The Cold Start Problem
Photo of david

david &
The Cold Start Problem by Andrew Chen

Edition
ISBN 9780062969750

Reviews

Photo of Julian Paul
Julian Paul@julianpaul
3 stars
Mar 24, 2024

This book could have been a 5-part essay series. Nevertheless the concepts shared are helpful to understand network scaling strategies for startups.

+3
Photo of Furculita Alexandru
Furculita Alexandru@furculita
5 stars
Dec 20, 2023

The best book on startups a founder needs to read before launching a product.

Photo of Jaiden Ratti
Jaiden Ratti@jaiden
3 stars
Dec 18, 2023

(3.7)

Photo of matej yangwao
matej yangwao@yangwao
4 stars
Aug 22, 2023

** spoiler alert ** Designing tokenomics in web3? You might like it. Other that nicely summarize stages of growth your business whichever are highly anticipated to go through so you can at least prepare well. Surprised noticed high seed network to have in this book mentioned, seems like verified tactic. > Network effects describe what happens when a brand's products get more valuable as more people use them, and one can apply Metcalfe's Law when studying these effects. >each time someone joins an app with a network behind it, its value doesn't just increase; it rises to the square of the number of nodes in the system. >The Cold Start Theory is crucial in building network effects. It's composed of five stages, which are important to creating a successful brand. These stages include: • The Cold Start Problem • Tipping Point • Escape Velocity • Hitting the Ceiling • The Moat Tipping point >However, this stage is critical because everything happens quickly, and if you start losing users, others will leave too. And if the user base is increasing, you're sure to keep having more people join. >Escape Velocity Once you've passed the tipping point, the next thing is to work furiously to strengthen your market's network effect and sustain the growth you're seeing. >Hitting the Ceiling Rapid growth can hit the ceiling even after reaching the tipping point. So you keep doubling down on efforts and tweaking what isn't working. >The Moat This final stage of the framework is all about using network effects to stay far ahead of competitors. >Gladly, the atomic network, which is the smallest network with enough people, can solve this challenge. Your atomic network should consist of early adopters who are passionate about the problem you're solving. Once they experience the practical value of your product, it will be hard for them to leave it for something else. >Usually, at the start of an atomic network, an important dynamic increases over time: a minority of users form the hard side of your network. They do more work and contribute more to your network, but they are harder to acquire and retain. >When they work, networks generate cross-side effects, where users on one side of a network benefit more than the other. More importantly, acquiring the hard side of the network and keeping them happy and stable is paramount. However, note that users on the hard side do more work and have higher expectations; hence it is challenging to engage and retain them. >If you want to build a global brand and dominate your market, it's not enough to create just one atomic network; you have to develop two or more. >Growth sometimes happens city by city or across teams. And when this occurs, the network gradually hits the tipping point when a product can grow rapidly and take over the entire market. >Dating apps suffer more cold start problems because if the app is booming, and matches the right people, ironically, you have two delighted people who will quit using your app. >To grow rapidly, you may want to take advantage of the invite-only strategy. Companies like LinkedIn and Clubhouse have utilized it to grow an extensive user base in their early days. >The invite-only strategy works this way: the most connected people get invited earlier, then they invite other highly connected people, and the cycle continues. >“come for the tool, stay for the network” >Then, over time, it directs its users into a network by collaborating, sharing, communicating, or otherwise interacting with other users. >While Hipstamatic built a great tool, Instagram used network effects to win the market. The Instagram versus Hipstamatic story exemplifies the “Come for the tool, stay for the network” strategy. >A popular strategy for bootstrapping networks is what I like to call come for the tool, stay for the network. >The crux of this strategy is to initially attract users with a single tool and then, after a while, get them to participate in a network. The tool helps get to the initial critical mass; the network creates long-term value for users and stability for the company. >“Come for the tool, stay for the network” helps overcome the cold start problem and makes it easier to create network effects. >escape velocity phase is when you work on an already successful product to make it realize its full potential. >An example of a company that hit escape velocity is Dropbox. At first, Dropbox partnered with a mobile company to provide photo backup services. The Dropbox team then realized that the partnership generated a lot of new sign-ups, but they were all low-value users. >The company learned the networks and attributes of its most valuable users throughout its founding journey, introduced vital features to appeal to businesses and added new marketing channels. These efforts helped Dropbox scale its network effects into escape velocity, leading it toward a successful outcome. >When new products become successful and start to scale, it’s often called hitting “escape velocity.” >Product teams struggle to address the underlying causes when a business hits its ceiling. Market saturation and network revolting are some forces that cause this, but lower-quality engagement from new users, regulatory actions, and spammers are also to blame. >Network effects can weaken and unravel just as fast as they are gathered, pulling down acquisition, engagement, and monetization all at once. >Businesses don’t suddenly hit the ceiling; they bump up against plateaus over time. >For instance, network revolt occurs over time. As a network grows, the hard side often gets more concentrated and potent and begins to act accordingly. The users on the hard side may start to demand more or even revolt, making it tough to keep everyone happy and causing grave consequences. >The Moat is the final phase of the cold start theory. It is about a thriving network that defends its turf, using network effects in a constant battle against smaller networks trying to enter the market. >When there's a battle between two networks, competitive drivers shift users from one to the other. The best place to focus in the market is the hard side of the network because it's where you can get the most leverage. >More extensive networks are fearsome not just because of the network effects of scale but also because of their ability to expand into novel categories and markets. Because they have preexisting networks, they can easily launch new products and features, using their earned network as launching pads. This phenomenon is known as bundling. >A good example of a company that used bundling well is Uber. They already had a good user base before launching UberEats, so getting customers onboard didn't feel like starting a new business from scratch. >Bundling is the phenomenon of selling different products together as a package. >Bundling works well for tech products because, at the minimum, it can drive customers into a new product or feature that could have otherwise struggled to get traction. Bundling is also an excellent way to beat the competition — when a new brand is springing up with a product your audience likes, you can simply add that feature to your existing product. The fact that you're already popular means people will likely prefer you to the new brand, something Meta (formerly Facebook) has been doing. >The good news is businesses can grow with the help of network effects. And as seen in this summary, there is the hard side and the easy side for every network effect. The hard side is the most important because it creates more value and helps a business scale up. Without this critical group, your network will struggle to make progress. >A trio of network effects engages and enables more users, thereby helping businesses hit escape velocity. They are the engagement effect, the acquisition effect, and the economic effect. >If you are completely illiterate but have a desire to achieve something big and have clarity and confidence, then you can achieve it with consistent actions towards your goals.

