistent scoring and clear tradeoffs, and connect discovery directly to Jira delivery so roadmaps stay aligned across teams as work progresses.

_{Thank you for supporting our sponsors, who keep this newsletter free.}

The what: A TLDR explanation of the strategy

On July 7th, 2025, I added another SaaS idea to the pile:

A Chrome extension. SimilarWeb for builders. Traffic, revenue signals, founder socials, ads - all in a sidebar you can trigger when you visit any company's website.

A 9-bullet sketch of something I'd personally use a dozen times a day, if it existed.

Nothing came of it. Got buried under many more (half-baked) ideas.

Fast-forward 6 months.

The Figma team reached out about a partnership.

They wanted me to try their new prompt-to-app AI tool, Figma Make.

I mulled over the angle for a while, then realised I had the perfect test case.

Could I take a 6-month-old idea that had never left the notes app and turn it into a fully interactive, clickable product prototype? Without writing a single line of code? Without opening a design tool?

The pipeline I came up with:

→ Notion (raw idea, 9 bullet points) → Claude Code + Wispr Flow (interview to flesh it into a full PRD) → Figma Make (PRD in, interactive prototype out)

One sitting later, I had my answer.

Here’s how it went down:

The how: The strategic playbook boiled down to 3x key takeaways

The traditional product design lifecycle goes something like: idea → sketch → wireframe → mockup → prototype → iterate → build.

When I was at Atlassian, one of the toughest aspects of bringing product ideas to life was this:

The longer the stretch of time between the original idea and something clickable, the greater the risk from untested assumptions.

→ Figma Make collapses this. A well-structured PRD goes in. A testable product journey comes out.

Not production code - nobody's shipping this to App Store review. But a credible, high-fidelity prototype that's ready for: usability testing, stakeholder buy-in, investor conversations, and the ability to click through your own idea to see if it sticks.

Here's how I built mine:

I opened Claude Code, pasted my Notion link, and asked it to interview me about the idea.

What followed was a 5-round interview. Claude pushed me on positioning ("How is this different from SimilarWeb?"), forced me to articulate the ICP ("Who specifically, and what's their burning pain?"), and pressure-tested the engagement model ("Why would someone use this more than once?").

I dictated 25 minutes of product spec. Other cafe patrons looked at me funny. Someone got visibly worried around 15 minutes in. I persevered.

The output: a comprehensive PRD covering everything from the core product experience and panel-by-panel UI specs to the success metrics and launch roadmap, etc.

Link to the full PRD here

At first, I was curious how Figma Make would go with a simple starter prompt. The nucleus of the idea, without the full PRD.

Pretty nice for a one-shot, ngl.

Time for the real test.

I fed the whole PRD into Figma Make. Positioning, personas, interactions, design principles. Even the pricing model and GTM strategy.

Everything.

A few minutes later, I was clicking through a passable MVP for a product I'd been imagining for six months.

With tools like these, the nuance matters. The quality of the input directly determines the quality of the output. The vague prompt produced something vague. A structured spec (clear user stories, information hierarchy, and interaction details) produced something that genuinely felt like my product.

That's the real unlock for anyone building things right now. Designers who value craft and speed get a turbocharged starting point that respects fidelity. PMs who've been blocked waiting for design resources can generate credible prototypes from their own PRDs and get alignment without the bottleneck. Founders and builders with ‘product sense’ but no design background can go from concept to clickable demo in a single session.

The bottleneck has shifted. It's no longer "can I build this?" It's "have I thought about this clearly enough?"

As much as I loved going 0 to 1 with a totally blank canvas, very little design work happens in a vacuum.

In practice, the modern designer always starts with context, constraints and reference points. Brand guidelines, user research, existing components, customer feedback, company strategy, etc.

Figma Make is built around this same instinct. Rather than generating in isolation like me, it lets product teams bring real context directly into the creative environment in 2 powerful ways:

1. Your other Figma assets. If you've already built a design language in Figma Design (colours, typography, components, spacing, logos) you can export that library directly into Figma Make and reference it as style context.

