Jump to ratings and reviews
Rate this book

The Cold Start Problem: How to Start and Scale Network Effects

Rate this book
A startup executive and investor draws on expertise developed at the premier venture capital firm Andreessen Horowitz and as an executive at Uber to address how tech’s most successful products have solved the dreaded "cold start problem”—by leveraging network effects to launch and scale toward billions of users.

Although software has become easier to build, launching and scaling new products and services remains difficult. Startups face daunting challenges entering the technology ecosystem, including stiff competition, copycats, and ineffective marketing channels. Teams launching new products must consider the advantages of “the network effect,” where a product or service’s value increases as more users engage with it. Apple, Google, Microsoft, and other tech giants utilize network effects, and most tech products incorporate them, whether they’re messaging apps, workplace collaboration tools, or marketplaces. Network effects provide a path for fledgling products to break through, attracting new users through viral growth and word of mouth.

Yet most entrepreneurs lack the vocabulary and context to describe them—much less understand the fundamental principles that drive the effect. What exactly are network effects? How do teams create and build them into their products? How do products compete in a market where every player has them? Andrew Chen draws on his experience and on interviews with the CEOs and founding teams of LinkedIn, Twitch, Zoom, Dropbox, Tinder, Uber, Airbnb, and Pinterest to offer unique insights in answering these questions. Chen also provides practical frameworks and principles that can be applied across products and industries. 

The Cold Start Problem reveals what makes winning networks thrive, why some startups fail to successfully scale, and, most crucially, why products that create and compete using the network effect are vitally important today.

395 pages, Kindle Edition

Published December 7, 2021

Loading interface...
Loading interface...

About the author

Andrew Chen

15 books57 followers

Ratings & Reviews

What do you think?
Rate this book

Friends & Following

Create a free account to discover what your friends think of this book!

Community Reviews

5 stars
1,239 (43%)
4 stars
1,065 (37%)
3 stars
470 (16%)
2 stars
74 (2%)
1 star
24 (<1%)
Displaying 1 - 30 of 285 reviews
Profile Image for Sebastian Gebski.
1,036 reviews996 followers
January 2, 2022
It's a book about building and scaling out a product, that is 100% focused on the product strategy. Forget engineering, building teams, working with backlogs, product-oriented culture/mindset - that (delivery) aspects are not on the plate. TBH there's absolutely nothing wrong about that - I really liked the crystal-clear focus of the book.

OK, what are the biggest pros?
1. it's written by the practitioner - a person who has gone through a scaling journey (when he worked for Uber)
2. I didn't catch any selfish motives here - no personal stories about family hardships, no emphasizing own accomplishments, no attempts to boost Uber grandiosity, etc. - that's a big pro
3. the author comes for his simple mental model for scaling product - he doesn't build any "legend' for that, he doesn't present it as a miraculous "framework" - it's just some mental frames and vocabulary to help you understand his message. I really liked it.
4. there are plenty of examples to illustrate all the concepts - taken from well-known, popular companies (like Uber, Airbnb, Reddit, or YouTube)
5. I really like the analysis of why certain things didn't catch on or have reached their plateaus (e.g. Google+, Usenet, Uber)

Is there anything I didn't like in the book? Well, there were at least a few situations when I had the appetite for more (details, numbers, deeper analysis), but well - every book has its capacity, right?

If you're interested in startups, successful product strategies, scaling out products - this book is a must-read for you. If you're interested in network effects, product engagement, so-called "hard-side" (of your user base) - definitely take a look. Don't get me wrong -> if you're into the topic this book won't provide you with a breakthrough, completely new and ultra-revealing theory: what it does is proper dissecting of the topic, naming the challenges, illustrating them with examples, and framing all that into a single mental model. IMHO it's still something totally worth paying for.
Profile Image for Rishabh Srivastava.
152 reviews186 followers
December 17, 2021
Solid read that talks about how to start and sustain network-based startups. Much of the framework was explained in the first 10% of the book, with the rest of the 90% giving concrete examples of the framework applied to practice.

The first part of the book is relevant for everyone who is interested in the space, while the rest of it is relevant mostly for people working as operators in a network-driven company.

My key takeaways were:

- Network effect = product gets more valuable with more users. In the dot-com boom, companies used Metcalfe's law (value of a network increases as a square of its number of nodes) as a way to justify valuations. This created FOMO. "Buy in early, and buy in soon, because the value of the network will soon explode in a fast growing startup". But this is an insufficient tool

- The Cold Start Theory is the author's framework for thinking through a network's growth. He postulates that there are 5 stages of a network, which are:

1. The Cold start problem: getting the right products and content on the same network at the same time. The problem to solve here is "how to get to the smallest possible network that can sustain and grow on its own". For this, need to have a dead simple (to use) product that still solves a sharp pain point in order to get initial users . The next big thing often looks like a toy that’s built for niche networks. Super important to start with a niche. Only then can you scale to other, larger, mainstream networks (by incorporating learnings from niche networks and improving products).
On the product side, it's important to focus on feedback loops here. What are the status, utility, and emotional feedback loops that you can build into your product?

