September 1, 2018

On Startup Ideas

I have been reading lately about what entrepreneurs go through before they come up with good startup ideas. Only few with amazing tech skills, such as Mark Zuckerberg, turn personal side projects into enterprises. Many of us neither have side projects, nor time to spend on them. Even those that have both of these resources often fail.

Let's try to go through some success stories to see whether we could extract a common theme from all of them to answer the question of how to come up with good startup ideas.

Ilja Laurs, the founder of GetJar, in his book "Verslas naujai" (eng. "The Modern Way of Doing Business") explains that he had been developing websites, mobile apps and games for several years before coming up with the idea for GetJar. The only way then to distribute mobile apps to clients was via mobile operators, which was a long, tedious process involving tons of paperwork. He decided to simplify it by developing an app store. The app store made it much easier to upload and download mobile applications.

Collison Brothers, founders of the payment gateway Stripe ($1.5 billion revenue in 2017), did pretty much the same thing.

In early 2010 John and Patrick began working on Stripe together. At the time Patrick was working on several side projects and they debated why it was so difficult to accept payments on the web. They sought to solve the problem and see if it was possible to make it simple - really simple. The next 6-months they played with it, showed it to friends, and saw how people interacted with it, iterating along the way.

Note: these Patrick's side projects must have probably been solutions for his clients, as at that time he was also a freelance developer.

Paul Graham, the so-called father of startups, also describes this phenomenon. He claims that there are two types of startup ideas, the ones I have described above being organic ones that satisfy some market needs. It's a good read, do check it out.

We can generalize all of these stories into a set of heuristics:

  • Work on some products for an extended period of time. Be genuinely interested about their users, the market and problems they solve. This would allow you to gather domain knowledge.
  • Identify tedious processes clients are paying for.
  • Turn them into products (e.g. SAAS apps). Sell to same, and potentially new clients.

Some possible extensions:

  • Oversee the whole product development chain, talk as much with clients as possible. Discussions with end users would allow you to identify those tedious processes.
    • E.g. Work for another startup, start a consultancy company, a web or app development agency. Something that's small.
  • Try to focus on one specific domain (e.g. high frequency trading, e-commerce, holiday rentals). It's a good idea to know one or several domains well instead of having a shallow view of multiple ones.

Share you stories - it would interesting to hear whether any of you have followed such "algorithm", and whether it has worked out for you.

  • LinkedIn
  • Tumblr
  • Reddit
  • Google+
  • Pinterest
  • Pocket