The Art of Identifying a Problem
Pitfall of falling in love with your next great idea
Often in the entrepreneurial world, founders fall into one of the most seductive traps:
Thinking about the solution before truly understanding the problem.
And it’s understandable —we’re human.
We fall in love with our product vision. We believe our design is sleek, our tech is exciting, and our features are obviously what the world needs.
Why Startups Fail?
But, the truth is that the leading cause, 42%, of startups fail because there’s no market need [2]. Not because the tech wasn’t good. Not because they couldn’t raise money. But because they solved a problem no one really had.
This means that many founders are building solutions in search of a problem, not the other way around.
The Solution
So how do you escape this echo chamber? How do you uncover real and urgent problems?
Luckily, this process has been studied, refined, and implemented by founders at Airbnb, Stripe, Notion, Dropbox, and many others throughout the history. So, you have a full literature sitting in front of you.
Early-stage investors such as Antler consistently emphasize that their most successful portfolio companies are those that begin with thorough and in-depth user discovery, engaging deeply with potential customers to validate real problems before building solutions [1].
So, instead of guessing what people need, use this proven path:
Start with users’ reality instead of going directly into a solution
Structure non-leading, behaviour focused interviews with the users
Identify jobs-to-be-done
Find the right ICP (i.e. ideal customer profile) and talk to them instead of a random pool of people
Map insights to a clear, urgent problem
Step 1: Start with the User’s Reality
“Make something people want.” [5]
But first, you need to know what they want.
Before you pull up a design draft or brainstorm features, step into the life of your potential user.
What does their process look like? Where do they struggle? What moments feel frustrating, broken, or tedious?
This is called “problem space research” [3] And it’s how companies like Spotify designed products that feel inevitable. Before building their recommendation engine, Spotify studied how people found music —manually, socially, passively— then designed around those discovery behaviors [8].
The goal is not to validate. It’s to understand.
Step 2: Structure Non-Leading Interviews
IDEO’s “Human-Centered Design Toolkit” emphasizes that the richest insights come from stories, not opinions. You want memories, not hypotheticals.
The best interviews don’t ask:
“Would you use this?”
They ask:
“Walk me through the last time you tried to solve this.”
“What happened next?”
“What was frustrating?”
“Where did you turn for help?”
Airbnb got this right. In the early days, the founders conducted interviews with hosts and guests about trust, travel horror stories, and empty apartments. They didn’t ask “Would you host a stranger?” or “Would you use a room-renting website?”.
Stripe did the same —they spent hours with developers observing the frustration of integrating PayPal and Authorize.Net [7].
Step 3: Identify Jobs-to-Be-Done
Use Jobs-to-Be-Done framework (JTBD) —which is a powerful tool for understanding why people use or don’t use a product. It is based on a theory that people “hire” products or services to get a specific job done in their life [4].
This mindset helps you design not just features, but solutions that fit real life goals and needs in every aspect such as functional, emotional, social.
Functional jobs
These are about what the user is trying to accomplish in practical terms.
“I want to grow my savings faster than a savings account.”
“I want to beat inflation.”
“I want to earn passive income without constant management.”
“I want to diversify my assets to reduce risk.”
“I want to retire early with enough money to live comfortably.”
Emotional jobs
These are about how the user wants to feel during or after the process.
“I want to feel smart and in control of my finances.”
“I want peace of mind knowing my money is working for me.”
“I want to feel secure, not anxious, about my future.”
“I want to stop feeling guilty about not doing more with my savings.”
“I want to feel proud when I check my portfolio.”
Social jobs
These are about how others perceive them or how they compare themselves to peers.
“I want to be seen as someone who understands money.”
“I want to have something to talk about with financially savvy friends.”
“I want to avoid looking careless or irresponsible to my family or partner.”
“I want to post or talk about investments confidently on social media or in groups.”
“I want to match or outperform people I follow online (e.g. Finfluencers).”
If you don’t understand all these basic three layers of a user’s “job,” you’ll likely build something that’s useful —but probably not sticky.
There are also extensions of these three basic jobs. Such as:
Contextual Jobs
These jobs can vary depending on life stage, income level, or recent events.
“I just started my first job, and I want to put my salary to good use.”
“I’m saving for a wedding in three years and want to grow the fund.”
“I inherited money and need to make smart decisions with it.”
“I’m switching careers and need a safety net.”
“I’ve seen people struggle in retirement, and I want to avoid that.”
Anti-Jobs (Negative Motivators or Hidden Fears)
These are jobs the user is trying to avoid.
“I don’t want to lose money because I picked the wrong stock.”
“I don’t want to feel dumb for following hype.”
“I don’t want to spend hours researching every investment.”
“I don’t want to be locked into something I don’t understand.”
A good example of understanding the emotional and social layers is Duolingo’s approach. It is what helped Duolingo turn language learning into a game. People never said “I want gamification”, but they wanted progress and motivation [6]. So, the job wasn’t just “learn a language”, but “Help me feel like I’m making daily progress — without feeling bored or dumb”.
JTBD is worth to consider before and also after talking to the people.
Step 4: Talk to the (Right) People
Interviewing people who are too far from the problem is a key mistake. It will just dilute the process of defining the problem.
Dropbox famously focused their early user research on individuals who frequently shared files via email or FTP, enabling them to identify a clear pain point and accelerate product-market fit [9]. That specificity probably saved them years of iteration.
Rather than asking people what they would do, study what they already do —and where they fail based on your prep with previous steps.
Step 5: Map Insights
Now turn your raw interview notes and JTBD into a compelling narrative:
WHO: Who is this person?
WHAT: What pain do they encounter again and again?
WHY (NOW): Why is this pain urgent, expensive, or growing?
“We talked to 12 Gen Z/Millennial investors. Most opened an investing app, got confused, then left it unopened for months. They weren’t unmotivated —they were overloaded. That’s the moment we’re solving.”
Conclusion
Most of the best ideas are triggered by a common pain point from others. Don’t fall in love with your idea and solution, but have a research mindset constantly. Don’t shut down the useful insights might be coming from understanding the problems of similar profiles.
References
Antler Academy. Validating articles, https://www.antler.co/library/validating.
CB Insights. “The Top 20 Reasons Startups Fail.” CB Insights, 2019, https://www.cbinsights.com/research/startup-failure-reasons-top/.
IDEO.org. The Field Guide to Human-Centered Design. IDEO, 2015, https://www.designkit.org/resources/1.html.
Christensen, Clayton M. Competing Against Luck: The Story of Innovation and Customer Choice. HarperBusiness, 2016.
Essays by Paul Graham. “Be relentlessly resourceful”. Y Combinator, 2009, https://www.ycombinator.com/library/94-be-relentlessly-resourceful.
Appel, Gilad. “Duolingo: How Gamification Took Language Learning to the Next Level.” Harvard Business School Case Study, 2020.
Reddit. “Stripe is Paypal circa 2010”, https://www.reddit.com/r/programming/comments/ycxpu1/stripe_is_paypal_circa_2010/.
Hackernoon. “Spotify’s Discover Weekly: How machine learning finds your new music”, 2017, https://hackernoon.com/spotifys-discover-weekly-how-machine-learning-finds-your-new-music-19a41ab76efe.
Goodwater. “Understanding Dropbox: Consumerizing the Cloud”, 2018, https://www.goodwatercap.com/insights/thesis/understanding-dropbox/