In practice: Every prototype Figma Make generates in that file then inherits your brand's design DNA - not as a loose suggestion, but as a concrete stylesheet governing how things look and feel.

For teams with mature design systems, this is significant. The AI isn't guessing at your brand. It's building with your brand.

You can go further by editing the extracted styles directly and adding a guidelines.md file - a free-form text document where you give Figma Make specific rules for how to apply your system. Creative direction that persists across every prompt in that file.

2. Live context through MCP integrations. This is where things get interesting for integrated product teams.

Figma Make supports MCP (Model Context Protocol) connections, meaning you can plug it directly into the tools your team already uses - Notion, Jira, Linear, Intercom etc.

That means a designer or PM working inside Figma Make can pull in live context without leaving the tool:

→ “Fetch the latest user feedback sitting in Intercom to inform the content in this panel”

→ “Pull requirements directly from our Jira board to design the navbar”

→ “Search Notion for research notes, prior explorations, and design principles my team has already documented”

I worked solo on this build, feeding in my PRD manually. But for teams, both Figma design libraries and MCP connections represent a genuine workflow shift. They collapse the number of tabs, the number of handoffs, and the number of "hey, can you send me that doc?" messages that slow down the iteration loop.

The designer doesn't need to chase down the latest context.

It flows directly into the creative environment.

For millions of product teams, Figma is the shared language.

Design happens in Figma Design.

Collaborative thinking happens in FigJam.

Presentations happen in Figma Slides.

Figma Make slots into that ecosystem natively - everything it generates is a first-class Figma object, with every layer, component, padding value, border radius, and colour token preserved and editable.

A prototype built in Make speaks the same language as each of these other tools.

→ Figma Design: Publish your Make prototype directly into Design to refine individual components, polish interactions, and take it from exploration to production-ready mockup.

→ FigJam: Drop the prototype into FigJam's interactive whiteboard where the team can annotate, reorganise, and build on the concepts collaboratively.

→ Figma Slides: Port the prototype into a presentation to align stakeholders, investors, or your team - no recreating screens as static images.

→ Figma Sites: Push your prototype to a live website using Figma Sites, where code layers let you add custom interactivity, connect real data sources, and ship directly to the web.

This matters because prototypes aren't endpoints.

They're conversation starters.

They're alignment tools.

They fit into a much longer process.

Before the prototype: research, strategy, requirements, design systems.

After the prototype: feedback, iteration, handoff, testing.

Figma Make is Figma’s strategy to ensure that artefact never needs to leave home. Generate it, refine it, present it, debate it, publish it, test it with real users - all within the ecosystem your team already lives in.

The where: 3x high-signal resources to learn more

If you've got a product idea sitting in your own "maybe one day" pile, this is your excuse to spend an evening on it.

Start free.

Paste your idea.

See what comes out.

Here's the complete Product Requirements Document I built with Claude Code - open-sourced for anyone who wants to PRD-to-app this weekend.

Covers everything from the ICP and engagement loop design to panel-by-panel UI specs, pricing model, and go-to-market strategy.

Steal the structure. Adapt it for your own idea.

Here's the final clickable concept.

I'm genuinely exploring whether to build QuickScout for real. Check out the prototype, and if you think you’d use it - click here to help me build the right thing!

Our flagship course on how to use free internet data to make better strategic decisions. Contains 5 years of strategy expertise, proven methods, and actionable tactics to accelerate your career with modern-day strategy skills.

We have a growing audience of 110,000+ strategists from top companies like Google, Meta, Atlassian, Stripe, and Netflix. Apply to feature your business in front of Strategy Breakdowns readers.

One of the most common questions we get asked is: “What tools do you use to run Strategy Breakdowns?” So, we’ve open-sourced our tech stack to give you an inside-look at exactly what tools we’re using to power each corner of this operation.