Field guide
How to find a problem worth solving
The best problems to work on are hiding in plain sight. They show up in forum complaints, one-star reviews, and spreadsheets people built because nothing else worked. This guide walks you through how to find them.
Step 1. Pick a domain to explore
Start broad, then go deeper. Every problem lives somewhere in a domain. Starting with one keeps your research focused and makes it easier to find related problems once you are in that world.
Physical World
Digital & Knowledge
Step 2. Research the domain
Go to these sources in order. You are looking for patterns. The same complaint showing up in multiple places is a strong signal. One mention is not a problem.
- 1.Search the topic in Reddit search
- 2.Find the 2 to 3 most relevant subreddits
- 3.Sort by Top, This Month or Top, This Year
- 4.Look for posts with 50+ upvotes and comments like "same problem here"
- 5.Search the topic plus "spreadsheet" and the topic plus "frustrating"
Review sites (G2, Capterra, Trustpilot)
Second stop- 1.Find the software category closest to the problem
- 2.Filter reviews to 1 to 3 stars only
- 3.Read the "What do you dislike?" field on every review
- 4.Look for recurring complaints — these are your proof signals
- 5.Note the names of tools people switched away from and why
Google Trends
Validate demand- 1.Search your core problem keyword
- 2.Check: is the trend rising, flat, or falling?
- 3.Look at Related queries and filter to Rising
- 4.Compare 2 to 3 keyword variations to find the most searched phrasing
- 5.Check by country if the problem seems regional
App Store reviews
Underrated source- 1.Find the top 2 to 3 apps in the relevant category
- 2.Filter to 1 to 2 star reviews
- 3.People are brutally honest here — granular, specific pain
- 4.Note recurring complaints that mention what is missing
- 5.Cross-reference with the Play Store for the same apps
Step 3. Run it through four gates
Before you go further, the problem has to pass all four gates. If it fails one, either dig deeper until you find the evidence, or drop it and move on.
Is it a real problem or just an inconvenience?
Pass if:
- Multiple people complain about it unprompted in forums
- There are existing workarounds — spreadsheets, duct-tape solutions
- People have tried to pay for a solution and been disappointed
Fail if:
- Only one person mentioned it
- The problem only exists in extreme edge cases
- People shrug it off and move on — no real friction
Is it large enough to matter?
Pass if:
- A searchable subreddit exists with 50k+ members and active posts
- Multiple Facebook groups are focused on this topic
- Google Trends shows consistent or growing interest
Fail if:
- The audience is fewer than around 10,000 people globally
- The problem is highly regional and unlikely to scale
- Interest is declining on Google Trends
Does the gap actually exist?
Pass if:
- Existing solutions are too expensive, too complex, or too old
- Top-rated tools have 1 to 3 star reviews citing this specific gap
- Forum threads asking for solutions get "I use a spreadsheet" as the answer
Fail if:
- A good free or cheap tool already exists and people know about it
- The gap is acknowledged but a major player just fixed it
- The gap is actually just a preference, not a structural problem
Can you document it with specifics?
Pass if:
- You can cite at least 2 specific forum threads with real discussion
- You found at least one stat: market size, a percentage, or a dollar amount
- You can name specific existing tools and explain exactly why each falls short
Fail if:
- Your proof is "I feel like people have this problem"
- You can only find one vague mention of the issue
- The evidence is all secondhand or AI-generated
Step 4. Test the interest before you build
Research tells you the problem exists. This step tells you whether people will actually pay or sign up for a solution. Do this before writing a line of code.
Build a waitlist landing page
One page. One sentence describing what you are solving. An email field. Share the link. If people sign up without you begging them, that is a real signal.
Post about the problem on social media
Describe the problem as you understand it and ask if anyone else experiences it. Comments, shares, and DMs from strangers are proof. Silence is also useful data.
Go back to the communities and ask directly
Post in the relevant subreddit or Facebook group. Ask how people currently deal with the problem. Do not pitch anything. Just ask. The responses will either confirm your research or redirect you.
Offer to solve it manually for one person
Find someone who has the problem and offer to help them solve it, by hand, for free or for a small fee. If you cannot get one person to say yes, a product will not change that.
Build a minimal version and put a price on it
A simple tool, a template, a short guide. Put it on Gumroad or a basic site with a payment button. Real purchase intent is worth more than a hundred email addresses.
The goal is not a perfect product. The goal is a clear signal that the problem is real enough that strangers will act on it.
Found something worth surfacing?
If it passes all four gates, we would love to feature it.
Submit a problem