2. The Tipping Point: what are the small networks you can start with to eventually get to a tipping point surging that network, and then into the larger ecosystem? For Tinder, this was colleges in the US and call centers in India. You can use many approaches to this – focused launch parties, an invite only approach, "come for the tool, stay for the network", Flintstoning, and others

3. Reaching escape velocity: consists of 3 things. The acquisition effect (can you get lower CAC with viral growth), the engagement effect (can you increase interactions in product as the network grows), and the economic effect (increases monetisation and conversions as the network grows). The 3 are interlinked

4. Dealing with issues as you hit the ceiling: CAC can spike due to market saturation and slowing viral growth. Other negative effects (like overcrowding and unpleasant interactions) can also manifest

5. The Moat: network based competition is a double edged sword, because other startups can take away your customer with the same tactics as you used to acquire them. What other moats can be built to work around this?
Profile Image for Jason.
Author 2 books8 followers
January 19, 2022
Introduction
A coworker recommended Andrew Chen’s new book, The Cold Start Problem, and I listened to it on Audible over the last few weeks. I felt there were enough gems to take notes and share in a review.

Analytics & Dashboards
The book starts with a helpful lesson in structuring analytics and reporting for any product with network effects. Product wide metrics might seem great, they are actually just vanity metrics. Knowing the total uber rides yesterday across the world is not helpful. Breaking it down by city, and potentially neighborhoods is actionable as you can know if you need more drivers in a local region, or scale up marketing to deliver more riders, or know about future trends like Friday night demand.

Metrics should not only be raw counts of booked trips, but instead different metrics that align with each side of the marketplace. For instance, if you are a rider you don’t want long wait times, and if you are a driver you don’t want long travel time to the pick up or cancellation rates. There are metrics to track for the experience that each side of the network is having.

Additionally, it is important to track the “zeros”, where the product delivers a very poor experience. For Uber that was no drivers available.

Atomic Networks
When starting a network based product it is important to begin with the smallest network possible, an atomic network. For Facebook and Tindr these were college campuses. For Uber it was a small area in San Francisco. The idea is to work out any issues early between both sides of the network before scaling, and even then doing so slowly.

The Hard Side
Any network product has a hard side of the network, the side that drives the engagement. For Instagram that was initially celebrities creating content driving usage for others to create accounts and (mostly) view other content.

Critiques of the Book
Although I learned a lot from this book I am continually frustrated by rich tech elite/VCs missing the impact that their products have on the people that use them. For Andrew to simply say that “All growth companies go through this” in relation to the hard side of the network revolting is a bit crass. He jokes about the protests he had to walk through to get the office and leaves off the ideas around prop 22 and the predatory role that Uber serves in this community. Additionally, at one point he talks about Travis leaving and “a new culture” instituted to put it mildly. His hero worship about working on Sundays until 10pm masks the toxic culture enshrined at Uber. Lastly, he proudly talks about the data that Uber uses about both its drivers and riders never noting anything about privacy.
Profile Image for Andy Rudd.
11 reviews1 follower
January 15, 2023
The saddest part of the Cold Start Problem is that Andrew Chen's effort to meet his page limit was not making the margins really big, but subjecting readers t0 the same four widely-known anecdotes over and over. Countless efforts to credentialize himself lead to unbelievable 20+ page descriptions of meetings at Uber that happened, can you believe it, after 10pm.

His discussion of network effects focuses on some strange choices - companies that cannot turn a profit, are flailing in the face of bigger players, or have been unquestionable net negatives for the world. It would be helpful if we were clear about his view of success. This is not an investigation of how to be profitable, or create value for end-users, but (as evidenced by Andreesen Horowitz printing and giving these books away like candy) possibly just how to generate big returns by any means necessary (including millennial subsidies with no real business value). My biggest takeaway is that anyone could be a venture capitalist as long as they're willing to pretend stuff like this is insightful.
Profile Image for Jano.
7 reviews3 followers
February 2, 2022
The book has some interesting anecdotes, but overall, I felt it is too anecdotal and relatively shallow.
Profile Image for Matej yangwao.
169 reviews10 followers
March 22, 2022
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.
This entire review has been hidden because of spoilers.
Profile Image for Walter Ullon.
265 reviews123 followers
February 26, 2023
I am pretty sure this book will one day either be required reading, or an entire course itself at some of the top business schools in the world. It is that good.

If you work on "networked" products, or are trying to implement network elements into a product, then you should read this before you go any further. Even if the subject itself is of little interest to you, the case studies (of which there are many and really well presented) are illuminating, and entertaining on their own.

