The Underlying Problem of Facebook’s Product Design Approach
It is undeniable that Facebook is one of the most influential companies in the world right now. And, the key reason behind their success is their continuous product innovation starting from the News Feed to Messenger Rooms. While they managed to maintain 2.7 billion monthly active users, they sure did face several vilifications due to their unintended consequences, Cambridge Analytica data leak scandal for example. I have been looking into and been studying in-depth about Facebook and was wondering if their product mindset has anything to do with or has any influence on it.
Early Facebook Days
Mark Zuckerberg is always the center of focus for Facebook. And, he had a lot to do with how they designed their products. And, this is one part that I particularly like about Zuck: he is a product-first person. To bring in a bit of evidence, you can watch his interview with Sam Altman from Y Combinator. He mentions how there is about tens of thousands version of Facebook out there and they test it on a different subset of users to learn and make decisions faster than anybody. As a piece of advice for future entrepreneurs, he said to stop thinking about the business in the first place. They should focus on building a great product first. Only if they succeed in it, they should think about building a business.
And, I completely agree with him on that. When you discover a problem and come up with an idea, try building a product out of it first rather than thinking about what business model to go for or how much Series-A you should go after. Building a great product requires a lot of effort and patience. If you succeed in it and find a good product-market fit, making money is only the logical sequence. And, that’s the very same thing Zuck and co. did in their earlier days. Starting with Harvard, they quickly expanded their services in different colleges in the U.S. There wasn’t any ads or revenue from Facebook for the first three years or so. Their only big focus was on building a great product that people loved to use.
Build Fast, Break Things
Facebook adopts the “BUILD (fast), SHIP, ITERATE” model for designing their products. And, they sort of uphold it as their design gospel. Every talk or lesson by a Facebook Product Designer you see on the internet, this is the one thing they will keep preaching to you. For example, I was watching this BBC documentary on Facebook made in 2019. At one point while talking about Facebook’s projects on AI, their Head of AI says, “Unlike companies like Apple who focus on making the most perfect product first and then sell it, we ship our products as fast as possible at Facebook. And then, make corrections over it.”
You won’t be surprised to know who originated and popularized this gospel. Took a guess? You’re right, it’s Mark Zuckerberg himself. The Y Combinator interview I talked about earlier was taken sometime in 2016. Almost 10 years back, Mark Zuckerberg joined the Harvard CS50 for a guest lecture about Facebook and computer science with Professor Michael D. Smith. (he starts his lesson with a yo) During the later part of this lecture, he talks about how Facebook has been moving so fast compared to its competitors like MySpace or Friendster. Constant product innovation and shipping them quickly to the users was the key. At one point he said something that might sound pretty ironic to you right now: “It’s really important to do something instead of worrying about what end-result will be.”
Built Fast, Broke Things
I believe, at this point, you can guess what I am trying to point out. Their approach or principle in designing products worked out unprecedentedly well for over a decade. It only started to fall short or fall apart from 2014/2015, ultimately leading to the Cambridge Analytica scandal. If we pinpoint the cause of the Cambridge Analytica scandal, it was because of a flawed API policy on Facebook’s end. Facebook Apps were pretty popular back in 2012/2013 and their Graph API was very popular among the developers. I too used to make Facebook Apps as freelance works. And, I clearly remember how fast they were adding new API endpoints for the developers. It was a very exciting time to be a Facebook App Developer at that point. And, I can only imagine how such a big flaw in their API policy went unnoticed. Because they were too busy building the next endpoint and shipping it.
And, Cambridge Analytica is not the only example of things breaking down for Facebook. The BBC documentary I talked about earlier has a very good example. It shows how a product designer on Facebook took up the project of making a blood donation product on Facebook. She led a team of researchers to map out the product detail after a series of brainstorming and empathy hunting. They took very little time to develop the product. In October 2017, they launched it in Hyderabad, India. It worked out pretty well. So, they started to roll it out pretty quickly on the other cities of India. And then, one sudden morning they found out their product was being used as a medium of black markets for blood!
Build (a bit) Slow, Fix Things
Facebook is undoubtedly one of the biggest and most amazing innovations of the 21st century. During the early days, Facebook faced numerous competitions from the biggest tech giants of that time like Yahoo!, Microsoft, and particularly Google. It made sense for them to build fast and break things back then. Perhaps it was the only way for them to become what they are today. And, that is understandable.
But, I think it’s about time Facebook rethink their mindset and approach to designing products. They no longer have any daunting threats from any of their competitions (if they have any!). It can even be argued that they are just too big to fail at this point. And, they no longer serve college students from the U.S. or U.K. anymore. One simple decision by a product designer/manager from Facebook has the capacity to influence millions (even billions) of lives around the world. Both positively and negatively. Just think about it: the outcome of the US Presidential Election 2016 might have been completely different if some product manager back in 2012/2013 did not ignore to implement one API policy for an endpoint.
I believe it’s about time they adopt the product design strategies of either Google (launch gradually) or Apple (build perfect). When you are at the position of influencing the whole world through your platform, you better take every bit of precaution rather than relying upon your gospel which has failed you in recent times.