Before you build a digital product, you want evidence that strangers will pay for it. You can get that evidence with three tests, run from cheapest to most expensive: check that people already complain about the problem, talk to five to ten people who have it, then ask for money or a firm commitment before the product exists.
A weekend is enough for a first pass: the first test and the invitations for the second. The whole sequence tends to take a couple of weeks, because interviews depend on other people's calendars and a pre-sale needs a few days to collect responses.
Each test below has a pass signal, a fail signal, and a suggestion for what to do with a "no".
Write down your pass line first
Decide what result counts as a pass before you run each test, and write it down. Once you've put hours into an idea, a weak result starts to look encouraging. A number you committed to on Friday is much harder to argue with on Monday.
Test 1: Are people already complaining about this problem?
This test costs nothing and takes a few hours. You're looking for people describing the problem in their own words, unprompted, in places you didn't create.
- Search suggestions. Type the problem into Google and YouTube and read the autocomplete suggestions and the "People also ask" box. Google bases them on real searches. The process in my guide to finding low-competition keywords works here too, with a different goal: you want proof of demand, and rankings can wait.
- Forums and communities. Search Reddit, niche Facebook groups, and industry forums for the problem. Note how often it comes up, how recent the threads are, and whether people describe workarounds.
- Reviews of existing products. Read the two- and three-star reviews of tools, courses, or templates that already address the problem. Those reviewers paid for a solution and can tell you exactly where it fell short.
Existing competitors are a good sign at this stage. They show that somebody is already paying to solve this.
Pass: you find repeated, recent complaints from different people, plus evidence that some of them spend money or serious effort on a fix, such as paid tools, spreadsheets they built themselves, or hired help.
Fail: you find a handful of old threads, or the only person describing the problem is you.
If it's a no: try describing the problem the way a customer would, since they rarely use your vocabulary. If you still find nothing after a second attempt with different wording, the problem is probably too mild to build a business on. Drop it or pick a neighboring problem that did show up in your searching.
Test 2: Talk to five to ten people who have the problem
Search results show that a problem exists. Conversations show how much it hurts and what people have already done about it. Find people in the communities from Test 1, or ask people you know for introductions to someone in the target group. Friends and family are poor interview subjects because they want you to succeed.
Why "would you buy this?" is a bad question
It asks for a prediction about the future, and it's an easy one to be kind about. Saying "yes, I'd buy that" costs nothing and ends an awkward moment. You walk away with a compliment and no information.
Rob Fitzpatrick's book The Mom Test is built around this problem. His fix is to talk about the other person's life instead of your idea, and to ask about specific things they've already done instead of opinions about what they might do. Questions like that are hard to answer with a polite lie, even for your mom.
Five questions that work
- "Tell me about the last time you ran into this." A specific recent story shows how often it happens and how much it matters.
- "What did you do about it?" If the answer is "nothing", the problem may not be worth paying to solve.
- "What have you tried, and what didn't you like about it?" This gives you your competitors and their weak points.
- "What does it cost you, in money or time?" You need this to judge whether a price is realistic.
- "Who else should I talk to?" People who care about the problem usually know others who do.
Keep your idea out of the conversation until the end, if you mention it at all. Take notes in their exact words. You'll want that language for your sales page later.
Pass: most of the people you talk to have already tried to solve the problem, and a few have paid for something. The strongest signals are someone asking when they can get your version or offering an introduction.
Fail: people agree it's annoying but have never looked for a fix. Praise for the idea with no story behind it counts as a fail too.
If it's a no: look at who did care. If two of your eight had a serious version of the problem and share a job or situation, your audience may be narrower than you assumed. Run a few more interviews with only that group before deciding.
