Joschua’s Notes

Search

SearchSearch

The Cold Start Problem

Jan 15, 2024

The Cold Start Problem

up:: Books

thumbnail|150

[!even-columns]

About

Reading

  • [status:: read]
  • [rating:: 3.5]
  • [added:: 2023-03-23]
  • [started:: 2023-04-13]
  • [read:: 2023-04-28]

Reflection/Review

As someone who isn’t very familiar with the startup world, this was a good introduction to the terminology and thinking. The author didn’t presume much prior knowledge but laid out new terms and explained concepts.

However, it was noticeable that it’s this author’s first book. At times, the book was quite repetitive, re-explaining a concept that it just explained a couple of paragraphs before. The writing was a bit clunky at times. In general, the book probably would have benefitted from being cut a bit shorter and having more concise points and illustrations.

While the stories generally kept me engaged, it was a bit of a slog to finish in the latter third of the book.

Since I started reading this book, right after reading Debt · The First 5000 Years, some of the stories of startup culture hit differently. Especially the chapter on flintstoning, essentially endorsing faking until you make it, was one I view critically. The methods described are deceptive at best, immoral and illegal at worst. I might be naive but I think companies should hold themselves to a higher standard.

In general, the idea of network effect is a valuable one. It is a good addition to the toolkit of mental models. The stories were interesting, even if the book could have been more concise in total.

Notes

  • network effect: the more users a product has, the better it gets
  • Allee curve: from ecology
    • One a population reaches a threshold it grows more rapidly. “Strength in numbers”
    • carrying Capacity: after a certain threshold, it stagnates. “Overcrowding”
  • In business terms
    • Allee effect → the Network effect
    • Allee threshold → Tipping Point
    • Carrying capacity → Saturation
  • Cold start problem: The hard things is building that first atomic network
    • this needs to be carefully curated
    • Tinder started by throwing parties at universities, requiring participants to show the downloaded app to the bouncer
    • LinkedIn wanted mid-level professionals, that’s why it started out being invite-only
    • The quality of the initial network matters. It needs to be tight enough for people to connect. That’s why a university by university expansion works.
    • Build one atomic network, then copy and paste.
    • “Come for the tool, stay for the network”. If the functionality is great, it can scaffold a smaller initial network.
    • Flintstoning: Doing the work of a network before having one
      • Reddit’s founders would post content all day long to make it seem like there is a flurry of activity
    • The “hard side” of a network is harder to acquire. In Uber that’s the rider, on YouTube the creators, in Dropbox the high value users. Focus on retaining and attracting, possibly subsidising, these users.
  • Phase 2 Escape Velocity: Scaling the network and generating profit
    • the network effect breaks down into three sub-effects: Acquisition network effect, Engagement network effect, Economic network effect.
      • Acquisition: use network to get new users
      • Engagement: the more users, the stickier
      • Economic: the more users the better finances
  • Phase 3: the ceiling
    • A way to break through the stagnation is to add more layers on. Adding fixed prices to auctions was controversial for eBay but lead to higher growth. So is carpooling and food delivery for Uber.
  • Phase 4: the moat
    • Cherry-picking: choosing valuable and poorly defended features from a competitor
    • “The Unbundling of Craigslist”: Indeed focused on Jobs, Airbnb on listings, Etsy on arts and crafts

Graph View

Backlinks

  • No backlinks found