Early on, the author explains that the classic view of network effects, as expounded by fans of Metcalfe's Law (the value/utility of a network is proportional to the square of its members) - is simply wrong, and that you must pay attention to the composition of the network and not just the raw numbers.

The rest of the book proceeds to explain why this is the case, and to teach you how to think about building and scaling network effects following clear, and effective exposition. He explains the theory, presents case studies on both the winner and loser side, then gives some advice.

Andrew Chen knows his stuff and it shows. There were chapters where my understanding of network effects was completely turned on its head and made me realize how fascinating this area can be.

I highly recommend that you read his blog and some of the writings of his fellow partners at a16Z as well. In a lot of ways, that content amplifies each chapter in the book.

It will for sure remain a reference for many future projects. Highest possible recommendation!
Profile Image for Ashwin Purushottam.
21 reviews5 followers
April 27, 2022
Chen bridges the gap between theory and execution with concrete examples from modern companies - along with the premise behind why it worked.

The first part of the book was particularly relevant to me as a founder of an early stage venture-funded startup in South East Asia. I'm currently focusing on getting those first 1000 customers. Pivoting and failing constantly is unavoidable, but finding a structured way to evaluate new launch methods is difficult. Chen provides exactly this - what to try, who does it work for, and why it should be done. His bite-sized chapters make the material easier to absorb too.

Later, he highlights the steps required for focusing on the smallest possible interwoven "Atomic network" which can sustain itself - this has helped our team focus on being just one thing for one type of demographic. After, he then you through the steps involved in scaling and reaching Malcom Gladwell's "Tipping point". Additionally, Chen shows how to understand which side of your business' network is harder to get started.

Two reasons I gave it a 4 vs 5:
- Chen's approach is based on his experience at Uber. The Softbank-fueled days are over where one can ignore unit economics and burn indefinitely, and some strategies in this book are geared towards doing just that. Having worked for a startup wholly owned by Uber myself with multiple ex Uber folk - their approach is to directly discount in order to push traditional sustainable competitors out. He repeatedly talks about Uber's culture of innovation resulting in the bullish "war room", but also how he got bored as Uber tapered down towards a more sustainable angle. You can see this approach throughout multiple chapter and also in Chen's real life investment strategies with recent startups he has backed at Horowitz, and this approach should be taken with an overall pinch of salt.

- Also, a great read for anyone at an early or even late stage start up, but I doubt it has value for those in big corporate industries as I found his few chapters on protecting larger moats a bit vague.
303 reviews217 followers
September 2, 2022
The first reading of this book was an ultra fast one. I had to make a quick decision on whether i want to participate in a book publishing project. Based on it i decided to do it and became a part of the team that published this book in Polish. Now i was able to dive into it in my own pace with a version that we have published. It was awesome. It's a very deep dive (400 pages) into network effects that is full of actionable insights based on real life examples and authors own experience from being a VC or growth lead responsible for the drivers side of things in uber.

If you are in consumer products, B2B products and especially marketplaces this is a must read.

PS. If you want to buy the polish version we have published feel free to do so at coldstart.pl
Profile Image for Cameron.
164 reviews8 followers
January 8, 2023
An excellent read for how to start a networked product, and probably also a helpful selling point for why networked products are so successful and the reasons to create one.
Profile Image for Anu.
391 reviews64 followers
February 6, 2022
Andrew Chen is a prolific tech figure, whose writings on growth and network effects are very popular. He brings it all together in this book with some intriguing and fascinating stories about tech cos in their early days. Great read for anyone in tech.
Profile Image for Dmitry.
896 reviews73 followers
November 18, 2023
(The English review is placed beneath the Russian one)

Российское издание: От одного пользователя до миллиона. Как успешные бренды и продукты наращивают аудиторию. Издательство: Манн, Иванов и Фербер, 2023 г.

Книга мне напомнила другую похожу книгу – «Бизнес на подписке. Почему будущее за подписной моделью и как вам ее внедрить». А напомнила она мне эту книгу потому, что обе книги являются сборником историй, в которых авторы коротко описывают как та или иная компания, используя конкретную бизнес-модель, построила успешный (прибыльный) бизнес. И это всё, что есть в этих двух книгах. Такой тип деловой литературы не является редким, однако чаще всего и лучше всего, если такой текст встречается в профильных журналах. Всё же книги должны предлагать читателю знания, а такой тип деловой литературы никакие знания предложить не может. Давайте подумаем, как информация об истории успеха компании Uber способна помочь создать совершенно новый и непохожий на Uber бизнес? Никак. Вы не сможете, полностью скопировав опыт Uber создать, условно говоря, второй Uber. Да и вообще не сможете создать прибыльный бизнес. А почему? Потому что каждый бизнес уникален. Это означает, что точное копирование одного успешного бизнеса не способно сделать другой, таким же успешным. Именно по этой причине все такие истории успеха и вся подобная бизнес-литература, в основе своей, в лучшем случаи является развлекательной, а в худшем – бессмысленной или даже вредной.

