Figma: The $65 Billion Haircut That Created the Best Risk/Reward in SaaS

The Conviction

Figma went public on July 31, 2025 at $33 per share, briefly rocketed past $142, and then lost 87% of its value in eight months (Yahoo Finance, April 16, 2026). That is roughly $65 billion in market cap erased. At $19.25 today, the stock sits 42% below its IPO price. The financial media has moved on. The narrative is settled: AI will eat design tools, and Figma is the main course.

The numbers tell a completely different story. In Q4 2025, Figma posted $303.8 million in revenue, beating consensus estimates of $293 million (Figma Q4 FY2025 earnings release, February 18, 2026). Full-year revenue hit $1.056 billion, up 41% year over year, crossing the billion-dollar mark for the first time. Net dollar retention accelerated to 136%. Adjusted free cash flow margin came in at 23% for the year. The company is sitting on $1.7 billion in cash and marketable securities with zero debt.

I am bullish on Figma at $19.25. The stock trades at 6.1x next-twelve-months revenue while growing 30%+. That multiple prices in permanent AI disruption, not the temporary margin compression from AI investment that management is actually guiding for. Figma controls 80-90% of the professional UI/UX design market, and every piece of quantitative evidence from Q4 2025 says the installed base is expanding, not contracting. The market has confused a real competitive threat with a near-term certainty.

The Business

Figma is a browser-based collaborative design platform. Designers use it to create user interfaces. Developers use Dev Mode to translate those designs into code. Product managers use FigJam for whiteboarding and Figma Slides for presentations. The platform expanded further in 2025 with Figma Make (AI-powered prompt-to-prototype), Figma Sites (design-to-web publishing), Figma Buzz (marketing asset creation), and Figma Draw (vector editing). It is no longer just a design tool. It is trying to become the product development operating system.

The business model is pure SaaS: seat-based subscriptions with per-seat pricing that escalates from free (2 projects) to $75/editor/month on the Enterprise plan (Figma pricing page, April 2026). What matters most for Figma is net dollar retention, which measures whether existing customers are spending more or less each year. At 136% in Q4 2025, every $1 of ARR from a year ago is now worth $1.36. That is best-in-class for a SaaS company at this scale, better than CrowdStrike (115%), better than Datadog (116%), and better than nearly every public software company outside of a few AI infrastructure names.

The user base is massive and deeply embedded. Figma claims that nearly every major technology company, startup, and digital agency uses the platform. That installed base is the real asset, and the question is whether AI tools chip away at it or whether Figma layers AI on top to make it stickier.

The Case

Three drivers underpin the bull thesis: AI credit monetization creating a new revenue stream on top of existing seats, platform expansion driving multi-product adoption across organizations, and the irreplaceable nature of collaborative workflow in professional design.

AI Credits Are a Revenue Unlock, Not Just a Cost Center

The market's biggest fear is that AI kills Figma. Here is the irony: AI might be Figma's best growth driver over the next two years. In March 2026, Figma began enforcing monthly AI credit limits across all paid plans, with customers able to purchase additional credits via subscription or pay-as-you-go models (Figma blog, March 2026). This is the company's first foray into consumption-based pricing layered on top of seat-based subscriptions.

The adoption data going into monetization was strong. About 75% of customers above $10,000 in ARR were consuming AI credits weekly before enforcement began (Figma Q4 FY2025 earnings call, February 18, 2026). More than half of customers spending $100,000+ per year had team members using Figma Make every week during Q4. These are not casual experimenters. These are power users who have integrated AI generation into their design workflows and will pay to keep access.

“Our growth and momentum show that our strategy is working. As AI gets better, Figma gets better, and we're shipping faster than ever.”

Dylan Field, CEO and Co-Founder — Q4 FY2025 earnings call, February 18, 2026

The bull case math is straightforward. If Figma can extract an incremental $5-10 per seat per month from AI credits across its paid user base, that is a meaningful revenue expansion on top of 30% organic growth. Management has not quantified the expected contribution, which is why it is not yet in the FY2026 guide. But the usage data suggests the demand exists. The revenue will follow as enforcement bites and AI surfaces multiply. For context, even a modest $3/seat/month blended across the paid base would represent hundreds of millions in incremental annual revenue at scale.

The cost side is real. Non-GAAP gross margin fell from 92% in FY2024 to 82.4% in FY2025 as AI inference costs hit the P&L (Figma FY2025 earnings release). That is nearly a 10-point compression in a single year, and it spooked investors who expected SaaS-typical 80%+ margins to be permanent. But the margin degradation happened before monetization started. As credit purchases ramp through 2026, the revenue from AI should begin catching up with the cost of serving it. This is the classic cloud infrastructure investment cycle: spend ahead of revenue, then harvest. Inference costs are also falling industry-wide as model efficiency improves, which means the gross margin headwind from AI should moderate even without price increases.

