What 35 Billion Sessions Taught Us About Building

Thirty-five billion sessions is a ridiculous number.

It’s easy to look at that and think it’s just a vanity metric, but it’s really 35 billion player decisions:  Someone chose to click, leave, come back, invite a friend. 

When you see that happen 35 billion times, patterns start to emerge.

At Basket, we’ve now powered more than 35 billion sessions across our network, and a few lessons keep showing up over and over again.

1. Retention is everything.

At launch, nothing matters more.

The Roblox algorithm is not a magical box. The algorithm is the players. This is often repeated by those in the UGC and social media space, but there is a lot of nuance to it. Roblox sets importance to certain levers based on what they believe will deliver the best results to grow Roblox as a whole. The most important of those levers is rightfully retention.  

Every game we’ve worked on that crossed 100 million sessions had a strong retention loop before it had a lot of content. We also deeply prioritize long term retention. We create long term incentives to play that don’t just keep our players coming back for a few days, but for months. 

2. The first 10 seconds matter the most.

Most games don’t lose players because they’re bad.They lose players because they’re confusing. Our best-performing experiences all focus on instantly immersing the player without overwhelming them. Roblox is already such a vast platform with so many different genres and games. There is no need to over complicate things. Keep it simple and engaging. 

If people are confused when they join, you will lose the majority of the people who join. 

3. Updates aren’t a growth strategy. They’re the product.

The fastest way to kill a game is to stop updating it.

The companies that consistently win on Roblox do not wait months between releases. They are producing quality updates consistently. So often developers will overthink things and get off consistent updates.

Players notice when a game feels alive and they notice when a games feel uncared for. Even just missing an update by a single day can kill a game. Players need to be constantly engaged and they need consistency. 

4. Small teams that ship beat big teams that plan.

The best games aren’t built by the biggest studios.

They are built by teams that move fast, listen to players, and execute relentlessly.

‍

Thirty-five billion sessions is an incredible milestone for our team, but the bigger takeaway is what those sessions taught us.

Make it easy to start. Give players a reason to come back.Keep shipping.

The rest tends to take care of itself.

‍