How to Score and Rank Product Ideas by Real Demand (Not Gut Feel)

Finding demand is only half the problem. The harder half is choosing. Most founders don't have one product idea — they have a dozen, and every one of them feels like "the one" on a Tuesday and like a dead end by Thursday.

The fix isn't more research. It's a scoring system that forces every idea through the same filter, so the one with the strongest demand signal wins — not the one you're most emotionally attached to.

Here's a repeatable framework for scoring and ranking product ideas by real demand, using the same public signals you'd use to find them in the first place.

Step 1: List every idea as a one-line problem, not a product

Before you score anything, strip each idea down to the problem it solves. "A CRM for freelancers" is a product. "Freelancers lose track of which client owes them money" is a problem.

Write each idea as: [Who] has [what problem] and currently [what workaround].

If you can't fill in all three blanks, the idea isn't ready to score — it's still a vague product fantasy. Park it and move on.

Step 2: Score each idea on four demand dimensions

For every idea, rate it 1–5 on each of these. The numbers come from evidence you can actually go look at, not from how excited you are.

1. Frequency — how often does the problem show up? Search your buyer's public feeds (Bluesky, X, Reddit, niche forums) for the problem phrased in their own words. Count posts in the last 30 days.

• 1 = a handful of mentions
• 3 = steady weekly mentions
• 5 = daily, recurring complaints

2. Intensity — how much does it hurt? A "nice to have" gets a shrug; a real pain point gets caps-lock and profanity. Read the tone of the posts, not just the count.

• 1 = polite feature requests
• 3 = frustration, "I wish this existed"
• 5 = "I'm losing money / hours / clients because of this"

3. Buying intent — are they looking to pay? Look for the phrases that separate a complainer from a customer: "is there a tool for", "what do you use for", "I'd pay for", "recommend me something".

• 1 = pure venting, no solution-seeking
• 3 = asking for recommendations
• 5 = actively paying for a workaround today

4. Reachability — can you actually get in front of them? Demand you can't reach is demand you can't convert. Do these buyers cluster somewhere you can show up — a subreddit, a hashtag, a community, a search term?

• 1 = scattered, no obvious channel
• 3 = one clear channel
• 5 = a dense, active community you can join today

Step 3: Multiply, don't average

Here's the key move most scoring frameworks get wrong: they average the four scores, which lets a 5 on one dimension hide a 1 on another. A product with intense pain but zero reachability is a trap.

Multiply instead:

Demand Score = Frequency × Intensity × Buying Intent × Reachability

The range is 1–625, and the multiplication punishes zeros hard. An idea that's a 5 on frequency and intensity but a 1 on buying intent scores 25 — not 3.5. That's the point: a loud problem nobody will pay to solve is not a business.

Step 4: Apply the two disqualifiers

Before you rank, run every idea through two hard gates. Either one kills the idea regardless of score.

The "already solved" gate. If a mature, cheap, well-loved tool already owns the problem, your demand signal is real but the market is taken. You can still enter — but only with a sharp wedge, not a me-too clone.

The "you can't reach them" gate. If reachability scored a 1, the demand is theoretical. You'll burn months trying to find buyers who never see you. Kill it or find the channel first.

Step 5: Rank and pick the top one — then commit

Sort your surviving ideas by Demand Score. The top one is your answer. Not the top three — the top one.

This is the part founders resist. They want to keep two or three ideas alive "just in case." But every idea you keep alive is a competing claim on your attention, and a founder's attention is the scarcest resource in the business.

Pick the winner. Write down the runner-up and the score that lost to it, so you can revisit it later with a clear head instead of a wandering one.

The takeaway

You don't have an idea problem — you have a prioritization problem. The founders who build the right thing aren't the ones with the best instincts; they're the ones who force every idea through the same evidence-based filter and let the strongest demand signal win.

Score your ideas on frequency, intensity, buying intent, and reachability. Multiply, don't average. Apply the two disqualifiers. Then build the one that wins — and let the runner-ups wait their turn.

That's how you stop guessing and start building what people will actually pay for.