Is my business idea good? Test the evidence first

Write: “For [customer] in [target country], [problem] happens when [situation], and currently costs [time, money or risk].” Choose a group you can reach. “Everyone who wants convenience” is too broad to test.

Speak with people who fit that description, not only friends. Ask about the last time the problem happened. What did they do? What did it cost? Who approves spending? Ask for concrete examples, without collecting private records. Avoid “Would you use my app?” Record what happened, separately from what you hope it means.

List the customer's current options: a competitor, a spreadsheet, a freelancer, an internal process or simply tolerating the problem. Compare effort, price and switching barriers. Existing businesses suggest something is being offered, not that your version has paying demand.

Use our guide to checking whether a business idea already exists for the discovery steps. Here, focus on why this customer would switch. Write a specific advantage you can test, rather than “ours will be better.”

Test the riskiest assumption first. Interviews test whether the problem occurs. A clear offer page tests whether people take a next step. A manually delivered pilot tests whether they use the solution. An actual purchase tests willingness to pay more directly than an email signup.

Explain what exists and what does not. Do not imply a finished product, invent reviews or hide pilot limitations. For a paid pilot, state the scope, delivery date and refund terms before accepting payment. An inquiry, signup or free trial is useful evidence, but it is not revenue.

Before starting, write the audience, offer, channel, time limit and decision threshold. Count everyone shown the offer, then count inquiries, starts, payments and repeat use separately. Low response could reflect the channel or message, not only the idea. Retest one changed assumption instead of changing everything.

Start with the price a customer would actually pay, not the size of the whole market. Deduct delivery costs, payment fees, expected refunds, support and your own delivery time at a stated hourly rate. Then estimate customer acquisition cost: money spent acquiring customers divided by new paying customers. With no paying customers, that cost is still unknown.

Contribution per sale is revenue minus variable costs. Break-even sales are monthly fixed costs divided by positive contribution per sale. Keep one-time setup costs separate. For subscriptions, also test retention; do not assume every buyer stays indefinitely. Write a cautious case with lower sales and higher costs. Label every untested input as an assumption.

Proceed to another limited test when customers show the problem, take meaningful action and your delivery economics look workable. Change the audience, offer or channel when the problem is real but the response or margin falls short. Stop this version when repeated tests find no meaningful commitment, or delivery cannot work within your time and budget.

Set thresholds before seeing results, so enthusiasm cannot move the goalposts. There is no universal conversion rate that makes an idea good. Your decision depends on price, acquisition costs, capacity and risk. A small positive test supports the next experiment, not a forecast of the whole market.