Validate a Digital Product Idea Before You Build Anything

Here is the most common digital product story on the internet: someone spends three weekends building a planner, a template pack, or a course — polishing, designing, recording — launches it to silence, and concludes that digital products "don't work." The product was not the problem. The order was. They built first and checked for demand never. Validation flips the order: you spend a few days proving that real people want the thing, in their own words, with a real commitment, before you build anything beyond a sketch. This guide is the exact sequence — steal the question, run the comment test, fill a $0 waitlist, and only then build the smallest version anyone would pay for. It assumes you already have, or are building, a small audience — a channel, a page, a newsletter list of any size. If you are starting from zero audience, our faceless channel guide is the step before this one.

Step 1: Steal the question, not the product

Every good product begins as a question people are already asking in public. Your job is to collect those questions word for word — because the exact phrasing strangers use is worth more than any brainstorm. Three mines produce the richest ore:

• Comment sections — yours, and those of creators in your niche. Look for the comments that are really product requests wearing a disguise: "Is there a template for this?", "I wish someone would just make a checklist," "I keep messing up the middle part." Screenshot them. Ten of those is a product outline.
• Reddit and forums in your niche. Sort by top posts of the year and read the complaint threads — the ones that start "I am so frustrated with..." People describe problems there with a precision they never use in surveys, because they are venting, not performing.
• Amazon reviews of related books and products. The 3-star reviews are gold: they liked the idea and hated the execution. Every "I wish it had..." in a 3-star review is a feature your version can promise on day one. The 1-stars tell you what to avoid; the 5-stars tell you what to keep.

One rule keeps this honest: you are stealing the question, never the product. Copying someone's planner page-for-page is plagiarism and platforms treat it accordingly. Stealing the underlying frustration — "beginners cannot keep a content schedule" — and solving it your own way is just market research. Write the stolen questions at the top of a document. Underneath each one, in the buyer's exact words, is your future sales page, because people buy products described in language they recognize as their own thoughts.

Step 2: The comment test — teach one piece, end with the question

With candidate questions collected, run the cheapest demand test that exists. Make one piece of content — a video, a post, a carousel — that teaches one slice of the solution completely and free: the first worksheet of the planner, the core checklist, the single template. Teach it properly; a tease teaches nothing and tests nothing. Then end the piece with the product's question, asked plainly: "I am thinking of building the full version of this — the complete 30-day system with all the templates. Would that be useful to you? What would it need to include?"

Now read the replies like a scientist, not an author. You are not counting compliments; you are counting specifics. "Yes, and it has to work for night shifts" is a vote. "Would pay for this if it came as a printable" is a strong vote. "Great video!" is not a vote at all — it is politeness, and politeness has launched a thousand dead products. A healthy comment test on a small channel looks like this: a handful of replies naming concrete features, constraints, or use cases. Those replies are doing double duty — they validate the idea and they are writing your feature list in buyers' language. If instead you get silence or pure compliments, that is data too: the slice did not hurt enough to matter. Rework the question or move to the next candidate before building anything.

Step 3: The $0 pre-sale — a waitlist page and the 10-stranger rule

The comment test measures interest. The waitlist measures commitment, and commitment is the only currency that predicts sales. Set up the simplest possible page — a free Gumroad or Carrd page, or even a form — that says what the product will be, who it is for, roughly what it will cost, and offers one action: join the waitlist for a launch discount. Total build cost: $0 and one evening. Then put the link everywhere the comment test ran, and drive your small audience to it for a week.

The benchmark is deliberately small and deliberately strict: 10 sign-ups from strangers — people who found you through content, not people who love you. Here is why strangers are the whole test. One hundred compliments from friends, family, and fellow creators cost those people nothing and predict nothing; your aunt will join any waitlist you send her. Ten strangers who watched a video, clicked through, read a page, and handed over an email address have each taken three real actions for an idea that does not exist yet. That behavior — not encouragement — is what a launch looks like in miniature. Miss the benchmark and the correct move is not a bigger build; it is a better question. Go back to your mined comments, pick the next candidate, and test again. Each failed test costs you days. Each unvalidated build costs you months.

Step 4: Build the smallest paid version — let buyers write v2

Ten strangers on the list? Now — and only now — build. And build small: the smallest version that fully delivers the one promise your waitlist page made. If the promise is "a 30-day content calendar system," v1 is the calendar template plus a short guide to using it — not the calendar plus a course plus a community plus a bonus vault. Every extra feature delays the only feedback that matters, which is buyers using the thing and telling you what is missing. Price it in the beginner sweet spot — $9–27 — and launch it to the waitlist first, with the promised discount. Our digital products guide covers the formats, the Canva build, and the platform trade-offs in detail.

Then comes the part beginners skip: let the buyers write version 2. Every support email asking "how do I do X?" is a missing instruction page. Every refund request carries a sentence that tells you exactly what the sales page over-promised or the product under-delivered. Every reply to "what would make this more useful?" is a feature request from someone who has already paid — the most valuable product advice in the world, and it is free. Collect them in one document. When the same request appears three times, that is your v2 roadmap, written by the market instead of your guesses. Release v2, raise the price slightly for new buyers, and give v1 buyers the upgrade free — you will never buy goodwill cheaper than that.

How long validation should take (and what it replaces)

The full sequence — mining, one comment-test piece, a waitlist page, a week of sign-ups — takes about ten days part-time. Compare that with the default path: three to six weekends building blind, a launch to silence, and a slow, demoralizing conclusion that the niche was wrong when the idea was simply never tested. Validation is not a delay added to your launch. It is the part of the launch that decides whether the rest of the work is worth doing at all. Experienced product sellers run some version of this loop for every single release, including the cheap ones — because the loop is cheap and being wrong is expensive.

Frequently asked questions

What if I have almost no audience yet?
Then your validation channel is other people's audiences: answer questions in niche communities with your free slice, post it where your buyers already gather, and watch the specifics in the replies. The waitlist benchmark scales down — even 5 stranger sign-ups gathered the hard way is real signal. Zero audience and zero response is the market telling you to build audience first.

Should I charge during pre-sale instead of a free waitlist?
A paid pre-sale is a stronger signal, but it comes with real obligations: a firm delivery date, clear refund terms, and the discipline to ship on time to people who trusted you early. For a first product, the free waitlist plus a launch discount gets you 90% of the information with none of the liability. Graduate to paid pre-sales on product two or three.

What exactly counts as a "specific" reply in the comment test?
Anything that mentions a feature, a constraint, a format, or a use case: "needs a printable version," "does this work for teams," "I would use this for my clients." Specifics mean the person pictured themselves using the product. Generic praise means they pictured themselves being nice to you. Only the first group buys.

Someone could steal my idea while I validate publicly.
They could — and it almost never matters. Ideas are the cheapest part of a product; the audience that watched you build it, the waitlist, and the v2 feedback loop cannot be copied by screenshot. Secrecy protects the one asset you do not have yet (a proven product) at the cost of the one thing you need (proof). Validate in public.