Меня поразили слова автора, суть который сводится к тому, что чтобы построить эффективный бизнес, основанный на сетевом эффекте, компании нужно предложить нечто такое потребителям, что они бы привели своих друзей. В самом начале книги автор пишет, что в таком бизнесе важным условием успеха является успешное завлечение потребителей. Это верно. Проблема лишь в том, как это сделать? Это настолько фундаментальная проблема, что, не решив её, всё остальное не имеет смысла. Более того, дело не в возможности компании суметь привлечь людей и их друзей, ибо для успеха «наличие друзей» вовсе не является основным фактором также как наличие небольшого числа пользователей вовсе не говорит о потенц��альном успехе (малое количество пользователей может таким и остаться, что является приговором для компании). Важным является умение дать потребителям то, в чём они нуждаются или будут нуждаться, когда познакомятся с товаром/услугой (т.е. они могут просто не знать, что нуждаются в этом). Компания Facebook стала успешной именно потому, что предложила такой продукт людям, который их и устроил и в котором они, как оказалось, нуждались. Так вот, читая все эти истории, невозможно определить, в чём же потребители могут нуждаться и в чём они определённо нуждаются (ведь если это очевидно, то определённо там уже присутствует некая компания, конкурент, а значит, вы уже не являетесь первопроходцем).

Да, я согласен с тем, что компания может не иметь достаточных мощностей, чтобы обслужить нахлынувший поток клиентов (к примеру, не справятся сервера с такой нагрузкой). Однако эта проблема свойственная всем�� бизнесу, а не только той бизнес модели, которую описывает автор. И да, проблема эта является менее важной, нежели нахождение конкурентной возможности.

Очень часто авторы бизнес-литературы приводят в пример Википедию, однако для меня всегда это выглядело очень странным, ведь по сути Википедия является единственным примером подобного проекта. Так что Википедию я бы рассматривал как исключение и не использовал её в качестве иллюстрации успешного бизнеса.

Примеры с Tinder, Zoom и LinkedIn являются более релевантными, но и они являются просто очередным ��ересказом истории успеха. Все эти проекты имеют уникальные предложения или свойства, благодаря которым они стали популярными. Другими словами, именно наличие уникальных свойств сделало их прибыльными и успешными.

Так что самой главной проблемой является самое начало книги, где автор говорит, что компания должна как-то заинтересовать потребителей. Ну, да, это суть любого бизнеса и одновременно это главная проблема любого бизнеса – увлечь (завлечь) потребителей своими товарами или услугами. Вот только как это сделать? Сетевая модель не обладает никаким магическим способом убеждения людей приобретать товар/услугу. Все примеры, что приводит автор, они по-своему уникальны. Ну а то, что их бизнес-модель основана на сетевой эффекте, об этом знает любой предприниматель, менеджер или маркетолог.

The book reminded me of another similar book - "Subscribed: Why the Subscription Model Will Be Your Company's Future - and What to Do About It." It reminded me of this book because both books are a collection of stories in which the authors briefly describe how a particular company, using a particular business model, has built a successful (profitable) business. And that's all there is to these two books. This type of business literature is not rare, but most often and best of all, if such text is found in specialized magazines. Books should offer the reader knowledge, but this type of business literature cannot offer any knowledge. Let's think about it: how can information about Uber's success story help create a new and different (from Uber) business? No way. You will not be able to create, let's say, a second Uber by copying Uber's experience. Nor will you be able to create a profitable business at all. Why is that? Because every business is unique. It means that exactly copying one successful business cannot make another one as successful. It is the reason why all such success stories and such business literature are, at best, entertaining and, at worst, meaningless or even harmful.

I was struck by the author's words, the gist of which is that to build an effective business based on the network effect, a company needs to offer something to consumers so that they would bring their friends. At the very beginning of the book, the author writes that in such a business, an important condition for success is to successfully entice consumers. It is true. The only problem is, how do you do it? This is such a fundamental problem that, without solving it, everything else is meaningless. Moreover, it is not about the company's ability to attract people and their friends because "having friends" is not the main factor for success, just as having a small number of users does not indicate potential success (a small number of users can stay that way, which is a doomsday for the company). What is important is the ability to provide consumers with something they need or will need when they become familiar with the product/service (i.e., they may simply not know they need it). Facebook became successful precisely because it offered a product to people that they were both comfortable with and appeared to need. So, reading all these stories, it is impossible to determine what consumers might need and what they definitely need (because, if it's so obvious, there's certainly some company already out there, a competitor, which means you're no longer a first mover).

Yes, I agree that the company may not have enough capacity to handle the influx of customers (for example, the servers can't handle the load). However, this problem is inherent to all businesses, not just the business model the author describes. And yes, this problem is less important than finding a competitive opportunity.

