The hardest question in early-stage startups is not "how do I build this?" — it is "should I build this at all?" The difference between ideas worth pursuing and ideas worth abandoning is often subtle, and the excitement of a new idea makes objective assessment almost impossible.
Here are 10 concrete, observable signals that an idea has real potential — not theoretical potential, but signals you can actually find in the world before writing a line of code.
Signs 1–3: Market Signals
1. People are already paying for a bad alternative. This is the strongest possible signal that an idea is worth pursuing. If you can find customers paying for a product with 2-star reviews filled with complaints, you have a validated market with a low quality bar to clear. The willingness-to-pay is already demonstrated; you just need to execute better than the incumbent.
2. The problem comes up repeatedly in niche communities. Search Reddit, Discord servers, Slack communities, and LinkedIn groups related to your target user. If threads about the problem keep appearing, with different people describing the same frustration month after month, the problem is chronic rather than occasional. Chronic problems generate repeat purchasing behavior.
3. Competitors are growing despite obvious flaws. Check competitor job listings, blog traffic trends on SimilarWeb, and App Store review volume over time. If a competitor with a notoriously bad UX or missing features is clearly growing, the market demand is strong enough to overcome bad execution. That is a category you can win by being competent.
Signs 4–6: Customer Signals
4. You can describe your customer in one sentence — and immediately name 10 of them. "Freelance graphic designers who use Figma and manage client feedback over email" is a real customer. "Small businesses that need better tools" is not. If you can name the customer concretely and immediately think of 10 specific people who match that description, you are solving a real problem for a real group.
5. Potential customers respond to cold outreach with interest, not confusion. Write a 3-sentence cold email describing the problem (not your solution) and ask if it resonates. If you send 20 emails and get 5+ responses saying "yes, this is a real pain," you have validation. If people are confused about why they would care, you have a positioning problem or a problem that is not as universal as you assumed.
6. Someone has already built a manual or spreadsheet workaround. When people build spreadsheet systems, Zapier workflows, or manual processes to work around a missing product, they have already validated that the problem is painful enough to invest time in solving. Your product replaces manual effort — the value proposition writes itself.
Signs 7–8: Competitive Signals
7. Competitor reviews mention the same 2–3 missing features repeatedly. When you read 50 reviews of a competing product and the same feature request appears in 20 of them, you have found an underserved need that the market has already validated and the incumbent has failed to address. Build that feature as your core differentiator.
Running this analysis manually is time-consuming. GoShipOrSkip pulls and analyzes competitor reviews from Product Hunt and Google Play at once, surfacing the most frequently cited pain points and unmet needs across an entire category.
8. No single player dominates with more than 30–40% market share. In fragmented markets, customers have not found a product they love — they have settled. Fragmentation is an invitation. In dominated markets, new entrants need a genuinely new category or angle, not an incremental improvement.
Signs 9–10: Personal Signals
9. You have domain knowledge or access that creates a genuine advantage. The best startup ideas come from people who deeply understand a specific domain — not because they thought the problem sounded interesting, but because they lived it. If you spent 5 years in healthcare operations and you are building a healthcare operations tool, you start with insights that a generalist competitor cannot easily acquire.
10. You can articulate exactly why existing solutions fail — in one sentence. "Existing tools are complex" is not a differentiation statement. "Existing tools require a 3-week onboarding process that small teams cannot afford" is. If you can describe the incumbent's specific failure mode precisely, it means you understand the market deeply enough to build something better. If you can only say the incumbents are "too expensive" or "hard to use," you need more research.
What to Do If You Score 7+ Out of 10
If your idea shows 7 or more of these signals, you have enough evidence to move forward with low-cost validation experiments: a landing page, customer interviews, and a simple prototype. You are not ready to build the full product yet — but you are ready to invest 30–60 days in proving or disproving your hypothesis.
If you score fewer than 5, the idea may still be viable, but you need more market research before investing development time. Go find the signals that are missing. If you cannot find them after genuine effort, the idea may need to be reshaped.