The idea sounds useful.
What would prove it?
When a new service idea takes shape, it is easy to picture the app that could run it. Before building, there is a smaller decision to make: which belief about customers or delivery needs evidence first?
You might need to learn whether people want the service, whether they will take the next step, or whether your team can deliver it well. Each question needs a different test.
Name the uncertainty
Write down the one thing that would make the project fail if it turned out to be wrong. Then choose a test that puts that belief in contact with real behaviour.
Do people have the problem?
Listen for how often it appears and how they handle it today.
Possible test: speak with people who recently faced it.
Will they take a step?
Interest in a conversation is useful. A request, booking, or return visit tells you something more.
Possible test: offer a small version to a defined group.
Can you make it work?
A service needs a reliable way to fulfil the promise, even when the first attempt is manual.
Possible test: deliver it for a few customers and record the work involved.
How Buffer tested before building
Buffer is a public example. It is not a Sproutzai client project.
Buffer tested interest before finishing the product. Founder Joel Gascoigne says he paused development and put up two pages to see whether people would consider using his social posting tool. He added a pricing page to see whether people would click a plan, and took feedback from the people who left their email. He built the first working version after those tests. Read his account ↗
Buffer’s pages tested interest and price. That did not prove every part of the business. It did tell the team what to learn next.
Design a test that gives you a decision
Before you commission software, try this short exercise with your team:
- Write the promise in one sentence.
Describe what a customer will be able to do or receive, without listing app features.
- Choose one question to test first.
Pick the uncertainty that could change the whole plan. Demand, action, and delivery are separate questions.
- Make the smallest real offer.
Use an existing page, a simple form, or a manual process if it lets someone genuinely take the next step. Be clear about what is available now.
- Set the decision rule before you begin.
Decide what evidence would make you continue, change the offer, or pause. Record what people do and what it takes your team to deliver.
Build from the evidence
A test does not have to look like the final product. It needs to answer a useful question honestly. When the evidence shows which part is worth building, the first version can be smaller and more purposeful.
