Find out if people care before you build too much
By now, you have shaped the idea. Now you need to see whether the right people care enough to do something.
Not "would you use this?" People are nice and that question is basically useless.
You're looking for stronger signals: replies, waitlist signups, discovery calls, pre-orders, pilot interest, community comments, repeated complaints, or people asking when they can try it.
By the end, you should have:
- a simple demand test
- the audience you'll test with
- the message you'll use
- what signal counts as interest
- what signal counts as noise
- what to do next based on what happens
1. Pick the signal you want
Not all validation is equal.
A like is not the same as a reply. A reply is not the same as a call. A call is not the same as someone asking for access, and someone asking for access is not the same as someone paying.
You do not need perfect proof yet, but you do need a signal that means more than polite interest.
| Weak signal | Stronger signal |
|---|---|
| "Cool idea" | "I have this problem right now" |
| Likes on a post | Replies from the right customer |
| Waitlist signups from random people | Waitlist signups from your starting segment |
| "I'd use this" | "Can I try it?" |
| "This sounds useful" | "How much will it cost?" |
Choose the signal before you run the test, so you do not convince yourself everything is validation after the fact.
2. Write the test around the pain
A good demand test starts with the problem, not the product.
If you lead with the feature, people will react to the idea. If you lead with the pain, you'll learn whether the problem actually resonates.
Weak:
I'm building an AI dashboard for support teams.
Stronger:
Do recurring customer issues keep getting buried across Intercom, Slack, and product feedback? I'm testing a lightweight way to spot patterns before they become account risks.
The second version gives someone something to recognize. That's what you want.
If the right customer does not see themselves in the pain, the product pitch probably does not matter yet.
3. Choose the smallest test that can teach you something
You do not need a huge launch to test demand. You need a small test with the right people.
That might be:
- 10 direct messages to people in your starting segment
- a short LinkedIn or X post written around the pain
- a landing page with a waitlist
- a Reddit or community post asking about the workflow
- a simple mockup shared with potential users
- a call request with people who already complain about the problem
- a lightweight offer for a manual service version
The best test depends on the customer. If they are active in niche communities, go there. If they are in your network, message them. If they search for the problem, test a landing page.
Do not hide behind the test that feels easiest if your customer will never see it.
4. Look for behavior, not compliments
People will say nice things. That does not mean they want the product.
Useful validation usually has effort attached. They reply with details, share their current workaround, ask for a demo, join a waitlist, forward it to someone, book a call, or offer to pay.
The more effort they put in, the stronger the signal.
A good test should help you answer:
- Do they recognize the pain?
- Do they describe the problem in their own words?
- Are they already trying to solve it?
- Do they want to see what you're building?
- Do they have urgency now, or is this a "someday" problem?
You're not looking for universal excitement. You're looking for enough signal from the right people.
5. Decide what the result means
A demand test is only useful if you know how to interpret it.
If the right people respond, ask follow-up questions and keep going. If the wrong people respond, your message or channel may be off. If nobody responds, the pain, customer, or positioning probably needs work.
| Result | What it might mean |
|---|---|
| Right people reply with details | The pain is resonating |
| Right people join the waitlist | There may be early demand |
| Wrong people respond | Your message or channel is too broad |
| People like it but do nothing | Interest is weak or low urgency |
| Nobody responds | The pain, customer, message, or channel needs work |
No response is still data. Annoying data, but data.
Prompt
This prompt is not here to make demand look better than it is. It's here to design a test and help you read the signal honestly.
Demand test
Take all of my notes and help me design a simple demand test.
Please structure it into:
1. Pain being tested
2. Starting customer segment
3. Best channel to test with
4. Test message
5. Strong signal
6. Weak signal
7. Follow-up questions
8. What result means keep going
9. What result means rethink
10. Biggest risks in the test
Then audit the test.
Please flag:
- where the message is too product-focused
- where the audience is too broad
- where the signal is too weak
- where I might mistake compliments for demand
- where the channel does not match the customer
- what I should change before running it
Be direct. If this test will not teach me much, tell me.
Before you move on
| Check | What to look for |
|---|---|
| You asked "would you use this?" | Ask about the pain, workflow, and current workaround instead. |
| The signal is too weak | Likes and compliments are not the same as intent. |
| You tested with the wrong people | Feedback from people outside your starting segment can be misleading. |
| The message starts with the feature | Lead with the pain so people can recognize themselves. |
| You ignored silence | No response is still a signal. It means something needs work. |
| You moved on too fast | If the signal is unclear, talk to more real customers before building. |