How to Validate a Digital Product Idea in 90 Minutes

I once built an entire product before checking whether anyone wanted it. Designed it, polished it, priced it. Then I finally stress-tested the competition and shelved the whole thing. The market was already crowded with better versions at lower prices, and an honest evening of checking would have told me that before I spent the month.

That mistake is why validation is Stage 1 of my own workflow, and why it is non-negotiable. Here is the whole thing: 90 minutes, four passes, and it ends in a real decision.

Pass 1: Demand Signals (20 minutes)

Search the platforms like a buyer, not a seller. Type the idea into Etsy and watch the autocomplete. Autocomplete is a directional signal, not proof: it tells you Etsy associates the phrase with buyer behavior, not how many people search it or whether they spend. So stack signals instead of trusting one. Do comparable products show recent reviews? Does the same product type exist on other platforms? Are people asking about the problem in communities you can find in five minutes?

Then run the first gate, and be honest. It is not about the raw listing count. A big category is fine if you can slice it: can you narrow the idea by buyer, use case, format, or outcome and still have an audience? "Planner" is an ocean. "ADHD-friendly weekly reset planner" is a beach you can actually land on. If every narrowing you try shrinks the audience to nobody, the idea is too broad for you, even if the category is huge.

Pass 2: The Gap Pass (30 minutes)

If demand signals stack up, stop asking what is selling and start asking what buyers keep requesting and not getting. Pull 10 to 20 relevant listings, not just the top screen, since marketplace rankings mix ads, trends, and personalization. Skip the 5-star reviews and read the 3-star and 4-star ones. That middle band is where buyers say "good, but I wish it had..."

Then step off the marketplace for ten minutes. Reddit threads, Facebook groups, Pinterest search suggestions, and reviews of comparable products elsewhere will tell you things Etsy reviews will not.

Two competitive readings to keep straight. Incumbents with thousands of reviews confirm demand and raise the cost of entry at the same time: enter only if you have a real difference in audience, outcome, or format, because you will not cross that moat with a better font. And near-empty niches are a warning, not a goldmine. Sometimes the niche is new. More often the buyers are somewhere else, or nowhere.

Whatever complaint keeps repeating across sellers is the gap, and the gap becomes your angle in the product, the listing copy, and the tags. You are not entering the niche. You are entering the hole in the niche.

Pass 3: Reach and Economics (20 minutes)

The pass most sellers skip, which is why products with real gaps still die. Four checks, one line each.

Buyer: Can you name the exact person? Not "sellers." The overwhelmed Etsy seller with 30 listings and no sales.

Pain: Is the problem costly, repeated, or stressful enough that they already spend money on it?

Reach: Do you know where these buyers search or gather, and can you actually show up there?

Economics: Does the likely price clear platform fees, promotion, and your support time? The pricing breakdown does this math.

Any check you cannot fill in honestly is homework or a kill, not a shrug.

The Decision (20 minutes)

Then pick one of three doors, not two.

Build. The sentence wrote itself, the signals stack, all four checks pass.

Test. Plausible but unproven. Do not build the full product. Put up a free sample or condensed version on Gumroad, make two or three pins with different angles, and track the whole chain for two weeks: impressions, clicks, page visits, downloads. Read the result honestly. No downloads after real qualified traffic is a warning about the idea. No traffic at all means the promotion failed, not the product, so fix the pins before you blame the demand. A freebie that nobody wants after a fair test just saved you a month.

Kill. No nameable buyer, no repeated pain, no reach, or no math. Killing at 90 minutes costs an evening. Killing after building costs a month, and it stings enough that most sellers ship it anyway and then wonder why it will not sell. I know because I have done it.

Where the AI Comes In

Honest version: a prompt cannot measure live marketplace demand for you. What it does is turn the evidence you just gathered into a decision. Feed the Niche Validator your search observations and it pressure-tests the gates and flags the assumptions you are making. Paste competitor review excerpts into the Gap Finder and it ranks the repeated complaints into an angle. The prompts compress the thinking, not the looking. Condensed versions of both are in the free Starter Toolkit, and the full Stage 1 and Stage 8 systems live in the Workflow OS guides.

Get the Free Starter Toolkit →