Photo of Art Lu
Art Lu@artlu
4.5 stars
Aug 20, 2023

Andrew Chen is great at coining memorable phrases: the Cold Start Problem, Meerkat's Law, the Law of Shitty Clickthroughs. He describes a framework for thinking about network effects, and shares inside/outside stories of why specific companies and products have won or lost.


3 stages:

- cold start (an "atomic network")

- tipping point ("network of networks")

- escape velocity (general adoption)


The 3 systems underlying network effects: acquisition, engagement, and economic


He covers famous success stories (e.g., Slack), provides quantitative/empirical metrics, and provides lessons from unsuccesful products:

- IE 1.0 did not succeed despite overwhelming network effects->product must be compelling

- Wimdu lost to AirBnB and went to zero after 2 years, despite a huge lead and hitting all the top-level metrics->didn't solve the Hard Side problem

- Google+ Big Bang rollout hit anti-network effects->fire hose of traffic hid the churn of unengaged users


Reading the book feels like consuming viral threads turned into short TED Talks. Very easy to consume 320 pages, feeling one's brain gain valuable insight with each bite-sized blob of millions of dollars of earned lessons.


Riffing on his a16z colleague Chris Dixon's essay, Andrew writes: The next big thing will start out looking like it's for a niche network. This was an unexpected takeaway for me. I have been dismissive of many early-stage platforms due to the cringe: Twitch, Clubhouse, Uber, even PayPal.


After reading the book, I'm left curious to get more of his candid thoughts on the Threads rollout, Farcaster/Lens, choices made by bitcoin/ethereum communities, alt-L1s, etc.

This review contains a spoiler
+1
Photo of Martin Kuzmanov Sandholt
Martin Kuzmanov Sandholt @sandholt
4 stars
Jun 6, 2023

Easy to read with good examples of network effects in different contexts. A bit boring if you have been following Andrew Chen's blog for while.

Photo of Daniel Bower
Daniel Bower@danielbower
3.5 stars
Feb 6, 2022

I’ve you’ve never read about network effects and the dynamics of multi-sided marketplaces/platforms then this book will provide both a good history of the business model and the strategies to employ the get over the “chicken or the egg” problem these businesses face. If however, you’ve read about these topics before there is very little new here. Chen has some nice turns of phrase that I’d not come across — and having a shared nomenclature is a helpful step forward — but otherwise, none of it felt particularly original.

