Problem Hunt was a crowdsourced project designed to help entrepreneurs discover problems beyond their own daily experience. As described in Joshua Kraus’s SitePoint profile, people submitted problems through a form, and the project planned to email ten compelling examples each week. The project’s historical account offers a useful starting point for anyone asking, “How do I find a problem to solve for a startup?”: gather real examples, then investigate what people actually do when those problems arise.
Why finding a problem can be harder than coming up with an idea
Entrepreneurs can readily identify frustrations they encounter themselves. The harder task is noticing problems faced by people with different routines, jobs, industries, or constraints. Cole McCollum, identified as Problem Hunt’s founder, put the challenge this way: “In short: identifying a problem to solve is really difficult. We make it easier.”
Kraus’s SitePoint profile, published July 16, 2015 and marked updated February 13, 2024, framed Problem Hunt as a way to broaden that search. The profile cited figures suggesting interest in entrepreneurship: 63 percent of people in their twenties wanted to start a business, attributed there to an Inc. survey from 2013, and 48 percent of Americans wanted to be entrepreneurs, attributed to a Fox Business study without a year given in the passage. It also repeated a claim that eight out of ten entrepreneurs who start a business fail within the first 18 months, without identifying a primary source in the passage. These are figures as reported by that article, not independently verified or current estimates.
How Problem Hunt was described
People could send problems to the project using a submission form. At the end of each week, Problem Hunt would send a newsletter featuring ten compelling problems. McCollum also tried to build a broader pool of examples by speaking with people in different industries and asking what problems they faced or noticed in everyday life.
The profile discussed user upvotes and comments as a possible future feature, not as a capability already in place. It does not establish whether Problem Hunt remains active today.
Examples of the problems people submitted
The SitePoint profile described everyday consumer frustrations rather than validated business opportunities:
- Wanting to check a phone too often.
- Not having a hands-free way to control Spotify while driving.
- Finding it difficult to plan events with friends far in advance, when tickets might still be inexpensive.
These examples can help prompt further questions, but the profile does not show that they led to successful products, paying customers, or viable businesses. A problem submission is a lead to investigate, not proof of demand.
How to investigate a problem with customer conversations
Broad questions such as “What problems do you have?” may produce general opinions. Questions about a specific recent experience are more useful because they uncover context, workarounds, and the effort involved. In a Y Combinator Startup School recap, YC partner and Pebble founder Eric Migicovsky recommends asking about the hardest part of a task, the last time someone encountered the problem, why it was difficult, and what they have already tried.
- Ask about the task. Find out what the person was trying to get done and what part was hardest.
- Ask for a recent example. Migicovsky’s prompt is: “Tell me about the last time that you encountered this problem.” Ask what happened, when it occurred, and what the person needed in that situation.
- Understand the difficulty. Ask what made that instance hard, inconvenient, costly, or time-consuming rather than assuming you know the cause.
- Explore existing workarounds. Ask what they tried, what they use now, and what remains unsolved. Existing effort can reveal how they respond to the problem, though it does not guarantee they would adopt a new solution.
- Ask what the problem costs them. When assessing a possible first customer, Migicovsky recommends finding out what the problem costs. The relevant cost may be financial or may involve time, effort, missed opportunities, or other consequences; let the person describe it rather than imposing a number.
These prompts help establish what happens in practice. They do not, by themselves, prove product-market fit or that a business can serve the need profitably.
How to compare problems without pretending there is a universal score
If several potential problems emerge, compare the evidence you gathered on four practical dimensions drawn from the interview guidance:
- Frequency: How often does the problem occur?
- Difficulty or pain: What makes it hard, and how significant is that burden to the person?
- Current response: What do people already do to manage or solve it?
- Cost: What does the problem cost the person or organization experiencing it?
The cited material does not define a scoring formula or a threshold for choosing a winner. Treat the dimensions as a way to organize evidence, not as a numerical test that guarantees a good business idea.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.When a structured interview resource may help
Strategyzer’s official Customer Jobs: Trigger questions page offers printable prompts for customer interviews and workshops. It notes that business-to-business discovery can involve different roles, including users, decision-makers, economic buyers, and channel partners. Those roles may have different problems and incentives, so a conversation with one person may not explain how an organization evaluates or pays for a solution.
The Strategyzer page advertised a Playbooks subscription for $299 per year when accessed on October 8, 2026; price and availability may change. This is a separate structured resource, not part of Problem Hunt.
What Problem Hunt’s account does—and does not—show
The historical profile presents a method for widening the search for problems: collect observations from others, including people outside your own industry, and follow promising examples with specific conversations. It does not report outcome evidence establishing that the project’s submissions became successful businesses, that any listed frustration represented a large market, or that a proposed solution would work. The useful lesson is the discovery process, not a promise that a crowdsourced list will identify a winning startup.
Quick Recap
Sources
- SitePoint: “Problem Hunt: Solving The Problem of Finding A Problem,” Joshua Kraus, published July 16, 2015; page marked updated February 13, 2024.
- Y Combinator: “Startup School Week 1 Recap: Kevin Hale and Eric Migicovsky”, September 4, 2019.
- Strategyzer: “Customer Jobs: Trigger questions”, official resource page accessed October 8, 2026.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