Very often authors of business literature cite Wikipedia as an example, but to me, it always looked strange because Wikipedia is the only example of such a project. So, I would consider Wikipedia as an exception and not use it as an illustration of a successful business.

The examples with Tinder, Zoom, and LinkedIn are more relevant, but they, too, are just another retelling of a success story. All of these projects have unique offerings or properties that make them popular. In other words, it was the presence of unique properties that made them profitable and successful.

So, the most important problem is at the very beginning of the book, where the author says that the company must somehow interest consumers. Well, yes, this is the essence of any business, and at the same time, it is the main problem of any business - to attract (entice) consumers. But how to do it? The network model does not have any magical way of convincing people to buy a product/service. All the examples that the author gives are unique in their own way. That their business model is based on the network effect is something any entrepreneur, manager, or marketer knows.
Profile Image for Rafael Ramirez.
120 reviews15 followers
April 12, 2022
Uno de los aspectos de mayor impacto de la economía digital es que facilita la conexión entre diferentes agentes económicos para crear, en conjunto, mucho más valor para los consumidores que actuando de manera aislada. Esto es lo que se conoce como un modelo de negocio de ecosistemas mediados por plataformas digitales; muy distinto del modelo de negocios tradicional en el que una empresa compra insumos y los transforma para crear bienes y servicios que luego vende a sus clientes.

A diferencia del modelo tradicional en el que la empresa tiene algo que vender a sus clientes desde su inicio, una plataforma digital debe convencer a proveedores a sumarse a su ecosistema basándose en la promesa de que, eventualmente, los clientes vendrán. Este es uno de los mayores retos a superar para una plataforma de nueva creación: el problema del huevo y la gallina, o como lo llama el autor de este libro, el problema del arranque en frío.

Andrew Chen tiene una amplia experiencia con start-ups que pretenden seguir este modelo de negocio, habiendo trabajado para Uber y posteriormente como inversionista en uno de los fondos de capital de riesgo de mayor prestigio en Silicon Valley, Andreessen Horowitz. Basado en esta experiencia, comparte diversas ideas para que los emprendedores que quieran crear plataformas puedan superar este problema y logren atraer los agentes necesarios a su ecosistema para lograr arrancar y posteriormente crecer de manera acelerada generando efectos de red.

Lectura obligada para cualquier emprendedor que quiera lanzar una plataforma o para directivos de empresas tradicionales que quieran dar el salto y transformarse en plataformas digitales orquestadoras de ecosistemas.
Profile Image for Ryan.
1,189 reviews169 followers
December 19, 2021
This is an excellent book for product managers and entrepreneurs building new businesses. It takes the high-level "this product has network effects" or "viral mechanic" and explores in a more rigorous and scientific way. The focus is on how a new product can get initial users and how existing products can continue to grow, what the underlying mechanics are, how to recover from small failures, etc, using real-world examples.

Uber is used as one of the main examples, as the author worked there (and damn, I wish I'd worked at Uber during the early days, even for $0 -- seems like an amazing team and place to work.). The expansion of Airbnb into Europe and the fight against Rocket Internet's now-forgotten Wimdu was an even better example, late in the book, of the positive side of good solutions to cold start and the weaknesses of bad solutions.

This book is also an excellent advertisement for Andrew Chen as a VC and for a16z as a firm. I assume the book will make hundreds of thousands of dollars or low-digit millions directly, but probably hundreds of millions of dollars in lifetime value for the author; interesting asymmetry there.

I would probably still say Innovator's Dilemma, Crossing the Chasm, Zero to One are more important, but along with The Hard Thing About Hard Things, this is a top-5 or at least top-10 startup/business book.
Profile Image for Deepak.
109 reviews20 followers
December 1, 2022
Absolutely brilliant text on network effects, giving the phenomenon a scientific language to work with.

Notes from the book (I know, a very long list):

Network effects is about connecting people, and as the number of people using the product grows, so does the potential to monetize and the defensibility of the product.

There are two tipping points in network effects - a lower end below which the effect will go to zero, and an upper end beyond which the network is overcrowded.

Stages of network effects
1. Cold start problem
2. Tipping point
3. Escape velocity
4. Hitting the ceiling
5. The moat

Cold start problem
- atomic network - the smallest possible network that makes the product relevant for a user. Zoom -2. Airbnb - 300 listings and 100 reviewed listings. Slack - 3 people or 2000 exchanged messages.
- Start with a simple product (with a dead simple value proposition), and do whatever it takes (including growth hacking) to reach the critical mass of an atomic network
- Smaller the size of the desired atomic network, easier is it build one. But also more difficult to build a moat around it.
- There is a hard side of the network - the select group of highly engaged users that create disproportionate value for the rest of the network. Eg content creators on YouTube, attractive women in dating apps or editors on wikipedia. 1-10-100 rule : 1 % creators, 10% synthesizers and 100% consumers
- Motivations for the hard side are different : eg in social apps, it could be communication, social proof and self expression (going up in the hierarchy).
- Hard side > demand > supply > supply > demand ..
- A killer product in the network world is usually a very simple one - one that is easily understood and easy to use by a large consumer base. And typically also free / on a freemium model, such that there is lesser friction for the users
- In order to solve cold start, one needs to start eliminating zeroes (when the customer did not get the desired experience eg no drivers around in a ride hailing app) and start creating moments instead. And this cold start problem needs to be solved over and over for different atomic networks.

