Equity

The episode delivers genuine data-backed insights - revenue-per-employee benchmarks, historical infrastructure vs. application market cap splits across tech cycles, and the shrinking lag between frontier and locally-runnable models - though these are interspersed with promotional portfolio narration and some generic VC framing that dilutes density.

“Nowadays you're getting companies doing 10, 15, $20 million of revenue per employee, which even 10 years ago or even five years ago that would have been unimaginable.”

“In the Web era, new entrants, meaning companies that were started during the Web era in the infrastructure market, produced $400 billion of new market cap. Application companies that were started during the web era created $3.1 trillion of market cap.”

The infrastructure-will-be-commoditized, applications-capture-the-value thesis is a well-circulated VC argument, and the pendulum metaphor is explicitly attributed to Bill Joy rather than original thinking; however, the nominal-terms infrastructure market cap claim and the trust-gap between social and financial services in Western markets are less commonly articulated.

“infrastructure market caps actually peaked in the year 2000 with 85% or uh, 90% of the market cap concentrated in infrastructure companies. But you fast forward 25, 26 years later, and in nominal terms in dollar terms, the market cap of infrastructure companies has not surpassed the 2000 peak.”

“right now we're in the command line era of computing as it relates to AI, It's a text based interface.”

Chi-Hua Chien is a genuine practitioner with 24 years of venture experience, an originating role in Excel's Facebook investment, a Kleiner Perkins partnership, and a co-founded fund with verifiable large-scale portfolio companies (Toss, Monzo, Facebook); he speaks from an actual investment track record rather than as a thought-leader.

“I was um, a 27 year old associate working at Excel when in the basement of Stanford Business School I found this little company that had six employees”

“I've been doing this job since 2002. So for 24 years.”

The episode is notably concrete: named portfolio companies with revenue and market-cap figures, percentage market cap splits across three technology cycles, Google's specific price drop, and Monzo's UK penetration statistics all ground the thesis in real evidence rather than hand-waving.

“Application companies that were started during the web era created $3.1 trillion of market cap. So 88% of the new value that was created in the web era went to application companies”

“they're now at 25% of the entire UK adult population, has a Monzo account and they're adding 7 to 8% of the UK adult population every single year”

The hosts land a couple of genuine follow-up challenges - pushing on whether infrastructure and application definitions have blurred and whether companies are subsidizing growth - but the conversation mostly flows in the direction the guest steers it, closing with 'I couldn't agree more' and no meaningful pushback on valuation frothiness or portfolio conflicts of interest.

“isn't it also true that many of these companies are giving it away and subsidizing tokens and on the back end investing so much in infrastructure that it's become uh, all out race to get to the public markets”

“it does seem like the definitions have melded quite a bit. You may call a company like OpenAI an infrastructure company, but it also is verging into the application layer”