High signal digest #2

The Cold Start TLDR, TikTok poison pill, the incompetent leader and more

2 things on the menu:

1. book summary = The Cold Start by Andrew Chen is about scaling networks. The book has useful terminology, first-hand anecdotes, and specific tactics for growing a network. My summary will focus on the tactics.

2. reading = TikTok poison pill and SEO, the playbook of the incompetent leader, thinking about thinking, and how to be a magnet for talent.

The Cold Start TLDR:

First some context. Andrew led driver growth at Uber pre-IPO. Drivers are the supply side of the network and is the hardest side to scale. Andrew has access to a # of people “who made it” and so the book is rooted in a # of quotes and anecdotes (from Airbnb to Twitch to PayPal) that make all recommendations real. I found the stories interesting, especially everything related to Travis-era Uber, but I am going to skip the stories and the terminology.

I will focus on recommendations because this is a TLDR. My hope is that if you are building something - you find it helpful and/or it makes you go “hmm”.

This is the 5-stage framework outlined by Andrew in The Cold Start

• Figure out how to get a small group of the right people (atomic) at the same time (concurrency) using the product the right way.

• Hard side is usually the supply side. Twitch=creator, Uber=driver, Airbnb=host, Wikipedia=volunteer.

• What are their hobbies and side hustles? What does your hard side do on weekends? Help them and solve for that to appeal to them.

• Create magic moments. Avoid creating zeros - a zero is an experience where a user churns from the app (ex. driver never shows up or airbnb host cancels).

• Network density matters. You need the right people on the network. 10 people from 10 different teams on a slack network is not helpful.

• Copy paste your solution to grow from one atomic network to another

• Killer product is an edge. “Just works” is an edge (ex. Zoom).

• Invite-only is a way to curate the network (ex. gmail, early FB). Start with a POV on who is best for your network and go after them.

• Come for the tool and stay for the network (ex. GitHub). Tip from a tool/product to a network → stronger position.

• Spend $ to grow via subsidies and growth tactics. This is okay if you believe your unit economics will work out in the long run.

• Flinstoning is faking the hard side. Start posting on your own (ex. reddit) → manual team → add some automation → full automation or replaced by market forces.

• Understand who are the power users and who are the tourists, but do not try to get every user to use all features (tempting, but wrong).

• To achieve escape velocity leverage (1) in product growth loops, (2) reactivation via email, (3) a/b testing, (4) subsidies, (5) segmenting users into high and low value, (6) in product viral loops.

• Growth curves are never smooth and all teams hit a ceiling. How to break through? (1) continue to innovate, (2) add more nodes to network, 3) adjacent user theory or layer cake = grow via new formats, new geos, new use cases. Layer cake cheat code for large companies is M&A.

• Be wary of law of shitty clickthroughs. Advertising decreases in efficacy over time. Best to solve growth via product and ops.

• Network revolt may happen. You want to nudge your supply side to professionalism. The solve is to embrace this.

• Network quality matters. It’s not always about who ships more features, who has more money, or a larger selection in their marketplace. How good is the experience? Wimdu vs Airbnb is a great case.

• How do you fight a larger player? (1) cherry pick markets from larger incumbent to win, (2) focus on your speed - you will move faster and you can be much more focused as David, (3) have a killer product.

• Avoid platform dependence. If integrated too closely - you become a feature of a larger network and/or you will be copied.

• Avoid illusion of a big launch (google+). Healthy networks are built from the ground up.

I’ve omitted everything related to early-Uber, but I found all of it fascinating. As for the recs, some of these are familiar (ex. avoid platform dependence, flinstoning, make a killer product), but it was helpful to see them framed from Andrew’s POV. Some I wish I’ve known earlier (ex. law of shitty clickthroughs, hobbies/side-hustles, big launch illusion) and will now apply.

On a personal level - having thought through this for 2 months - my biggest takeaways are that for all future products I will be (1) more cognizant of the 5 stages with a plan for how our product will break through each stage and (2) rely even less on paid media/press and more on product/ops to grow.

Reading snippets:

• TikTok - 2 new things that I found interesting. First, TikTok is a poison pill that isn’t just pushing all social media giants to lean into short-form video, but also pushing FB/YT/TWTR (if you see shirts that go hard on your twitter feed you know) into ditching their social graphs in favor of a data-driven engagement model. Founder-led Meta is making the boldest pivot. Evolve or die right? Except in his New Yorker article Cal Newport concludes that as the social media giants ditch the social graphs, they also ditch their competitive moats, and end their dominance as they are now susceptible to new engagement models. TikTok is also not immune to this paradigm shift. Think this is all long-term healthy.

• Second, TikTok SEO is a thing. Watch this 15 sec TikTok. Tested this trying to find a dinner recommendation in NYC and it was a good search experience. Expect us to see more on this.

  

  source: no-good.io

• Great thread on the playbook of the incompetent leader who empire builds, achieves nothing, and fails up. It is scary how accurate this is. I know you are nodding. There is a near 100% chance we’ve all worked with someone like this, right? What worries me the most is how good these folks are at faking, saying the right things, and getting the next job. What do we think about the solutions proposed in the thread?

• Thinking about thinking: people underestimate how enjoyable and engaging just waiting is. Good research piece from the American Psychological Association. Direct contradiction to Pascal, but I strongly side with the APA. Think one of the true measures of happiness is when you are happy just being alone. Time for your walk or workout of the day if you’ve yet to go?

  

• Pointed set of recs on how to be a magnet for talent. I strongly agree with looking for tenacity/grit and the “sell-at-all costs” mentality. At a macro level I am convinced that spending a lot of time thinking through recruiting is worth it. All early hires have such a large impact on the business and the culture. And above all, I think it all comes down to what we are working on and who we are doing it with.

I have more interesting reads on impact of early tech decisions, how to learn anything, dall-e implications BUT think we might be at capacity for screen time together.

be good 🤝