The tipping point
- Invite only strategy not only creates FOMO, but also helps build a high quality dense network, and at times also gives the team a chance to build for progressively bigger audiences
- Come for the tool, stay for the network (eg Instagram photo editing, google docs)
- To build network effects upfront, it might make sense to put in large investments upfront (eg Uber heavily incentivising drivers in the initial days) even at the cost of unprofitability in the short term
- Flintstoning - replacing the actual code / product feature with a manual effort to overcome the cold start. Eg in food delivery, getting customers to place orders, picking up yourself and delivering, without even signing up the restaurant.
- In the beginning of the journey, try things that wont scale (eg Uber Icecream)

Escape velocity
- Network effects broken into 3 elements - acquisition, engagement, economics (monetization)
- Acquisition : viral factor - no. Of new users that can be brought in by the existing user
- Engagement : retention curves that look like a smile (initial churn which is natural, but then churn reverses with more stickiness) are rare and great targets for investment. Engagement increases with new use cases, reactivating dormant users. Reactivating can sometimes be as big as acquiring new users.

Ceiling
- Reason 1 - market saturation (absolute TAM that could be brought onto the network comes down), plus network saturation (each incremental add to the network does not have the same incremental value to existing users, as earlier)
- Solve for adjacent users (adjacent user theory) eg lighter version of the app for those that have low end phones / penetrating new geos via new language options
- Solve for new use cases for existing users eg buy now option on ebay, in addition to the auction based approach
- Or acquire start ups that are hitting escape velocity and can be integrated into the network

- reason 2 - the law of shitty clickthroughs - consumers acclimate to ads / brands / marketing messages after a period of time, and start tuning them out

- reason 3 - sometimes the hard side of the network revolts and can take down the rest of the network. Eg Uber power drivers for better pay / benefits.
- Further, while the intent is to organize the hard side of the network, it might lead to issues eg concentration of power on the hard side

- reason 4 - rise of spamming, unwanted content (eternal september). Plus “context collapse” (too many contexts for a content creator to deal with. Eg would you want to share something if your boss was on the same network). “Share with close friends” on Insta is an example to mitigate this

Building a moat
- a global network presents a stronger moat (eg airbnb) than local (eg Uber). Or a slack (intra company) vs zoom (inter company)
- There is a winner-takes-all possibility with network effects. Eg doordash focus on suburbs and college towns had led it to have 2x the market share of Uber Eats
- Cherry picking from a bigger network - eg airbnb from Craigslist
- Sometimes going big bang breadth wise may not work - it could lead to multiple weak networks even as the overall numbers look big and attractive - eg in the case of google plus. It is better to build atomic networks first and then do the big bang “top up”
- Cracking the hard side of the network - cracking the top drivers (Uber) that were also dual- apping and getting them over with incentives. Double whammy for competition - they would lose these drivers and further experiencing surge
January 24, 2022
I had high expectations of that book, let me admit it. Well-known VC partner, sound experience at Uber... In fact, the whole book could be packed into just 2-3 pages article. It feels like the author was an observer but not an active participant in the things he described or even happened to observe them by luck; in most cases, it is full of verbiage, as if the author tried to create a new vocabulary out of (or for?) just one simple idea. Unreasonable namedropping here and there is also quite annoying. 2/5, poor book about simple but powerful idea though.
Profile Image for Mark.
73 reviews1 follower
November 25, 2022
Andrew Chen has clearly "walked the walk," through his experience at early-Uber and otherwise, and the subject matter here is critical to any network-based startup. That said, the book is more "war stories" as examples than it is a how-to. Missing was a step after each war story that distilled down the lesson-to-be-learned and shared more directly how to execute like the example or how to avoid the misstep of the example. The success examples were clear, but without directives on execution, they felt more like stories of startups that "got lucky" with their strategies.
Profile Image for Vikrama Dhiman.
159 reviews97 followers
January 26, 2022
Last 3 units make it good read

The Escape Velocity, The Ceiling and The Moats save the book from almost there to good. If you e followed the latest in tech, the first units will be a breeze - there is some spark but only some.
Last 3 units though come into their own. The Ceiling esp is a really good collection.
It's an exhaustive read but still leaves much in theory and takeaways uncovered. Look forward to Edition 2.
57 reviews2 followers
December 25, 2021
Amazing book

