Testing Business Ideas: 4 most common tests to get you started

Great innovation teams are able to gain momentum and build up stronger evidence over time with a series of experiments. In this blog post, we share one four-step example of how to execute and analyze multiple experiments in order to increase the strength of your evidence. 

We believe our customers have a problem...

At the start, all you have is a business idea, something that you think will address a particular problem or pain point. You may have a basic inkling of who your customer segment should be and have even created a Customer Profile of them, but how do you know if you’re right? The first thing you should do is get out of the building and talk to your prospective customers. Customer Interviews are ideal for quickly gaining initial qualitative insights into whether the problem you imagined for your customer segment exists and if the value proposition you’ve designed resonates with them. 

At this point, your evidence strength is very weak because people are just telling you what they may or may not do. However, it is very useful to capture some initial insights especially when you know nothing. We typically recommend a minimum of 15-20 interviews. 

What you can do: 

• Interview in pairs to ensure one person is actively listening and asking following up questions while the other person takes good notes. 
• Do a quick debrief after each interview to see what did and not work. This is a good opportunity to refine your interview questions and style to improve your next interview

What not to do: 

• Don’t ask people what they want. It's not their job to tell you what they want and they may not know what is possible. Focus instead on understanding the job they need to get done and how they accomplish this work.
• Don’t try to “interpret” what people say. Instead, write down quotes verbatim, then analyze these, looking for patterns in what people say. Typically you may start to see patterns after 15-20 interviews. 

We believe this problem exists at scale…  

After gaining some qualitative insights from interviews, you should next aim to test your value proposition quantitatively at a larger scale. One way to do this is by spending a few hundred dollars a day on Google Display, Google Search, and Twitter Ads. This is a good way to test at scale if a particular pain or gain of your customer resonates stronger by creating two versions of the same ad, also known as Split Test. We have explained in detail how you can create an online ad using templates. 

What you can do: 

• Don’t forget your value proposition canvas! You can pull your ad copy directly from all the hard work you’ve already done to develop your value proposition canvas. 
• Try your ad on multiple channels to ensure you’re reaching the right customer segment..  
• Use this as an opportunity to build up your list of early adopters. It is always useful to go back and test and interview them. 

What not to do: 

• Don’t develop the ad in isolation. Be sure you tie it back to a hypothesis you are testing and whether the results will support or refute that hypothesis. Keep your eye on the prize. 

We believe our customer is actively looking for a solution to their problem… 

Each of these testing steps should build upon the strength of evidence from the previous test. Interviews were good at capturing first insights but what people say and what they do are often very different. It is important to follow up with stronger evidence by asking people to do something, whether it is buying something, using your service, or making an investment of time or other resources. 

At this point the uncertainty of your idea is still very high, you don’t want to waste a lot of resources building something. The Concierge is a test to create a customer experience but manually without having to build anything. It is not possible to scale, since everything is done manually but it can give you firsthand learning on what is needed to create, capture, and deliver value to a customer. 

What you can do: 

• Prepare all the steps needed to create the product or service manually
• Integrate web page analytics if you’re taking orders online. This will also help you determine the conversion rate. 
• Keep your eye out for patterns. The point of this test is to gather statistically significant data points and this means seeing if any patterns emerge from how people are interacting with your minimum viable product (MVP). 

What not to do: 

• Take on more orders than you can handle. If you are unable to deliver your value proposition to your customer then you won’t be able to collect any useful evidence. 

We believe this feature will solve my customers' problems… 

Typically when designing your value proposition, you may have many features you want to include in your value proposition but you may not know where to start or if your customer is even interested in all of them. Creating a single-feature minimum viable product (MVP) is a great way to test whether a single feature addresses your customer problem. It will take a lot less time and resources as well to build out your entire value proposition. Releasing features one at a time also allows you to isolate and test a particular hypothesis. 

What you can do: 

• Design the smallest version of your feature that solves for your customers most important job.
• Use the prospective customers you have acquired from your previous Online Ad. 

• Always design your tests in a way that will allow you to track the specific data you will need to support or refute a hypothesis. This way, if your product or service picks up traction, you’ll know why.  

What not to do: 

• Don’t build a minimum version of your product. Creating a MVP simply means a minimum viable option to allow you to test a particular hypothesis. There is a difference.
• Do not let the people who did not follow through get away. Always try to find a way to capture feedback from them. Learning about why it didn’t work is just as important as knowing what did work. 

Creating an experiment sequence does not mean it is a linear process. At every stage of development always try to find a way to get feedback from your customer to make sure you are addressing a problem they have. Be sure to increase the strength of evidence by getting your customer to do something but there is nothing stopping you from conducting interviews along the way to capture those insights.