Julie Zhou on how Facebook product designers think
Some products and features that we build are destined to succeed, and others to fail. We want to minimize the time we’re spending on the flops, and maximize the time we’re allocating to the winners.
But of course, distinguishing the successes from the failures is much harder in prospect than in retrospect.
Former Facebook design head Julie Zhou set out to study this question. How do we know, early on in a project, whether this is an idea that’s going to be successful? What came out of her research was a framework that Facebook teams use to minimize time spent on projects that are off track.
At every product review, Facebook teams ask and answer three questions.
- What people problem are we solving?
- How do we know this is a real problem?
- How will we know if we’ve solved this problem?
1. What people problem are we solving?
What’s a “people problem”? It’s a real problem that real humans have. It’s something your users actually want a solution for. It’s the way someone would describe their experience if you asked them on the street.
One thing that happens, JZ says, is that you tend to think in the mentality of your team or company. So you say things like, “The problem is we need to optimize the clickthrough rate of the page.” But that’s not a people problem — that’s a company problem. If you start your design thinking that way, you’re liable to wind up with a solution that users don’t care about, and thus don’t use.
A similar mistake that I’ve observed and made myself is to think in the mentality of social impact, rather than in terms of the individual. So you might say things like, “The problem is we need better collective decision-making tools.” Again, not a people problem. Sure, our society might be better off if we had and used such tools — but products succeed or fail based on whether or not individual people use them and like them. So you need to think in terms of individual people.
JZ gives five questions that you can ask to make sure you have a valid people problem.
- Is it human, simple, straight-forward?
- People don’t typically describe their experience with words like CTR, optimize, integrate. Talk like people talk.
- Is it solutions-agnostic?
- Often we start by saying I’m going to build an app that x, or I’m going to design a website that blank. And already in that statement is an inkling of what the solution is. But what if an app is not the right way, or it’s not supposed to be a website?
- A good people problem statement gets away from already trying to constrain it into a particular solution.
- Is it “We win!”-agnostic.
- It shouldn’t be about your company or your team winning.
- Classic example is “Our service is going to be the best at blah”.
- A person on the street doesn’t care if your service is the one that is the best at that. They just want to know for this problem that I have what is the best solution.
- Does it get at the why?
- Often you have a problem like “people aren’t discovering this page.” But that doesn’t get at the why — the root cause. Is it too hidden? too confusing?
- Functional, emotional, social problems are all fine.
- Sometimes people want to feel like they belong. Feel validated. Those also are things that people might say.
Examples of valid people problems that led to Facebook solutions:
- “I want to talk about an interest with other people who are interested, but I don’t know where to find these people.” Led to a discovery tool for Facebook Events.
- “Not everything I see in News Feed is likable — I want more ways to easily express how I’m really feeling about something.” Led to Facebook Reactions.
- Sometimes a people problem is more of an opportunity, not something people would bring up as a hardship. Example: “Desire to share spontaneously and authentically.” Led to Facebook Live.
2. How will we know this is a real problem?
There are thousands of problems we could devote our life’s energy into solving. Why this one?
- Real problem: a problem that a lot of people have. Not just something we experience individually.
- Know: supported by quantitative and qualitative evidence.
Example: “Not everything I see in News Feed is likable — I want more ways to easily express how I’m really feeling about something.”
- Want to make sure that this is actually a problem that many people faced, and that the solution we had would actually address it.
- Talked to a lot of users. Had them go through feeds, and for each story, describe for us, what was your reaction, what were they feeling. People would say well there should be more ways for me to just say something. What I like about the Like button is that it’s so simple. I don’t have to go and comment and the keyboard comes up and two-handedly type something. I can in one gesture scroll through things and say that I like it. But the only thing that I can do is like and there’s gotta be ways for me to express other emotions.
- We looked at usage of emojis and stickers.
- Looked at short comments like “awesome” or “that sucks”.
- Took the top most used reactions, what we wanted to be universal, and that’s what you see in that tray.
- I want to share spontaneously and authentically in the moment.
- We had the FB live product out for celebrities. Built it for them because it was something they told us they wanted.
- Question for us: is this something that regular people want?
- So we tested it internally. Saw lots of use among Facebook employees. Learned about use cases. Relay race. Weekend adventures. Team meeting.
- Then launched as a test on Android to a small percentage of users. Saw a lot of really creative uses right away.
3. How will we know if we’ve solved this problem?
This is a really important question to ask up front, JZ says. Too often we build it, launch, results are coming in, and there’s tons of data points that we’re tying to interpret and put together. It’s hard at that point in time to be very very objective about whether or not we solved the problem.
So up front we should ask: What would be different in the world if we’re successful? How are those worlds in which we’re successful different from those in which we’re not?What exactly should our expectations be?
TLDR: Set measurable goals. If possible, use metrics and targets.
Measurable is key. This doesn’t always mean numbers or data. It does mean that there’s a clear criterion: if I did this thing and got this result, I know what to make of that result.
Example: “Not everything I see in News Feed is likable — I want more ways to easily express how I’m really feeling about something.” For this people problem, JZ’s team envisioned the Reactions product. And they asked: what would it look like if we were successful?
- People share more feedback with Reactions than they did previously.
- Each Reaction gets a fair amount of usage.
- People who receive Reactions on their posts consider them valuable. Did it make you want to share more in the future?