A deep operator (scaling through Uber) describes the properties and requirements of companies delivering in network markets. Competition is brutal and constant but the rewards are enormous and global. His examples are detailed and actionable in was highlighting almost every page on my kindle with things I want to use with my colleagues.
Profile Image for Denis Vasilev.
681 reviews97 followers
December 31, 2021
Проекты с сетевым эффектом и сопутствующей проблемой начальной аудитории. Интересная тема, но не так много полезных рецептов и обобщений, как я ожидал от такого знающего автора. Слишком много отвлеченных рассказов про Убер и прочих
Profile Image for Kashish Sharma.
25 reviews2 followers
January 24, 2022
Highly recommend this book to every operator who wishes to learn about network effects and dive deeper. Definitely recommended.
Profile Image for George.
11 reviews
April 21, 2022
Got about 30% through the book before stopping. All fluff with no real substance imo. It can be summarised in a few lines that describe "network effects".
Profile Image for Jacek Bartczak.
196 reviews64 followers
August 6, 2022
I was skeptical about a book on network effects. I was afraid that it would be a book "about Growth hacks for Growth hacks". - what to do to increase the effectiveness of an advertising campaign by 2% , how to more cleverly invite more app users.

The network effects are much broader. From a certain point, the network becomes a market, and its members become companies employing several hundred people. E.g. the network used to be Android / Google Play, and the members of this network became companies, e.g. Square, AirBnB. Some members of this network.... also have their own networks. E.g. AirBnB is a network of landlords and tenants.

Even if you do not work in a company based on network effects:
- PART of the tips will help you acquire customers, better manage the customer path, etc...,
- the book will expand your knowledge of how than business works. Network effects-based businesses are all around us. Internet and mobile networks are also networks - only that they are huge and very mature.

If you are a software house that codes products for startups or an agency that does campaigns for startups, this book will help you bring "consulting value" to your collaborations. Or at least better understand your target audience.

There is a lot about how to start building a network, as well as how to get through the most difficult phases. It also shows why some networks died (UseNet) and others didn't (Uber). Lots of tips on where to get your first customers from, and what to do to make it easier for customers to recommend your product.

Great book that shows what comes out of the merger:
- a veteran practitioner. Andrew Chen has been associated with Uber, Substack, Andreesen Horowitz,
- methodology. The big concept of "network effects" was broken down into steps, pages, factors, etc. So that anyone interested can have a framework.
Profile Image for AV.
88 reviews9 followers
April 16, 2023
If you’re looking to take away insights from this book and apply to real world scenarios, you need to be either a founder of a consumer business or someone responsible for the growth of a consumer business, typically, product and marketing folks.

But not just any consumer businesses. The ones that are built on or have networks effects at its core. Think Uber, Tinder, Airbnb, Reddit, Instagram, LinkedIn, Dropbox, Slack, YouTube, etc.

The book outlines extremely insightful methods of the different stages of the growth problem, taking multiple leaves out from how the aforementioned businesses grew. And I truly believe that these methods and tactics can be applied again and again, for similar types of businesses still.

The book is divided into the following 6 parts,

1. Network Effects - What is it
2. The Cold Start Problem - How to get started with a small network, focusing first on the “hard side”, i.e., the supply side of the network.
3. The Tipping Point - Scaling the initial network and using it to grow the user base to a point of repeatability where growth becomes automatic
4. Escape Velocity - Maintaining the growth by leveraging its existing networks as well as acquiring new users but focusing more on revenues and profitability
5. The Ceiling - How after years of consistent growth, the product hits a ceiling post which, companies need to innovate and make hard decisions to fight the slowdown and continue on the growth trajectory
6. The Moat - The competition which results in your victory, where you finally reach a point where due to the size of your network effects, innovation, marketing strategy, or something else, you become extremely hard to beat, essentially establishing a monopoly.

I liked how Andrew Chen went through each stage in a lot of details and including both the success and the failure stories. Although it'd have been great to include a few more failure stories than just Usenet and Google+. Surprisingly, there was no mention of the deteriorating state of Clubhouse, which would have been interesting to read about from an insider's perspective.

