How to get and evaluate startup ideas

The following are my notes from YC startup school session on getting and evaluating ideas to build a startup.

It’s hard to predict whether an idea would succeed or not. But there are some ideas are more likely to succeed than others. This is a guide to find such ideas.

A good resource: How to Get Startup Ideas – Paul Graham

Four most common mistakes with startup ideas

Not solving a real problem

The idea would be a nice to have, but it wouldn’t be solving a real problem
Or, it’s solving a problem that people don’t care about
This is a Solution In Search of a Problem (SISP)
Example: AI is cool, what could I apply it to?
Instead, founders should fall in love with a problem and try to solve it rather than falling in love with a solution

Getting stuck on a tarpit idea

Tarpit ideas are those ideas that have been around forever
These are widespread problems that lots of potential founders encounter
Seems like it could be easily solvable with a startup
Apparently, there’s a structural reason why it’s actually not solvable
Example: An app to make casual weekend plans with your friends
Tarpit ideas are not necessarily impossible, but these are common ideas that are surprisingly difficult to build
Instead, start with Googling the idea, find out who has worked on it, and talk to them to get a perspective

Jumping into an idea without evaluating if it’s a good one

Think of your idea as a “good starting point”
It’s like writing a research paper, you have an assumption and you validate it as you go on writing the paper
10 questions to ask before proceeding with an idea:
- Do you have founder-market fit?
  - Probably the most important criteria
  - Are you the right team to work on this idea?
  - Instead of picking a good idea, pick a good idea for your team
- How big is the market for this idea? What would it look like in the future?
  - Check if the markets are big now
  - Check if the markets are small yet growing rapidly
  - Note that a market typically needs to be growing 50% or more every year to be growing “explosively”
  - Good read – How to Model Total Addressable Market (Template Included) | Visible.vc
- How acute is this problem?
  - See if the problem is real and people actually care about it
  - If the problem is real, the alternative to your solution should be close to nothing
- Do you have entrenched competition? If so, do you have a novel insight?
  - If there is already existing competition, do you have a new novel insight that they don’t know about?
- Do you yourself want this? Do you know people who want this?
  - If it’s a no then probably nobody wants this
- Did this only recently become possible or necessary?
  - Due to regulation change or change in circumstances such as COVID-19
  - If so, then it’s a good idea to try out
- Are there good proxies for this business?
  - Proxies are successful companies that do something similar, but not exactly the same thing
  - Like food delivery companies are the world
  - Looking at DoorDash, Zomato was started in India
- Is this an idea you’d want to work on for years?
  - Often, the passion to work on an idea develops overtime
  - A good video: Should You Follow Your Passion?
- Can your solution scale?
  - Pure software businesses – great
  - Pure services businesses – bad
- Is this in a good idea space?
  - An idea space is a class of closely related startup ideas
  - Example: Idea – Credit card for truck drivers. Idea space – Trucking industry
  - Different idea spaces have widely different hit rates
  - You would want to pick a good idea space which has a good hit rate and a founder-market fit
  - Example: Fivetran a company that started with data analysis pivoted to another idea in their idea space – an easy to use hosted ETL service

Three things that make startup ideas seem bad, but actually make them good

Hard to get started. Example: Stripe
Boring space. Example: Gusto – payroll software
- Side-note – Boring ideas have much higher hit rate than “fun” ideas as most people wouldn’t choose them
Existing competitors. Example: Dropbox
- Innovation in the space of ease of usage
Good read: Schlep Blindness by Paul Graham

How to come up with startup ideas

The best startup ideas are noticed organically
Ways to put yourself in a position to have organic startup ideas:
- Become an expert on something valuable
- Work at a startup
- Building things you find interesting

Here are 6 recipes for generating startup ideas:

Start with what your team is especially good at
- This would have the perfect founder-market fit
Start with a problem you’re personally encountered, ideally one you were in an unusual position to see
- Vanta founder Christina has to go through a security audit to realise that they were painful
- Think of every job + internship + life experiences you’ve had
- What problems did you come across? What did you learn that other people don’t know of?
Things you personally wish existed
- This can be dangerous and might become a tarpit idea.
Look for things that have changed in the world recently
- Like how COVID-19 make vaccines and masks very important boosting their sales.
Go to people and ask about the problems they have
- Talk to potential customers and also founders in that idea space to understand the problems and their solutions
Look for big industries that seem broken

Remember, the best way to find out if your idea is good is to launch it. Also, 25% of successful YC companies have pivoted from their initial ideas.

There are more chances of finding a good idea if you look in your field of expertise.