part0010

The Twitch safety team wasn’t happy. They were responsible for making Twitch a safe video platform for creators and viewers to interact around live streams. We were reviewing raids, a new product that would allow any creator to send their viewers to another creator's channel. During beta testing, most creators used raids to help each other grow. However, in a few cases, creators raided each other to throw insults and abuse.

The safety team was mad because I wanted to cut a feature that would allow creators to shut down abusive raids. My rationale was simple—there were already existing safety tools that creators could use to manage raids and we needed to cut something to meet our launch date.

I listened to the safety team’s argument. “Yes, it might be true that 99% of raids are good,” they said, “but the 1% of raids that are bad are some of the worst experiences that people can have on our platform.” I started to think about the worst-case outcome: We ship this product, and people abuse it on day one. Even if creators can shut down this abuse with existing tools, building better tools will show the community that we cared about protecting them against bad behavior. I realized that the safety team was right.

I said, “You’re right, we can’t cut the safety feature. But can we work together to design a simpler solution that will still allow us to meet our ship date?” Suddenly, the meeting became very productive. We worked on simplifying the solution, and the safety team even agreed to help build it. A few weeks later, we launched raids on stage at Twitch’s annual creator conference and made safety our primary message. Our community loved that we built safety tools on day one.

When you get into disagreements with others, remember that your job as a product manager is to find the truth, not to be right all the time.

Because people expect PMs to know everything about their product, I used to spend hours looking at customer research and crafting a polished document before sharing it with anyone. I would then go into a meeting with a goal to convince everyone else to see things my way. This approach is inefficient.

It’s inefficient because no matter how much preparation I do, there’s always a chance that I could be wrong. So instead of going into a discussion with a goal of “How can I convince this person to see things my way?” I now have a goal of “How can we discover the truth together?”

The fastest way to find the truth is to seek knowledgeable people who are willing to disagree. Knowledgeable people could include your customers, your teammates, or anyone with relevant experience. After you form an initial opinion about a decision, find others to debate the decision. During these debates, identify significant unknowns together (“How do we know that users want this?”) and follow-up on these unknowns as quickly as possible to find an answer.

With the help of other knowledgeable people, you’ll quickly find questions you’ve missed or holes in your logic. If you’re wrong about something, admit it early. There’s no shame in changing your mind to get closer to the truth.

Even with the help of other knowledgeable people, you’ll likely won’t have all the information that you need to make a decision. It’s useful to think of decisions as one-way or two-way doors. Two-way doors are decisions that can be easily reversed, so prioritize decision speed even if you don’t have perfect information. One-way doors are rarer. These decisions are hard to change, so try to gather more information if you’re uncertain.

Every decision has two phases: 1) gather and debate; and 2) commit and execute. During the first phase, if you genuinely believe that the decision is wrong and have the evidence to back it up, then disagree openly. It’s always easier to compromise with the decision-maker to avoid damaging a relationship, especially if that person is your manager or an executive. But your goal should be to find the truth no matter what, even when doing so is uncomfortable .

After a decision is made, you must commit yourself and your team to execute on it. Even if you don’t agree with the decision, it’s your job to make sure you understand the rationale behind the decision and explain that to your team. It’s never acceptable to say, “Because my boss said so.”