I used to think 0→1 meant starting from a blank slate.
Creator Rewards at Pinterest taught me that it usually doesn’t.
When I joined the project, there were already years of decisions behind it. Existing creator tools, technical constraints, previous experiments, and a lot of opinions about what the program should become.
And then we had a few weeks to get an Alpha out.
It was messy, but I think that’s why it’s still one of the projects I learned the most from.
We started with creators, not the reward
One thing I think we got right was not jumping straight into designing the streak.
Before deciding how someone should earn a reward, we spent time understanding what being “consistent” actually meant for different creators.
Someone who posts every day and someone who spends a week making one really thoughtful piece of content can both be consistent. But a streak designed around one of them can easily make the other feel like they’re doing something wrong.
That changed how I thought about the product.
We weren’t just designing a way to reward activity. We were deciding what Pinterest was telling creators was worth doing.
That made the mechanics feel a lot less neutral.
How often someone needs to publish, what counts, what happens when they miss a week. Those decisions don’t just shape the product. They shape behavior.
I would’ve brought some people in earlier
We spent a lot of time designing how rewards and payouts could work, only to learn later that some of our assumptions didn’t hold up once legal and policy got involved.
We adjusted, and we shipped. But we also spent time unwinding decisions we probably wouldn’t have made if we’d understood those constraints earlier.
If I were starting over, I’d bring those teams in sooner.
Not because I think more people need to be in every meeting. Usually I think the opposite.
But when you’re designing around rules you don’t control, it’s worth knowing where the edges are before you get too attached to what’s inside them.
The first version doesn’t have to prove everything
I think this was the harder lesson for me.
There were things I really didn’t want to cut from Alpha because they felt important to making the whole experience work.
But we didn’t need Alpha to prove the entire Creator Rewards program. We needed to know whether creators understood what we were asking them to do, whether the reward actually changed behavior, and where our assumptions were wrong.
Some things we cut came back later. Some didn’t need to.
I used to think an MVP was a compromised version of the thing you actually wanted to build.
Now I think about it differently.
The first version isn’t there to prove you were right. It’s there to show you where you weren’t.
And you need enough room left to do something about it.
The case study for this work: Pinterest Creator Rewards.