HOW TO GET AND EVALUATE STARTUP IDEAS
YOUTUBE SUMMARY : YCOMBINATOR
CHAPTER INDEX
HOW TO GET AND EVALUATE STARTUP IDEAS
A PRACTICAL FRAMEWORK FOR FOUNDERS
A startup idea does not need to be perfect.
It needs to solve a real problem, fit the founders, and create a strong starting point for learning.
Execution will reshape the idea. Your first job is to begin from a position where the odds are in your favor.
AVOID BAD STARTING POINTS
CHAPTER 1
START WITH THE PROBLEM
TECHNOLOGY IS NOT THE STARTING POINT
Do not begin with: “AI is powerful. Where can I use it?”
That often creates a solution searching for a problem.
Begin with a specific problem people already care about. Then choose the technology that solves it better, faster, or more cheaply.
TEST WHETHER THE PAIN IS REAL
A PLAUSIBLE PROBLEM MAY STILL BE WEAK
Founders can explain many problems convincingly.
The real test is user behavior. Are people actively trying to solve it? Are they spending money, time, or effort on alternatives?
A problem that sounds logical but creates no urgency rarely becomes a strong business.
AVOID TAR-PIT IDEAS
SOME SIMPLE IDEAS HIDE STRUCTURAL DIFFICULTY
Tar-pit ideas attract founders because the problem is common and the product seems easy to imagine.
Yet many teams have tried for years without progress.
Research previous attempts. Find the hidden constraint. Talk to past founders before assuming better execution is enough.
DO NOT CHASE PERFECTION
CHOOSE A STRONG STARTING POINT
Taking the first idea without evaluation is careless.
Waiting for a flawless idea creates paralysis.
Choose an idea with promising qualities, launch a basic version, and let evidence reshape it. Strong companies often emerge through several informed changes.
EVALUATE THE IDEA
CHAPTER 2
1. FOUNDER–MARKET FIT
ARE YOU THE RIGHT TEAM?
The best idea is not simply attractive in the abstract.
It must fit your team’s knowledge, skills, network, and lived experience.
Ask: Why are we unusually equipped to understand this problem, build the solution, and reach the first customers?
2. MARKET POTENTIAL
BIG NOW OR GROWING QUICKLY
A startup needs room to become large.
The market may already be substantial, or it may be small today but expanding rapidly because of technology, regulation, or behavior change.
Evaluate where the market is heading, not only its current size.
3. PROBLEM INTENSITY
HOW PAINFUL IS THE ALTERNATIVE?
Strong problems create urgency.
Look for users facing lost revenue, blocked work, high costs, risk, or an unavailable solution.
The best signal is often an unacceptable alternative. When users currently have no workable option, a useful product can become essential.
4. COMPETITION AND INSIGHT
COMPETITION CAN VALIDATE DEMAND
Existing competitors are not automatically a warning.
They can prove that customers spend money in the category.
The key question is whether you have a new insight: a missed user, better workflow, new technology, lower cost, stronger distribution, or a radically better experience.
5. PERSONAL DEMAND
DO REAL PEOPLE AROUND YOU WANT IT?
Ask whether you personally need the product or know people who clearly do.
A “no” does not kill the idea, but it increases the need for direct customer research.
Do not build from assumptions when conversations, observations, and purchase behavior can give stronger evidence.
6. WHY NOW?
WHAT RECENTLY CHANGED?
New opportunities appear when something becomes possible or necessary.
Look for shifts in technology, regulation, infrastructure, cost, and customer behavior.
A strong “why now” explains why the problem can be solved today and why older companies may be poorly designed for the new reality.
7. PROXIES
FIND PROOF IN ANOTHER MARKET
A proxy is a successful company solving a similar problem in another country, industry, or customer segment.
It is not a direct competitor.
A good proxy shows that the model can work while leaving room for a localized or specialized version.
8. LONG-TERM COMMITMENT
CAN YOU STAY WITH IT FOR YEARS?
You do not need instant passion for the industry.
Many valuable businesses begin in ordinary areas such as payroll, accounting, logistics, or compliance.
Interest often grows when customers depend on the product and the company gains momentum.
9. SCALABILITY
CAN GROWTH SEPARATE FROM LABOR?
Software can often serve more customers without increasing cost at the same rate.
Service businesses are harder to scale when every new customer requires scarce expert labor.
If the model depends on people, design systems, automation, standardization, or a marketplace that expands capacity.
10. IDEA SPACE
CHOOSE FERTILE TERRITORY
An idea space is the broader category around one product, such as hospital software or financial infrastructure.
Choose a space with meaningful problems, growing demand, and room for adjacent ideas.
A fertile space lets the company pivot without abandoning everything it has learned.
GOOD IDEAS THAT LOOK BAD
CHAPTER 3
HARD TO START
DIFFICULTY CAN PROTECT THE OPPORTUNITY
Some ideas require regulation, partnerships, complex infrastructure, or painful operational work.
That difficulty scares away capable competitors.
When the problem is important and the hard work is solvable, the barrier itself can become an advantage.
BORING MARKETS
NEGLECTED PROBLEMS CAN BE VALUABLE
Founders often chase exciting consumer ideas while avoiding payroll, tax, compliance, and back-office software.
Boring markets can contain deep pain, recurring demand, and weak products.
The daily work of building any startup is similar: code, users, bugs, sales, and iteration.
CROWDED MARKETS
A BAD CATEGORY MAY CONTAIN A GOOD OPENING
A crowded market can signal strong demand.