That and the fact that I felt the book is lengthier than necessary, by around 20%, is why I gave it 4 stars. The concepts and ideas are all great and very useful but due to the repetitiveness of some of them and some unnecessary details, the book becomes more than what’s needed. For example, the examples of Tinder, Uber, Dropbox, etc. have been repeated multiple times, which might be contextual citing the different categories of context, but I don't think it was necessary.
Profile Image for nalini singh.
30 reviews4 followers
August 25, 2023
As I build a network based product (localhomeydotcom) , I couldn't have found a better book to avoid common pitfalls as I get to the first 500 users. There are very few people better placed than Andrew Chen to talk authoritatively about the topic in such comprehensive detail given his time at Uber, his investments and his research about the stuff he didn't personally experience is commendable.
He keeps things interesting with case studies and real quotes from real heroes behind our mobile screen addiction ;) I wish more VCs would write books! I don't 100% agree with all the approaches examples in this book took to solve the cold start problem, but it gives a great snapshot of behind-the-scenes VC and VC backed start up employee's perspective of their start ups burning cash so fast ! Would i replicate the strategies in this book- some, but definitely not the stuff Uber did. I just hope readers don't take away that these are the ONLY ways to solve a cold start problem. I am sure the stories of bootstrapped or just angel funded start ups that survived might have dozens of other strategies that worked. I hope there's a part 2 of this book in the works where we get to read those stories.
All being said, my favorite business book this year- well structured, dense with information still fun to read, a rare combination. That's why 5/5 for the book despite its "biased Uber ex-employee" perspective. 10/5 for Andrew's commitment to finding time to write it despite his busy day job.
Profile Image for Matthew.
93 reviews
November 5, 2023
A super readable, well-written book about scaling network-driven companies. Chen interspersed principles (e.g., to build “atomic” networks, start small and grow sustainably; or solve for the hard-side first when building a multi-sided marketplace).

Perhaps consistent with Silicon Valley’s, “move fast and break things” ethos, I thought Chen may have glossed over some "gray areas" in corporate history. Specifically, he prided LinkedIn on early user acquisition with an invite-only approach, but made no mention of its predatory strategy of sending referral emails to entire contact lists without permission (for which it lost a class action lawsuit)

Another case study was Uber, harking to Chen’s experience with the company. Did it grow quickly and was it disruptive? Sure, but I thought the book lacked an adequate discussion of well-documented cultural issues at the company that ultimately caused co-founder/CEO Travis Kalanick to be exited from the company.

Finally, I’d like to see an index in a future iteration of this book.
Profile Image for Susan.
171 reviews
June 20, 2022
A comprehensive book on the hardest journey of any network Startup: how to go from zero to one when there are no users. The stories of many now household names are discussed, like Uber, Tinder, Wikipedia, Zoom, Airbnb. Competition in the tech market is unique in the sense that product features are relatively easy to copy, since it's just software, but what's difficult is the organic network a product relies on. The author also introduces 5 stages of the growth framework, including the coldstart, tipping point, escape velocity, hitting the ceiling, and the moat.

I think the approaches discussed in the book not only apply to companies, but also new features and projects that are facing adoption issues. One downside of the book is that it felt pretty long and some parts are repetitive, like the Uber example was mentioned multiple times. It tried to illustrate each point with examples of a prominent network product, which is a good concept but could be more concise.
Profile Image for Gepser Hoil.
10 reviews3 followers
January 23, 2022
A must-read for entrepreneurs and people creating products and/or communities.

Each chapter had unique insights with specific examples, talking about what worked and what didn't work.

It's a long book, so a summary won't be enough, but if you want to know how to create a successful product or service, or how to compete (and win) with another product doing the same (no matter how big), what should be your focus when working with a product (individual network metrics vs global metrics).

Of course, it also explains what are network effects? How do I know if my product has it? How do I know if other products don't have it? Can I add it later? How can I compete with a network effect product?

It also covers some topics like the tipping point, anti network effects, smallest possible network, reactivation, retention, keeping the network healthy, network vs network, and more.
11 reviews
August 22, 2022
This was indeed a pretty good read.
The author has in-depth knowledge about the subject, particularly how network effects helped Uber become the main ride-sharing app. The book explains detailed concepts and ideas for measuring networked systems.
The book divides itself into 4 parts
1. Cold start
2. Tipping point
3. Escape velocity
4. The ceiling
5. The moat
All these parts are well explained with ample examples.
The most insightful takeaway from the book is about scaling the hard side of a network is often the game changer for most products.
Insights about how Dropbox combined features and network and then partitioned the network to sell to only power users is a nice little insight into monetizing a product.
Overall this book presents enough counter-intuitive ideas and simple metrics to track the progress to be worthy of a read and re-read.
Profile Image for Selena S.
69 reviews
January 9, 2024
There were some really interesting case studies here, such as the birth of credit cards and coupons, and these were by far my favorite part of the book. There were also some helpful high level concepts such as starting with the smallest possible network, and the consumerization of b2b products.

However there was a lot of the whole, reinventing-the-wheel-and-claiming-it-as-yours. For example, optimizing an “engagement loop” is the same as optimizing a plain ol’ user journey. I guess Chen needs a fancy name to fit into his chapter structure?

There was also a weird take that things like labor issues a la Uber are inevitable for growth and that you should embrace them.
Displaying 1 - 30 of 285 reviews

Can't find what you're looking for?

Get help and learn more about the design.