Coframe Just Raised $35 Million to Make Websites That Rewrite Themselves


Coframe Just Raised $35 Million to Make Websites That Rewrite Themselves

 

Every website you've ever visited is frozen in time. Same layout, same buttons, same words — whether you're 25 or 70, whether you're buying your first backpack or your fiftieth.

Why is that still true in 2024?

A startup called Coframe thinks it shouldn't be, and investors just handed them $35 million to prove it. The round includes a $25 million Series A co-led by Khosla Ventures and Hanabi Capital, with money also coming from people connected to OpenAI, Anthropic, and Google DeepMind. That's a serious lineup for a company most people have never heard of.

So what are they actually building?

The "Living Internet"

Founder and CEO Josh Payne calls his vision the "living internet." Instead of every visitor landing on the same static page, Coframe plugs into a company's website or ad campaigns and keeps generating new versions of them — testing which ones work, then deploying the winners automatically. Two people can hit the same site and see completely different experiences, built for who they are.

Here's the part that caught my attention: it's not one AI doing this. It's a team of them. Coframe runs a specialized model alongside AI agents that each play a role — a strategist, a designer, a software engineer, a QA specialist, a data scientist. Together they can dream up, build, test, and refine hundreds of website and ad variations in minutes. What used to take a human team weeks now happens before you'd finish a coffee.

Does It Actually Work?

Claims are easy. Results are harder. So let's look at the numbers.

Coframe says it's generated more than $300 million in extra revenue for its customers. Dropbox saw conversion rates jump 158%. Replit saw 410%. Cash App got a 60% bump. One Fortune 500 client saw a 70% lift in conversions within 30 days of signing the contract.

At Dropbox specifically, an experiment that normally ate up four weeks of engineering, design, and QA time got built and launched in 30 minutes. Across thousands of test variants, Coframe's win rate sits at 39% — more than triple the industry average for A/B testing. Customers are testing ten times faster than before.

Twenty percent average improvement in conversion rates across clients that include Dropbox, BCG, Block, Backcountry, and Sur La Table. That's not a rounding error.

Why This Matters Beyond Silicon Valley

"For decades, the web has been frozen in place," said Vinod Khosla, founder of Khosla Ventures. "Coframe is building the living internet... This is a huge leap beyond traditional A/B testing and the next iteration of the internet."

I've built and rebuilt enough web pages over the years to know how painful redesigns are. You guess what visitors want, you wait weeks for a developer, you launch, you hope. Coframe is betting that guessing is obsolete. Payne put it simply: "Instead of waiting months for the next redesign, companies should have websites and applications that get better every day."

What happens when the audience isn't even human anymore? Coframe is already thinking about that. A growing slice of web traffic now comes from AI agents that research, compare, and buy on a person's behalf — think of an AI assistant shopping for you. Coframe is working with OpenAI on its ChatGPT ads product, so a company's ad and its landing page can optimize together as one learning system, built for both human eyes and AI agents.

Payne's path here is its own kind of unconventional. He studied math and computer science at Stanford, then dropped out of both the master's and MBA programs to keep building. Before Coframe, he co-founded Autograph with Tom Brady, a company that hit a $2 billion valuation, and created GPT-Migrate, once the top-ranked project on GitHub. Coframe has also acquired companies founded by two early Optimizely executives, Bo Mohazzabi and Steven Ebin, who now run go-to-market and revenue optimization.

So, What Now?

Co-founder Pavlo Razumovskyi said something I keep turning over: "The way most companies build and manage their user interfaces today is wrong, and in a year or two it will be obvious why that is."

I don't know if he's right. But I know the internet I grew up watching change has barely changed its basic shape in 20 years. Static pages, waiting for a redesign that never comes fast enough.

What happens when the web starts improving itself while you sleep? We're about to find out.

Source: MarTech Cube

Our last article on this: When the Musician Hired to Fight AI Turned Around and Used It