I’ve seen founders create beautiful products for problems that people don’t have, and I’ve seen founders create ugly products for problems that people desperately want solved. The second group makes money. The first group writes post-mortems.
The difference is demand. Not manufactured demand — you can’t convince someone they need something they don’t need, at least not for long. Real demand — the kind that exists before you show up and will continue to exist whether you build something or not.
Demand testing is the step between “I think this is a good idea” and “I’m going to build it.” It answers one question: are people actively searching for, paying for, or working around a solution to this problem right now?
The Four Types of Demand Signals
Not all demand looks the same. Here’s how to recognize it.
Signal 1: Search demand. People are typing queries into Google looking for a solution. You can measure this directly using Google Trends, Ubersuggest, or even Google’s autocomplete suggestions. If people are searching for “[your category] tool” or “how to [thing your product does],” search demand exists.
Quantity matters: 1,000+ monthly searches for a relevant keyword indicates meaningful demand. 100 or fewer might not sustain a business through search alone — though the problem might still be real, just not expressed through search behavior.
Signal 2: Spending demand. People are already paying money for solutions in this space. Competitors exist. Products are being purchased. Services are being hired. This is the strongest demand signal because money has already changed hands. Your job isn’t to create a market — it’s to serve the existing market better.
Customer interview techniques help you validate spending demand: “How do you currently handle this? How much do you spend on it?” If people are spending real money on bad solutions, the demand is proven.
Signal 3: Complaint demand. People are actively complaining about the current state of things. They’re writing negative reviews of existing tools. They’re posting frustrated messages in forums. They’re venting to colleagues. Complaint demand tells you the problem is emotionally charged, which predicts willingness to pay for a better solution.
Signal 4: Workaround demand. People have cobbled together makeshift solutions using tools not designed for the purpose. Spreadsheets being used as project management tools. Email being used as a CRM. Post-it notes being used as a kanban board. Workarounds are demand expressed through behavior rather than words.
The ideal opportunity shows all four signals: people are searching for solutions, spending money on bad ones, complaining about the experience, and building workarounds. When all four are present, demand is not a question — only the right solution is.
The 48-Hour Demand Test
Here’s my rapid demand testing protocol. It takes 48 hours and costs nothing.
Hour 1-4: Search demand research.
Open Google Trends and enter 3-5 keywords related to your problem space. Is the trend flat, rising, or declining? Rising means growing demand. Flat means stable demand. Declining means you might be entering a dying market.
Then check Google autocomplete: type the beginning of a relevant query and see what Google suggests. These suggestions are based on actual search volume. If Google suggests exactly what you’re thinking of building, that’s demand.
Hour 5-12: Spending demand research.
Find every competitor in the space. Use Google, Product Hunt, G2, and Capterra. For each competitor, note: how long they’ve been around, their pricing, their customer count (if available), and their most recent updates.
If multiple competitors have survived for years and are charging real prices, spending demand is confirmed. If no competitors exist, be cautious — it might mean nobody wants to pay for this, not that you’ve discovered an untapped market. Empty markets are usually empty for a reason.
Hour 13-24: Complaint demand research.
Search Reddit, Twitter, and relevant forums for complaints related to your problem space. Use searches like “[problem] is frustrating” or “[competitor] sucks” or “I wish there was a tool for [thing].”
Catalog the complaints. Note the language. How many different people are expressing similar frustrations? The volume and emotional intensity of complaints predict the urgency of the demand.
Hour 25-36: Workaround demand research.
Ask 5-10 people in your target market: “How do you currently handle [problem]?” Listen for workarounds — manual processes, misused tools, or hacked-together systems. Each workaround is proof that the demand exists but the right solution doesn’t.
Hour 37-48: Synthesize and decide.
Score each signal type on a 1-5 scale based on what you found. Sum the scores.
- 16-20: Strong demand. Proceed to pre-selling or MVP.
- 10-15: Moderate demand. Needs further investigation through customer conversations.
- Below 10: Weak demand. Consider a different problem or a different market.
When Demand Tests Mislead
Demand tests can produce false positives and false negatives. Know the failure modes.
False positive: Demand for free, not for paid. People might search for and complain about a problem but refuse to pay for a solution. This is common in consumer markets where alternatives are free (even if terrible). The demand exists but the willingness to pay doesn’t. Revenue is the only real validation, and demand testing is a proxy, not a proof.
False positive: Demand from non-buyers. The people complaining online might not be the people who make purchasing decisions. In B2B, the end user and the buyer are often different people. Demand from users without buying authority is a weaker signal than demand from decision-makers.
False negative: Hidden demand. Some problems are so accepted that people don’t complain about them, search for solutions, or spend money on alternatives. They just… endure. These hidden demands are real but invisible to the 48-hour test. They usually surface only through deep customer conversations where you ask about behavior rather than frustration.
False negative: Demand under the wrong keyword. If you’re searching for “AI scheduling assistant” but people are searching for “stop having so many meetings,” you’ll miss the demand because your keyword doesn’t match the customer’s language. Always search using the customer’s words, not your product terminology.
The demand test is a filter, not a verdict. It tells you whether the signal is strong enough to justify spending time on deeper validation. It doesn’t tell you definitively whether a business will succeed. That requires the next step: actually putting an offer in front of people and seeing if they buy.
The Demand Landscape Map
I use a visual tool to synthesize demand research. It’s a simple 2x2 grid.
Vertical axis: Problem urgency (from “mild inconvenience” to “hair on fire”) Horizontal axis: Solution satisfaction (from “well-served by existing options” to “completely unserved”)
Plot the problems you’ve identified on this grid based on your demand research.
The top-right quadrant (high urgency, low satisfaction) is where opportunities live. Problems that are urgent and poorly served represent the strongest demand. The bottom-left (low urgency, well-served) is where you want to avoid building.
Most ideas fall in the middle — moderate urgency, moderate satisfaction. These can work but require sharper positioning. You need to serve a specific niche where the urgency is higher than the average, rather than trying to serve the whole moderate-middle market.
Key Takeaways
- Demand isn’t created, it’s discovered. Test for demand before building. The four signals: search demand, spending demand, complaint demand, and workaround demand.
- The 48-hour demand test costs nothing and provides directional clarity. Score each signal type on 1-5 and sum. Below 10 = weak demand. 16-20 = strong demand.
- Watch for false signals. Demand for free isn’t demand for paid. Demand from users isn’t demand from buyers. And hidden demand won’t show up in search or complaints.
- Use the demand landscape map. Plot problems on a urgency vs. satisfaction grid. The top-right quadrant (urgent + unserved) is where the best opportunities live.
- Demand testing is a filter, not a verdict. Strong signals justify deeper validation. Weak signals justify moving on. The test saves time; the customer conversation provides certainty.
