Clubhouse Investor Andrew Chen on How Startups Can Stop Bleeding Users

  • Clubhouse investor Andrew Chen says a “killer product” sometimes isn’t enough to retain users.
  • While working at Uber, he learned to measure what happened when the user experience breaks down.
  • This data helped Uber build a reliable service, Chen writes in his book, “The Cold Start Problem.”

Andreessen Horowitz general partner Andrew Chen was working from home when he first stumbled onto Clubhouse, then a simple app that let people speak to other users, all in a single chat room.

Soon after, Chen logged on and found friends he hadn’t seen in months, because of the pandemic lockdowns. Other times, he eavesdropped on riveting conversations about robotics, Bitcoin, or the history of tech. In his new book, “The Cold Start Problem,” he describes these spontaneous interactions as “little bursts of magic.”

Chen became a devout user and, later, an investor in the startup that rose in popularity during the pandemic. In the following months, Clubhouse accrued millions of users a month as it offered the homebound online masses a way to spend some of their suddenly abundant time. All of that led to the startup achieving a $4 billion valuation within a year.

In his book, Chen credits the company’s trajectory to “a killer product, with a release at the right time.” But he admits that a tech startup’s work isn’t done after it’s secured new users. The founding team has to invest in those networks to keep users from leaving.

Clubhouse’s popularity has faded with vaccinations rising and more people returning to normal life. Daily downloads of the app have plunged more than 90% since a peak in June, while daily average users are down almost 80% since February, Apptopia data shows.

And recent Insider interviews with creators, advertisers, and startup investors show a startup struggling to build an audience and keep it.

In his book, Chen, who previously led the rider-growth teams at Uber, shared a lesson that could help the app stop bleeding users.

Zero to hero

The work “doesn’t stop once a networked product has established its first atomic network — it needs to be continually solved by the network as it grows,” Chen wrote. In other words, it’s not enough for an early sample of users to find utility from a product. The product needs to engage more networks for other users to join and enjoy it.

Chen recalled the concept of “zeroes” at his past employer.

“A zero at Uber was the worst experience you could have, when a rider opens the Uber app with the intent to pick an address and pick up a ride — but there aren’t any drivers in the area,” he wrote. 

For Clubhouse, a zero might be when a user opens the app but can’t find a room that interests them. Or worse, they see only rooms driven by controversy, known as “drama rooms,” and quit the app.

Chen stressed the importance of measuring a product’s shortcomings, in order to mitigate them.

The product teams at Uber had dashboards that showed them how often these breakdowns happened and where, Chen said. They tracked the percentage of users that were “seeing zeroes,” and used the data to suss out where they needed to deploy more drivers. When Uber made sure users could hail rides, fewer users churned. 

The same approach could help companies like Clubhouse discover what they need to do to provide users more “little bursts of magic.”

“Thinking about zeroes and unfulfilled requests was such a useful concept at Uber that we baked it into many of our common dashboards, split by city and region so we could understand how often it was happening,” Chen wrote. “I encourage product teams to develop their own form of this metric, laid out as a dashboard of networks — whether that’s divided by geography, product category, or whatever else makes sense.”