How ideas are researched and assessed

We start with problems people appear to have, not random product ideas. We look for repeated signals, group them into a clear problem, research the opportunity and only then propose and score a product worth testing.

Where researched ideas come from

Daily Product Idea does not begin by asking AI for random startup ideas. We first look where customers, professionals, operators and business owners discuss real work: repeated questions, complaints, workarounds, requests and relevant market changes. When a promising signal appears, we look for the same underlying problem across related communities and other evidence sources. Similar signals are grouped into a clear problem. AI can then help propose a focused product, research the opportunity and structure the assessment. Every researched idea is reviewed before publication, and evidence is kept separate from assumptions.

1. 1Observe focused customer and professional audiences
2. 2Find and group repeated problem signals
3. 3Expand promising problems across related communities and sources
4. 4Propose a focused product
5. 5Research the opportunity
6. 6Score the evidence
7. 7Review it
8. 8Publish one idea

Subscriber counts, upvotes and comment counts are useful context, but they never stand in for repeated independent evidence. A problem raised once in a large community is weaker evidence than the same problem raised repeatedly across several smaller ones.

Score dimensions

To score your own idea on these six dimensions, see is your product idea worth building? A practical 6-point test.

Score bands

• 80–100: Strong opportunity, still requires testing
• 65–79: Promising
• 50–64: Mixed
• 35–49: Needs a sharper angle
• 0–34: Not currently recommended

Scoring happens after problem discovery, clustering, candidate generation and research — not independently of it. Where evidence is weak, conflicting or absent, the score notes say so explicitly.

Evidence levels

Each researched idea's level is counted from its recorded sources, never judged, so the same evidence always gives the same level.

Illustrative:

An example showing the format. Not researched.

Limited:

Fewer sources than Indicative requires.

Indicative:

Two or more independent voices or sources describe the problem.

Supported:

Corroborated by two or more sites outside the original communities: reviews, job or search demand, or issue trackers.

Direct:

Observed direct evidence from customers, such as interviews, sign-ups, pre-orders or use.

Community posts count once per author and research findings once per site. Vendor and industry pages never count, and every level above Limited also needs the supporting sources to outnumber those that weaken the idea.

Only Direct evidence can justify the word "Validated," and the direct evidence must be described alongside the idea.

Every idea page lists the sources it was assessed from under "Evidence and sources", and records its methodology version and when it was last assessed. See what Daily Product Idea is or browse the ideas.