Photo of BC
BC@b-c
2.5 stars
Mar 8, 2024
Photo of Seyfeddin Başsaraç
Seyfeddin Başsaraç@seyfeddin
4 stars
Jul 23, 2023
Photo of Masharty Tembo
Masharty Tembo@masharty
5 stars
Jan 1, 2023
Photo of Jerric Lyns John
Jerric Lyns John@jerric
5 stars
Oct 25, 2022
+1
Photo of Kwadwo Adu
Kwadwo Adu@kwadwo
5 stars
Oct 20, 2022
Photo of Tharun
Tharun@tharun
5 stars
Oct 15, 2022
Photo of Colin Das
Colin Das@clocs
5 stars
Aug 13, 2022
Photo of Alexis Leon
Alexis Leon@alexisleon
4 stars
Jun 24, 2022
Photo of Yuval Shoshan
Yuval Shoshan@yuvals
3 stars
Apr 18, 2023
Photo of Anthony
Anthony@amorriscode
3 stars
Mar 25, 2023
Photo of joseph
joseph@uncanney
4 stars
Mar 24, 2023
Photo of Jimmy Cerone
Jimmy Cerone@jrcii
4 stars
Feb 4, 2023
Photo of Keven Wang
Keven Wang@kevenwang
4 stars
Feb 4, 2023
Photo of Drew Spartz
Drew Spartz@drewspartz
5 stars
Jan 26, 2023
Photo of Oz Lubling
Oz Lubling@ozlubling
4 stars
Jan 20, 2023
Photo of Simao Freitas
Simao Freitas@simao
5 stars
Jan 19, 2023
Photo of Andre Schweighofer
Andre Schweighofer@dre
5 stars
Jan 13, 2023

Highlights

Photo of M J
M J@mjhere

When you look at the small group of writers who make more than 100+ edits in a month, it's about 4,000 people. As a ratio, it means that active contributors represent only 0.02% of the total viewer pool. The motivations of these contributors are worth studying. Within the most hyperactive group of editors isa contributor named Steven Pruitt, a records and information officer for US Customs and Border Patrol Protection. Steven edits Wikipedia in his spare time-but when I say spare time, it's almost like a part-timne job. CBS News covered his contributions to Wikipedia in 2019, as the editor with the most edits on English Wikipedia …

Page 83
Photo of Piet Terheyden
Piet Terheyden@piet

For example, in Twitch’s journey, the team deeply focused on creators, giving them better tools and monetization, which in turn caused them to become more active. More satisfied creators meant they would broadcast live video streams more often, bringing in more viewers, which drove further engagement and monetization. It might have been easy to think they should just double down on their marketing spend, but instead the team looked toward amplifying the network effects that engaged its streamers.

Photo of Piet Terheyden
Piet Terheyden@piet

Had LinkedIn started with an undesirable set of users, it likely would not have become a magnet for the true believers that continued to onboard their friends over time. Had Tinder begun somewhere besides USC—let’s say in a small rural town—it wouldn’t have been able to build campus to campus, and then large cities and then on from there. It would have changed the whole strategy. For networked products, the curation of the network—who’s on it, why they’re there, and how they interact with each other—is as important as its product design. Starting with a deliberate point of view on who’s best for your network will define its magnetism, culture, and ultimate trajectory.

Photo of Piet Terheyden
Piet Terheyden@piet

There could have been a world where Reddit built many internal studios—one for their “cute” sub-Reddit community, another for sports, yet another for music—and hired full-time moderators as employees of those studios to create the necessary content. While this isn’t a common strategy for social networks, it’s also not crazy. In recent years, we’ve seen players like YouTube in video and Spotify in podcasts begin to license and create more first-party content to accelerate their services.

Photo of Piet Terheyden
Piet Terheyden@piet

The other question to ask is, if a user wants to reactivate, how hard is it? At Uber, we had a staggering statistic where several million users were failing their password recovery per week—how do you make this much easier, and treat reactivation with the same seriousness as the sign-up process? While reactivation is typically not a concern for new products—they should focus on new users, since their count of lapsed users won’t be large—for products that have hit Escape Velocity, there will be a pool of many millions of users to draw upon. Reengaging them can become as big a growth lever as acquiring new users.