52 people use AI to make PPT on a platform that serves 70 million customers and makes $700 million a year

52 people, an AI PPT tool, annual revenue of hundreds of millions of dollars. In Silicon Valley AI companies generally “burn money for growth” in 2025, but they rely on PPT to do a miracle valuation of 2.1 billion dollars. And the name of this company is--Gamma.

With 70 million users, Gamma is a newcomer to the industry. It has not become the next PowerPoint, but is making PowerPoint become the “previous self”.

as a result, whenKapasi (city in Malaysia)Posting to ask for recommendations for the PPT version of Cursor, the following clear are the netizens of the spontaneous brush screen: Gamma!

Come invest more, but earn more.

Gamma has just officially announced that it has closed its Series B round of funding$68 million, led by A16Z, escalating the company's total valuation to $2.1 billion.

The reason given by the investors, too, is straightforward:

In the age of AI, there's no reason to continue to use vintage 1987 software, and we're leading the charge to invest in, and support, anti-PowerPoint.

There's no doubt that Gamma's recent headwinds are attracting many investors to look at it sideways.

Take the last $12 million financing as an example, Gamma hadn't even spent a penny of the money and new investment was already coming to them.

But Gamma's founders are not impressed:

We now have more money in the bank than all the money we raised before combined.

To put it humanly, Gamma is super profitable.

Beginning in 2023, Gamma isIt's been profitable.Cumulatively, $100 million in annual recurring revenue has been realized(ARR).

Keep in mind that the team only has about 50 employees in total right now, which equates to two million dollars in ARR generated by each employee alone.

So the hot VC circle is just a sprinkle of water for Gamma. Relying on VCs is not as good as relying on oneself, and Gamma has been upholding this point since she started her career.

The early days of a business that was once on the brink of extinction

Gamma was first established in 2020, with three foundersGrant Lee,Jon Noronha,James FoxThey were all working together at Optimizely.(bushi)Being an executive.

After his wife took over the only desk in the house during the home office period, Grant Lee had to run to a nearby park bench and struggle to stare at the PowerPoint on his cell phone screen for meetings.

As it turns out, never in a million years, talking to good buddy Jon Noronha reveals that he feels the same way:It's painful to make Google Slides, and it's painful to look at Google Slides made by others.

Thus, a group of “pet lovers”, “woodworkers” and “board game players” with different personalities got together because of similar interests and started their business in San Francisco. They started their business in San Francisco.

Their original idea was to make a prettier alternative to PowerPoint, the online typesetting tool.

It's a huge blue ocean market, with nearly a billion people creating PPTs every month, but either making them is time-consuming and laborious, or the results are less than satisfactory.

Although these years there are Keynote, Google Slides and other tools have appeared one after another, PowerPoint's position is still as stable as Mount Taishan, because the Office family bucket is really too powerful ......(DOGGIE)

So they're re-targeting the newparadigm shiftOn, hopefully, a return to the essence of PPT production.

First started a lot of user research and prototyping, forced myself to use my own product every day, and had to do weekly debriefing with Gamma, and finally summarized the three major pain points of the traditional process:

• Formatting: Users usually spend 80% to beautify PPT, only 20% to focus on the content.
• Viewing Review: if the PPT doesn't do a good job, the audience won't buy the content.
• Structural rigidity: The fixed structure of PowerPoint forces people to start designing, but the vast majority are not professionally trained.

After two full years of pure product iteration, they launched a public beta in 2022.

Initially the numbers were so impressive that upon launch it took the day, week and even the month'sNo. 1 on the Product Hunt listEnrollment jumped from 100 per week to 3,000 per week, a 30-fold increase.

After a few weeks, the growth rate started to slow down and user retention wasn't enough, as the vast majority of people still felt that theGamma making PPTs is too hard.

And by this time, they had so little operating capital left that investors even pressured them to switch tracks.

Jedi: Black Traffic is also Traffic

tillChatGPTThe emergence of AI made them realize that AI might be able to solve users' biggest headache - beautifying slideshows.

When all was said and done, they chose to throw in the towel and announced, under pressure, that they would not lay off any staff, would not transform, and would leave themselves only three months.100% all in AI.

It was also during these three months that they completed six core features, including AI generation, intelligent layout and visualization styles, AI design suggestions, PDF export, etc., which later became the core of Gamma's play to retain users.

So things took a turn for the worse when the official version arrived.

They engaged in an extremely bold marketing campaign, producing promotional videos that exaggerated the product's current capabilities and clutching highly controversial tweets:

The most valuable skills in business are about to become obsolete. Designing a killer slide deck used to take days, now AI can do it for anyone in seconds.

Unsurprisingly or surprisingly.Paul GrahamSaw this one and retweeted it angrily to denounce them on the“Nonsense.”, a rumor spread and various AI netizens came down to discuss it, eventually making the tweet explode ......

