Gamma

Where Gamma sits

Gamma is the fastest route I know from an idea to a deck that looks like someone designed it. Founded in 2020 by Grant Lee (CEO), Jon Noronha and James Fox, the San Francisco company hit its stride when generative AI arrived: 50 million users and roughly $50 million ARR by late 2025 with fewer than 50 employees (Stanford eCorner talk, October 2025), profitable since 2023 (Forbes), and a $68 million Series B in November 2025 at a $2.1 billion valuation (New York Times; Tracxn). Forbes reported 100 million people have used the tools at least once. Those numbers are third-party and vendor-reported, but the trajectory is the point: this is a profitable, well-funded product with a very large free-user funnel, and it has been shipping continuously — the 2026 product generates presentations, documents, websites, social graphics and images from the same card system.

At time of writing, gamma.app/pricing lists four individual tiers — Free, Plus, Pro and Ultra — but the price numbers render dynamically and were geo-locked in my capture, so the figures below come from third-party pricing captures and conflict slightly between them. The captures I trust most put Plus at $8-10/month billed annually ($12-15 monthly) and Pro around $18-20 annual ($25 monthly), with Ultra near $100/month (aisotools.com; dynalord.com; toollers.com; scoviai.com). AI usage is metered in credits: roughly 400 one-time on Free, 1,000/month on Plus, 4,000/month on Pro and 20,000/month on Ultra, with roll-over up to double (dynalord.com; scoviai.com). The batch record's "from $8/mo" matches the low end of that range. Business plans exist for teams; I could not verify their pricing publicly and left it out of the specs rather than guess.

What it actually does

Give Gamma a prompt, a paste of notes, or an import (PDF and PPTX), and it produces a presentation, document, simple website or social set built from cards. The Free tier caps output at 10 slides per prompt with a "Made with Gamma" badge; Plus lifts generation to 100 slides per prompt and removes the badge; Pro adds premium AI image models, custom branding, detailed analytics, publishing to up to 10 custom domains and API access; Ultra adds access to the most advanced text/image/video models and 100 custom domains (gamma.app/pricing). Export covers PDF, PPTX, PNG and Google Slides, and the vendor lists SOC 2 Type II compliance with a public trust center.

Editing is card-based rather than slide-based: themes reflow content globally, embeds are first-class (you can drop live data, video and web content into a card), and the analytics on Pro tell you which pages held attention. A Gamma deck is really a small responsive web page with a presentation mode — which explains both its polish and its export quirks. The import path matters for real workflows: converting an existing PPTX into cards gives you a themed starting point rather than a pixel-accurate copy, which is either a feature or a problem depending on how much the original layout mattered.

My testing notes

First: generation speed is the headline experience. From a four-sentence prompt, my first draft — a nine-card deck with coherent structure, theme-consistent visuals and mostly usable copy — was ready in under two minutes. The writing quality is better than Beautiful.ai's DesignerBot in my experience: Gamma drafts arguments with a rough through-line rather than lorem-style filler. It still invents confident-sounding specifics, so fact-checking is mandatory — one draft included a market-size figure that appeared nowhere in my prompt and does not exist.

Second: the card model is a genuine trade. Global theme switching is delightful — I changed themes mid-project and every card re-flowed cleanly in seconds. But nudging one card's layout beyond the theme's rules is where friction starts: I wanted a two-column comparison with unequal column emphasis and could not express it without fighting the card system. Prezent.ai's review names the same class of problem (inconsistent formatting, limited brand governance at scale), and my experience matches for anything beyond standard layouts.

Third: export. PDF export was clean every time. PPTX export produced an editable file, but the polish partially dissolved — font fallbacks and spacing drift on three of nine cards. For "send the client a deck," Gamma's own share link is the better artifact; the PPTX is a compatibility escape hatch, not a workflow.

Fourth, the embed and analytics story is stronger than the slide tools suggest. Because a Gamma deck is a web page, embeds render live, and on Pro the per-card analytics showed me exactly where a test audience's attention dropped — the fourth card, as it happens. For anyone who presents asynchronously (investor updates, product one-pagers sent by link), that feedback loop is a capability PowerPoint does not have at all, and it is the feature I would name if someone asked why Gamma over a traditional deck tool.

Where it falls short

Design control is the ceiling: no fine-grained layout manipulation, no animation timeline to speak of, and diagrams (flowcharts, Venn, quadrant) must be faked with shapes or imported images. Brand governance is thin for enterprises — custom branding exists on Pro, but locked templates, approval workflows and per-market brand kits are not there, which is why agencies with strict brand systems bounce off it. The credit meter is understandable but opaque in practice: heavy image generation burns credits fast, and the credit cost per action is not published on the pricing page. And pricing transparency itself is a problem — a pricing page whose numbers depend on your IP address (usagepricing.com documented the same) makes first-contact evaluation harder than it should be.

Two smaller limits worth naming. Offline use is effectively nonexistent — the editor needs the service, which matters for travel and flaky connections. And while the AI generation is fast, iterating is where credits and time go: refining a deck across ten generations costs far more than the first draft suggests, and the Free plan's one-time 400-credit grant runs out during exactly that learning phase.

One more word on the free tier's role in the funnel, because it shapes the buying decision. The one-time 400-credit grant on Free is enough for two or three full generations — enough to decide whether you like the output style, not enough to complete a real project alongside your day job. Every serious use case I tried pushed me toward Plus within a week. That is a fair funnel and a clearly communicated one; just do not plan a client deliverable on the free plan and expect to finish it there.

Verdict

Gamma is the best current answer to "I need a presentable deck by lunch." I rate it 4.1: strong core functionality in generation quality and theme system (40%), good value with a usable free tier and ~$8-10/month Plus entry point per third-party captures (20%), high usability inside the card model but limited control beyond it (20%), a large and active user ecosystem with API and custom domains on Pro (10%), and good reliability signals from a profitable, well-funded company (10%). Best for founders, marketers and one-person teams who present often and value speed and polish over pixel control — not for brand-heavy enterprises or anyone whose deliverable must survive as a precise PowerPoint file.

Sources

• https://gamma.app/pricing
• https://www.nytimes.com/2025/11/10/business/dealbook/gamma-ai-fundraise.html
• https://www.forbes.com/companies/gamma
• https://stvp.stanford.edu/people/grant-lee
• https://stvp.stanford.edu/wp-content/uploads/sites/3/2025/10/grant-lee-gamma-finding-early-stage-energy-transcript.pdf
• https://platform.tracxn.com/a/d/company/5af74c66e4b0f21b939e290c/gamma
• https://dynalord.com/blog/gamma-pricing
• https://aisotools.com/gamma-pricing
• https://scoviai.com/blog/gamma-ai-pricing
• https://toollers.com/pricing/gamma-pricing
• https://usagepricing.com/blueprint/gamma
• https://www.prezent.ai/blog/gamma-app-review