Is your product idea worth building? A practical 6-point test

Published 7 October 2026 · By Daily Product Idea · Our methodology

How to use the test

Score your idea from 0 to 100 on each of the six points below, using only what you can show: a post, a price, a search result, a conversation. Where you have nothing, score low and say so. These are the same six dimensions, with the same weights, that Daily Product Idea uses to score every idea it researches. Problem evidence counts for 25% and the other five for 15% each.

Give yourself an afternoon. Most of the work is looking things up. If you find yourself scoring from memory or enthusiasm, stop and go and find the source.

This is the quick version. For the fuller method, including writing a reason for every score and separating findings from assumptions, see how to evaluate a product idea before you build it.

1. Problem evidence (25%)

The question: can you show, from independent sources, that specific people have this problem repeatedly and that it costs them something?

What passes

• Several unconnected people describing the same difficulty in their own words, recently.
• Signs of a cost: hours lost, money spent, mistakes made, deals missed.
• A workaround: a spreadsheet, a script, a paid freelancer, a manual routine they repeat.

What fails

• The problem is yours, or a friend's, and you have not seen anyone else describe it.
• One popular thread, or several posts by the same person.
• People agree it would be nice, but nobody has done anything about it.

This point carries the most weight because the other five depend on it. If it fails, stop here: a product nobody needs is not rescued by being easy to build. Flight Deck scores 70 for problem evidence because three separate issue-tracker threads describe the same difficulty. That is the kind of evidence to look for. For where to find it, see how to find problems worth solving.

2. Customer clarity (15%)

The question: can you name the customer precisely enough to find them, and do your sources agree on who they are?

• Passes: "bookkeepers handling invoices for several small clients", "consultants rebuilding the same weekly report from several exports". You could search for these people by job title or community.
• Fails: "small businesses", "developers", "anyone who…". If everyone is a customer, you cannot find the first one.

Also check who pays. The person with the problem and the person with the budget are not always the same, and an idea can be clear about one and vague about the other. LedgerLine scores 55 here because the research was less clear about exactly who the customer is.

3. Market potential (15%)

The question: is there evidence that people spend money or paid time on this problem, or only that they complain about it?

• Passes: paid tools, templates, courses or freelancers already sold for the problem; people describing hours of paid staff time spent on it; steady search demand for a fix.
• Fails: no sign of spending anywhere, or only free tools that people seem content with.

Complaints are cheap; spending is a commitment. Explain-Before-Merge scores 42 for market potential, higher than ideas with similar problem evidence, because its research found people already paying for a manual alternative. Dropbox for Agents has stronger problem evidence, at 78, but scores 30 here because no comparable spending was found. To test this point properly, see how to test whether people will pay.

4. Differentiation (15%)

The question: after listing everything people use today, can you say in one sentence what your product would do that none of it does?

List the direct competitors, but also the free and open-source tools, the features built into platforms your customers already use, and the manual workaround. The workaround is often the real competitor.

• Passes: a narrow, specific gap you can name, and evidence that customers feel it.
• Fails: "it's like X but better", "nobody else does this" (before you have looked), or a gap that only you seem to mind.

Flight Deck scores 20 here: the research found several free tools covering most of what it proposed. See how to assess the competition for a fuller method.

5. Buildability (15%)

The question: can a small team build the smallest version that someone would use, and what does it depend on that you do not control?

• Passes: a first version you can describe in a paragraph and build in weeks, on data and platforms you can reliably get at.
• Fails: it only works at scale, needs a partnership or data you cannot get, rests on a platform that could shut you out, or depends on accuracy nobody has shown is possible.

Score this one for yourself, not for an imaginary team. LedgerLine scores 80: its core is reading invoices with a vision model, which can be tried in days. Trade Reality Dossiers scores 30, not because the writing is hard, but because it depends on small operators sharing their real margins, which no source showed they would.

6. Customer access (15%)

The question: where exactly would you find your first twenty customers, and what would it cost to reach them?

• Passes: named communities, newsletters, directories, marketplaces or people you could contact this week.
• Fails: "social media", "content marketing", "go viral", or a channel that only works once you already have customers.

This is the point most often skipped and most often fatal. Deal Floor scores 34 for customer access, the lowest of its six scores. Like buildability, this point depends on you. If you already know the customers, score it higher, but only if you could name them.

Adding it up

Multiply each score by its weight and add them up. Here is the arithmetic for an illustrative idea, not a real one: a tool with a fair amount of problem evidence, an easy build and no obvious way to reach its customers.

Then read the total in 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

The example comes out at 56, "mixed". The total hides the most useful thing the scores say: the idea would be easy to build, but there is no plan for finding customers. That is what to work on, or test, next.

Reading the result

A total is a summary. Before deciding anything, look at the shape:

• Problem evidence below about 40: go back to point 1, whatever the total. Find more evidence or a different problem.
• One point far below the rest: that is your riskiest assumption. Design the cheapest test that could prove it wrong, and run it before building anything.
• High scores you cannot point to a source for: mark them down. A score you cannot explain tends to drift upwards.
• A high total: means "worth testing", not "validated". Only what customers do, such as sign-ups, pre-orders or use, can show that.

Comparing two ideas? Compare them point by point, not by total: see how to choose between two product ideas. For the fuller method behind the six points, including writing reasons and separating findings from assumptions, see how to evaluate a product idea.

The test on one page

1. Problem evidence (25%): can you show independent people having this problem, and what it costs them?
2. Customer clarity (15%): can you name the customer precisely enough to find them, and who pays?
3. Market potential (15%): is anyone already spending money or paid time on it?
4. Differentiation (15%): what would you do that every current option, including the workaround, does not?
5. Buildability (15%): can you build the smallest useful version in weeks, on things you control?
6. Customer access (15%): where exactly are the first twenty customers, and what does reaching them cost?

Every researched idea on Daily Product Idea is published with its six scores and the reasons behind each, so you can see the test applied to real ideas. See how ideas are researched and assessed or browse the researched ideas.