Raj Shamani is one of the biggest podcasters in the world. His YouTube channel Figuring Out has nearly 20 million subscribers, putting it just behind Joe Rogan and Steven Bartlett’s The Diary Of A CEO .

But the way his team decides what to make is almost as interesting as the scale. Shamani says an episode of Figuring Out can begin with a thumbnail, before anyone has booked the guest.

His team first studies what people are discussing across Reddit, X, LinkedIn, Instagram and YouTube. It mixes those conversations with news and data from Shamani’s own channels, identifies a handful of questions that could matter in the coming weeks and turns them into 30 to 40 thumbnail concepts. The best-performing ideas are passed to another team, which looks for the person best equipped to answer the question.

“We are not trying to predict virality,” Shamani told me during a recent discussion on my Business of Creators podcast. “We’re just trying to understand human curiosity and how is that shaping up on a month to month basis?”

If the operation works as Shamani describes, Figuring Out has pushed YouTube’s culture of testing upstream into commissioning. Audience data helps decide what gets made, rather than simply measuring what worked after publication.

Just this week, WME has signed Shamani for representation across all areas and territories outside India. The agency says it will work with him to expand his business and profile internationally, connecting the audience he has already built with opportunities around the world.

“Where you are born should never determine what knowledge you have access to or how big you’re allowed to dream,” Shamani said.

Figuring Out began as the room Shamani wanted to enter

Shamani grew up in Indore, in central India, and began working in his family’s consumer chemicals business as a teenager. He has since become an entrepreneur, angel investor and author. He launched Figuring Out in 2020. In 2023, Forbes India named him to its 30 Under 30 list for House of X, the company he co-founded to help creators build consumer brands.

The podcast came from a more personal frustration. Shamani was ambitious, he said, but didn’t have access to people who could explain how larger companies and careers were built. “Figuring out for me became the room I wish I had access to growing up,” he said.

The early version was wonderfully basic. Shamani recorded on his phone in his room, sometimes with a blanket over his head to reduce the echo. He says his first 53 episodes attracted about 53 listeners in total, and jokes that he was probably half of them. In another early episode he forgot to start one of his 2 cameras, leaving the audience able to see the guest throughout but never the host.

That homemade show now sits inside a much larger company. Figuring Out Media employs more than 120 people across content, distribution, investments and education. The podcast operation includes research, production, guest outreach, social distribution and what Shamani calls an “errors and experiments” department.

Shamani describes himself as an entrepreneur. “I would say I’m a founder who happens to host the flagship product,” he said. His ambition is to build an education-first media company, with the podcast as one part of it.

A broad promise gives the show permission to change subjects

Figuring Out is built around long interviews about business, money, health, relationships and public life. Shamani’s framing is wider still.

“Figuring out was and is built around just one specific question. How do the world’s most extraordinary people think? And what do they understand about the world that the rest of us don’t?” he told me.

The recent guest list gives some sense of the range. Over the past few months, Shamani has interviewed neuroscientist Andrew Huberman, author Robert Greene, NYU professor Scott Galloway, cricketer Smriti Mandhana, astronaut Shubhanshu Shukla and former Austrian chancellor Sebastian Kurz.

The show started with a much narrower business brief. Shamani says it moved into health when his team realized that its pursuit of work and growth was coming at a cost. Relationships followed, then wider questions about the world.

“The breakthrough was when people stopped following us for a topic and started trusting the promise that we will help them build an extraordinary life by understanding really good people around the world,” he said.

That promise is commercially useful. A show tied to one category eventually runs into the limits of that category. A show built around the host’s ability to extract useful knowledge can move with its audience, provided the audience continues to trust his choices.

The question can come before the guest

Shamani calls the internal audience-analysis system the Wave Machine. It monitors conversations across large platforms and smaller communities, combines them with news signals and then compares them with performance across approximately 100 experimental channels.

The output is 5 or 6 subjects the team believes people could care about next. The system then proposes 30 to 40 thumbnail variations. Once those have been tested, the 3 with the highest click-through rates go to the guest team.

Interview shows have traditionally booked an interesting or famous person and then worked out what to ask. Shamani’s version starts with a live audience question and shops for the expertise.

Thumbnail testing is already standard practice for many large YouTube creators. YouTube itself lets creators compare up to 3 thumbnails , using watch-time share rather than clicks alone to determine a winner. Shamani’s team says it takes that discipline further back, using packaging signals before the interview has even been commissioned.

Figuring Out has added about 5 million subscribers in the last 6 months.

Research turns the host’s curiosity into a company process

Shamani’s preparation system is as detailed as his commissioning system. Depending on the guest, he says his researchers prepare between 40 and 100 pages drawing on previous interviews and public material. The team sorts potential material into 4 zones: gold, red, green and boring.

The gold zone is the target. “And the gold zone is can we steal something from their head which they also haven’t spoken anywhere else?” Shamani said.

That description gets to the craft of a good interview. Research gives the host somewhere new to go. It also gives him the confidence to listen, abandon the prepared sequence and follow an answer when it becomes more interesting than the next question on the page.

Shamani says the company records 220 to 230 interviews a year and publishes around 150. “The rest we just reject and delete them,” he said. “We never put them online because the guest wasn’t able to teach something new.”

There is an appealing tension here. The Wave Machine tries to quantify audience curiosity, while the research team and host decide whether a conversation has earned publication. The data finds an opening. Editorial judgment decides what passes through it.

That judgment becomes more valuable as the tools become easier to copy. Any well-funded show can monitor social platforms, test thumbnails and use artificial intelligence to process a large volume of audience comments. Far fewer can choose the right question, persuade the right guest to answer it and hold the audience’s trust while moving from entrepreneurship to neuroscience or international affairs.

Many creator businesses struggle when the founder’s instinct has to become a repeatable process. Shamani is trying to encode his instincts across a team. He says his role alternates between “chief support officer” and “chief standards officer,” clearing obstacles for staff and then raising the bar for what they produce.

The wider company now includes education, investment and social distribution businesses alongside the show. Each draws on the same basic proposition: find people with hard-won knowledge, turn that knowledge into something useful and distribute it to a large audience.

The WME signing gives Shamani more machinery for taking that proposition overseas. It also arrives with Figuring Out within 1.3 million subscribers of Joe Rogan on YouTube.

“By joining forces with WME, we can take this mission to the world and build a truly global institution together,” Shamani said.

If Shamani maintains anything close to his recent growth, he is likely to pass both Joe Rogan and Steven Bartlett before the end of 2026 and become YouTube’s most-subscribed podcaster.

It would be a huge achievement. An interview show started with a phone and a blanket in Indore would have overtaken the 2 English-language podcast channels that have set the international benchmark for this format.

“For me, everything is a ten year game minimum,” he told me. “So we’re halfway there. We’re still early.”

The subscriber race may be settled within months. Building an institution that can carry Shamani’s methods beyond its founder will take much longer. Its strongest asset today is his editorial instinct: knowing which audience signal deserves attention, which expert can answer it and which recorded conversation should never leave the hard drive.

This article is based on an interview with Raj Shamani on my podcast The Business of Creators .