Keith Rabois' opening keynote went back to first principles, and asked that we all not conflate growth with growth hacking - growth hacking is what you do only after you have growth. You need to prove out that you have a core value prop, an asset that people like and use. Once you begin to see signs of growth, only then you can do some growth hacking.
value -> growth -> growth hacking
Keith gave some great examples of how he's seen this play out at top companies, and summarized his learnings into 3 big lessons.
Early 'Growth Hacking' stories at LinkedIn, PayPal, and YouTube
At LinkedIn, one of the early drivers of growth was search engine indexing. In 2005, this was a pretty radical and novel idea. At the time, Keith was also on the Yelp board and he had seen that they had been having success indexing restaurants. It was working well because restaurants are great long-tail content - which Keith recognized was also a strength with LinkedIn profiles. After spending 3 years collecting users and building profiles, that monopoly that had been created by their underlying, steady growth allowed them to launch indexing, adding 20-30% daily growth for LinkedIn.
The 'Pay with PayPal' button was another growth hack that looks trivial today, but came from an important core insight. The PayPal team noticed that some eBay sellers were typing 'pay me with PayPal' multiple times across their eBay listing pages - which was a bad, inefficient way for them to get their customers to actually pay with PayPal. They found a total of 54 eBay sellers were doing this. Even though they didn't think that eBay sellers were an interesting or overly important market, PayPal COO David Saks disagreed and thought it was a good market. They went ahead and created the buttons, the initial step that would lead to PayPal's hyper growth on the eBay platform.
Finally, Keith gave a final example with early YouTube. In their case, the 'growth hack' was an embeddable video player that would allow YouTube content to be played anywhere (at the time, a lot on MySpace). However, they needed a critical mass of videos that people actually want to embed, before it makes sense to start having an embeddable player.
First, grow based on real value - then worry about how to amplify.
Growth hacking is an observational science. These examples were logically driven, not empirically driven. In the PayPal example, counting to 54 isn't empirical.
Feature design is the new marketing. This is more radical, and maybe why growth hacking is more popular.
Keith spoke about how when you launch, it's your brand that drives you and creates your company, and it develops online. You have to design it first, but as more and more people pay attention, it changes - and the bad news is, there's often tension between growth hacking and a high fidelity brand.
As an example, Keith brought up an old LinkedIn email. It won all of the tests, was by far the best performing email… but it had a grammatical error. The LinkedIn team ran it anyway. He contrasted this with Square, where the belief is that the brand is more important than any short term, quantitative improvement. He gave this example to emphasize that today, you really need to understand what your brand is, and what you want before you start down the path of growth hacking.