Retention Is Key Product Indicator

When launching Instagram's Bolt app, Peter Dang discovered that retention—not user numbers—is the ultimate product success indicator.

The Instagram team had created Bolt, a camera-first app designed to reduce the pressure to share. They had every advantage: Instagram's world-class design team created a beautiful interface, their engineering team built a lightning-fast app, and they had the distribution power of Instagram behind them. On paper, it should have succeeded.

But when they launched Bolt in New Zealand/Australia, the data told a different story. Peter explains: "Retention is the key indicator in any product that you build—it's not the number of users, it's not the volume, it's actually retention and cohort retention."

The team closely monitored their cohort retention graphs, looking for the pattern that signals product-market fit. As Peter describes it: "You plot the line, and if it asymptotes, then you're in a good spot because that means people over x period of time will continue to stay on the app."

For Bolt, that asymptote never materialized. Despite having Instagram's best talent, resources, and distribution, they couldn't predict what would resonate with users. The retention curves showed people weren't sticking with the product.

This experience reinforced a crucial lesson for product teams: you can have the best team with impeccable product taste, but you can't always predict what will succeed on the first attempt. The key is to view these situations not as failures but as learning opportunities. As Peter quotes, "It ain't a loss, it's a lesson."

The team salvaged what they could—some technologies built for Bolt were later incorporated into the main Instagram app—and moved on, having gained valuable insights about user behavior and retention patterns.

For product teams today, this highlights why retention metrics should be your north star when evaluating new products. User acquisition numbers might look impressive initially, but without that retention curve flattening out and asymptoting over time, you don't have a sustainable product. Focus on building something people keep coming back to, not just something they'll try once.