Study whether customers still dislike the available products or avoid using them entirely.
The opportunity is strongest when you understand exactly what competitors missed and can deliver a step-change improvement.
HOW STRONG IDEAS APPEAR
CHAPTER 4
NOTICE IDEAS ORGANICALLY
THE BEST IDEAS OFTEN EMERGE FROM EXPERIENCE
Many successful founders did not invent an idea during a brainstorming session.
They noticed a repeated problem while working, building, or living inside an industry.
Direct exposure reveals details that outsiders miss and makes the problem easier to validate.
BECOME AN EXPERT
WORK NEAR VALUABLE PROBLEMS
Expertise creates access to overlooked opportunities.
Work at the frontier of a useful field. Join a startup. Learn how an industry operates. Observe where people waste time, money, and attention.
The deeper your understanding, the better your ability to identify real gaps.
BUILD WHAT INTERESTS YOU
PROJECTS CAN EVOLVE INTO COMPANIES
Programmers can build tools, experiments, or products without forcing an immediate business model.
Useful projects create technical insight and attract early users.
Sometimes the business becomes visible only after people begin using the thing you built.
SEVEN WAYS TO GENERATE IDEAS
CHAPTER 5
1. START WITH YOUR ADVANTAGE
USE THE TEAM’S STRONGEST CAPABILITIES
List what each founder understands better than most people.
Combine domain expertise, technical ability, operational strength, and trusted relationships.
Generate ideas inside that intersection. Every idea found this way begins with stronger founder–market fit.
2. REVIEW YOUR OWN PROBLEMS
MINE JOBS AND LIFE EXPERIENCE
Review every job, internship, project, and unusual life experience.
Ask:
• What repeatedly frustrated people?
• What did I learn that outsiders do not know?
• Which process still feels outdated?
Personal exposure can reveal problems hidden from the startup ecosystem.
3. BUILD WHAT YOU WISH EXISTED
USE PERSONAL DEMAND CAREFULLY
A product you personally want can be a strong starting point because the problem is concrete.
Still ask why it does not already exist.
The gap may be an opportunity, or it may be blocked by behavior, economics, regulation, or distribution.
4. FOLLOW RECENT CHANGE
NEW CONDITIONS CREATE NEW NEEDS
Map what has changed in the last few years:
• New technology
• New regulations
• Lower costs
• New work patterns
• New consumer behavior
Then ask which old products are no longer designed for the current world.
5. ADAPT A PROVEN MODEL
USE SUCCESSFUL COMPANIES AS PROXIES
Study companies that recently succeeded.
Look for a version that serves another geography, customer type, workflow, or price point.
Do not copy blindly. Identify which assumptions transfer, which must change, and what local advantage makes your version defensible.
6. INTERVIEW THE MARKET
TURN CONVERSATIONS INTO A MAP
Choose a promising industry before interviewing people.
Talk to customers, workers, operators, suppliers, and founders. Ask about repeated problems, existing workarounds, spending, and failed solutions.
Over time, these conversations reveal where demand is strong and where ideas repeatedly fail.
7. FIND A BROKEN INDUSTRY
LARGE INEFFICIENCY CREATES ROOM
Look for major industries still dependent on phone calls, spreadsheets, paper, manual coordination, or fragmented vendors.
A broken industry does not guarantee a startup opportunity.
Find one specific workflow where a focused product can create measurable value.
TURN RESEARCH INTO EVIDENCE
CHAPTER 6
TALK TO PEOPLE CLOSEST TO THE WORK
EXPERTISE CAN BE ACQUIRED
You may enter an industry without deep expertise, but you cannot stay uninformed.
Spend time where the work happens. Observe users directly. Learn the language, incentives, economics, and constraints.
The goal is to build a reliable mental map before committing to a solution.
STUDY FAILED ATTEMPTS
PAST FAILURE IS USEFUL DATA
Search for companies that previously tried the same idea.
Ask what blocked adoption, destroyed margins, slowed distribution, or created operational complexity.
A failed predecessor may reveal that the idea is impossible—or show the exact assumption your new approach must change.
FOLLOW BEHAVIOR, NOT COMPLIMENTS
COMMITMENT IS STRONGER THAN INTEREST
Users may praise an idea without changing behavior.
Stronger signals include paying, switching tools, sharing data, accepting a pilot, or repeatedly returning.
Measure commitment. A startup becomes real when customers take action, not when they say the concept sounds interesting.
ALLOW INFORMED PIVOTS
STAY INSIDE A FERTILE SPACE
A weak first product does not always mean the entire direction is wrong.
Keep speaking with the same market. Each failed offer can reveal what customers actually value.
When the idea space is strong, learning compounds and guides the team toward a better product.
THE FINAL TEST
CHAPTER 7
LAUNCH A SMALL VERSION
REALITY RESOLVES UNCERTAINTY
Frameworks improve your odds, but they cannot prove an idea will work.
Build the smallest version that tests the main assumption. Put it in front of users. Ask for a meaningful commitment.
Launch before certainty, then let evidence decide what to keep, change, or stop.
A GOOD IDEA IS A STARTING SYSTEM
PROBLEM → INSIGHT → LAUNCH → LEARNING
Start with a painful problem.
Match it with founder advantage, a credible market, a clear “why now,” and a specific insight.
Then launch, observe behavior, and iterate. The strongest startup idea is not the one that sounds perfect. It is the one that creates fast, useful learning.