Platform Expansion Is Multiplying the Addressable User Base

Figma started as a tool for designers. It is now sold to designers, developers, product managers, and marketers. That is a dramatically larger addressable market within each customer account. FigJam and Figma Slides are included with all paid seats, creating entry points for non-designer roles. Dev Mode launched in 2024 as a paid add-on for developers who need to inspect and extract design specifications. Figma Sites turns design files into published web pages.

The net dollar retention acceleration from 129% in Q2 2025 to 136% in Q4 2025 tells you this expansion is working (Figma Q2 and Q4 FY2025 earnings releases). Existing customers are not just renewing. They are adding seats across new job functions and adopting new products within the Figma suite. That 7-point NDR acceleration in two quarters is one of the most bullish signals a SaaS company can produce. For comparison, most SaaS companies at Figma's scale see NDR slowly decline as they saturate their installed base. Figma's is reaccelerating. The multi-product strategy is clearly driving incremental spend per customer.

“I think the case is that humans will continue to use software, and increasingly agents will, too. I think this creates more surface for designers to work with and designers to think through.”

Dylan Field, CEO and Co-Founder — Interview with Calcalist, March 2026

The platform strategy also creates switching costs that matter more than the design tool itself. When an organization runs design, whiteboarding, presentations, developer handoff, and AI prototyping on Figma, the cost of migrating to a competitor is not just learning a new tool. It is rebuilding an entire cross-functional workflow. That stickiness shows up in the 97% gross retention rate, meaning only 3% of revenue churns annually (Figma Q4 FY2025 earnings release).

The Collaboration Moat Is What AI Cannot Replicate

Google Stitch's March 18, 2026 "Vibe Design" update lets users generate polished UI designs from text or voice prompts. The stock fell 8.8% in a single day on the news (CoinCentral, March 2026). Anthropic announced a competing design tool in April 2026. These are real threats, and dismissing them would be dishonest.

But generating a UI mockup from a prompt is not the same as building a product. Professional design is a collaborative process involving iteration, feedback, version control, design systems with thousands of reusable components, developer handoff specifications, and organizational workflows that connect design decisions to shipped code. A solo designer generating a one-off screen from a prompt does not need Figma. A team of 200 people building a complex product does.

This is the distinction the market is missing. AI design generation is a feature. Collaborative product development is a platform. Figma's real competition is not "can Google generate a button faster?" It is "can Google replicate the collaboration infrastructure where 80-90% of professional designers already work, including their design systems, component libraries, team permissions, commenting threads, and developer handoff workflows?" That is a much harder problem.

Dylan Field has framed this directly. Figma's philosophy is that "as the models get better, we get better" (CNBC interview, February 17, 2026). Figma Make already lets users generate designs from prompts, then refine them collaboratively in the same environment where the rest of the team works. The AI generation layer is an input into the collaborative workflow, not a replacement for it. If that framing proves correct, Figma captures the value of AI design tools rather than being displaced by them.

Consider the analogy to code editors. GitHub Copilot can generate code from prompts. It did not replace VS Code or the entire development toolchain. It made developers more productive inside their existing tools. Figma Make could follow the same pattern: an AI layer that speeds up the starting point while making the collaborative finishing process more valuable, not less. The 97% gross retention rate suggests enterprise customers are not yet defecting to prompt-only alternatives, even eight months after AI design tools entered the public consciousness.

The Risks

The AI risk deserves honest assessment. Google has effectively unlimited resources, Stitch is free, and the "vibe design" capability will improve rapidly. If the market for UI design tools bifurcates into "quick generation for solo makers" (won by AI-native tools) and "collaborative product development for teams" (won by Figma), the company keeps its core business. If the line between those categories blurs, Figma's moat narrows. I lean toward the former, but this is a genuinely uncertain call, and position sizing should reflect that.

The Bottom Line

Figma at $19.25 is a bet that collaborative workflow is harder to disrupt than individual design tool features. The stock is priced as though Google Stitch and Anthropic have already won. The Q4 numbers say they have not even started to dent the installed base. Revenue grew 41%. Net retention accelerated to 136%. Free cash flow margin hit 23%. Those are not the metrics of a company facing imminent disruption. Those are the metrics of a company investing aggressively in the AI race while its core business compounds.

The base-case DCF, using a 12.7% WACC and 8x exit multiple, produces an implied price of $22-28, representing 14-45% upside from today (no priors DCF model, April 2026). The bull case, requiring sustained 25%+ growth and successful AI monetization, gets you to $55-65, or roughly 3x the current price. The bear case, where AI tools erode growth to low-teens and margins compress, puts fair value at $10-14. At the current price, the asymmetry tilts in favor of the bull.

Analyst consensus sits at roughly $42, with a range of $30 to $63 and a split between "Buy" and "Hold" ratings (MarketBeat, April 2026). What to watch: Q1 2026 earnings in May will be the first quarter with partial AI credit monetization revenue. Look for commentary on credit attach rates and whether gross margin stabilizes. If 75% of high-value customers were using AI credits weekly before enforcement and a meaningful fraction converts to paid, the monetization narrative shifts from "cost center" to "growth driver." That is the catalyst this stock needs.