And when passing eaters are curious enough to click in, they find out:Gamma generated content is actually “interesting”.

(Plan Pass ✅)

Immediately, the product Gamma began to virally sweep the social networks. First 5,000 new users signed up every day, then 10,000, 15,000 ...... all the way up to hundreds of thousands.

The retention curve has also changed significantly, by as much as five times as much as previously unattainable. And the vast majority of users are willing to pay for the product.

This was the moment they knew they had made the right bet.

In a week's time, they jumped from obscurity to the top tier of AI content tools in Silicon Valley, and harvested a large number of users and investment from Accel, and broke a million dollars in ARR.

But at this time the company had only 16 employees, and even with the later expansion, it only grew to 52 today.

Small team, big revenueGamma's products and management team are all based on the same idea of “fighting big with small”.

Operational Philosophy of Fighting Big with Small

What makes Gamma unique is not the text generation, but theRedesigned presentation interface.

Whether it's PowerPoint or Google Slides, the underlying logic is editing, whereas Gamma focuses on the user's expression - theThe user inputs the inspiration and the AI takes care of the form.

One-click generation from ideas to slides, no need to design from scratch, you can create a brand style consistent and beautiful PPT, but also export format seamlessly to PowerPoint, Google Slides.

As Jon Noronha said:

We're not replacing PPT, we're helping mankind eliminate the fear of the blank page.

In a productivity track dominated by giants like Microsoft and Google, Gamma has found a cutout to stand out:AI Generation + Extreme Beauty + Instant Sharing.

Large companies rely on resource stacking, small companies focus on user experience, find the user's immediate needs “let AI help me express”, and then a fatal blow.

So when it comes to talent acquisition, the company's credo is:Painfully slow recruitment.

Maintaining high standards and recruiting only “A” players ensures that every employee shares the same values as the company. Adopting a "player-coach" model that requires all managers to be involved in both the basics and the direction of the team.

The result is to flatten the organization so that one person leading three people can top the output of the original 20.

In terms of growth strategy, founder Grant Lee uses the following four points:

1. Influencer marketing:

The 90%'s product impact depends on less than the 10% going viral, and it's no accident that this has to be triggered.

Startups first need to list the product's target KOCs(Key Opinion Consumers), offer them a generous bonus for at least six months and let them open new accounts, don't meddle in script direction, and allow creators to be able to play freely.

Then follow up with the top performing creators among them and even hire them as consultants to train others.

Take care that all social platforms are placed, and you can also learn about lead generation channels from onboarding interviewers.

2. Performance marketing and branding:

In order to build a brand, startups have to pay a huge price in time and money and need to be persistent in testing ideas.

Until one of the specific use cases resonates, then work around it.

3. User testing:

Extensive user testing is conducted during the prototype stage of the product to understand how the user experience feels, and if random users agree that the product is simple and effective, then it is released.

4. Dogfooding:

When teams have multiple ideas, Dogfooding can assist in finding the optimal solution.

In short, “ability + luck + believe in yourself” is indispensable, build up your strength and then seize the opportunity, so Gamma realized the turnaround against the wind.

One More Thing

But it's a lot harder to win a turnaround battle like Gamma did in the AI market right now.

Former CEO of Reddit Yishan WongJust recently there were two long posts on 𝕏 singing the praises of AI startups, and even Musk came across the table: that's so true! 👍

In short, his point is that the base model holds the underlying advantage of arithmetic power and other things, and that almost as soon as a new application comes along, the base model can learn it in a few days.

This also means that for the vast majority of startups, the development window is being shortened extremely fast, and to make money, you either need to cash out quickly or play the differentiation route and wait to be acquired by a giant equity buyout.

But either way, it's also highly unlikely that startups will grow into the next generation of AI giants, with the best chances forFinding real-world application segments in highly specialized fields.

For example, Gamma, Perplexity, Runway, Cursor, and other Silicon Valley wonders that went small to make it big have all gone this way.

While the giants are fighting for “stronger models,” they are fighting for“A product that understands the heart better.”.

So once the base model is added with features, startups go up in smoke? It doesn't seem necessarily so.

Reference Links:
[1]https://www.nytimes.com/2025/11/10/business/dealbook/gamma-ai-fundraise.html
[2]https://techcrunch.com/2025/11/10/ai-powerpoint-killer-gamma-hits-2-1b-valuation-100m-arr-founder-says/
[3]https://www.youtube.com/watch?v=VRv8g_44q1U
[4]https://x.com/thisisgrantlee/status/1966874658680303880
[5]https://www.youtube.com/watch?v=i6J9PmZZTq8
[6]https://mp.weixin.qq.com/s/FfNwSqEKtBIeex6rUImrfw
[7]https://x.com/yishan/status/1987787127204249824