Got a good idea? Test it!

You left another year behind in the blink of an eye, and you entered 2019 with fresh ideas. Do you want to get to the physique you dream of with a gym membership and a planned regular training program? Or, quitting your job to which you feel increasingly estranged, and bring that business idea to life?

The characteristic feature of these two scenarios is the need to make certain investments or sacrifices to achieve the goal. Most simply, you have to pay six months or 1 year of a gym membership on the first day. 

To pursue your dream of entrepreneurship, the free time left over from your work hours most likely will not be enough, and you will have to resign from your work. Before you take these steps, wouldn’t it be nice to know if you can actually go to the gym regularly or whether your business idea has considerable potential?

Let’s look at the subject in the context of brands and products/services they offer to their customers. Suppose that an innovation competition was held in the company, and creative ideas emerged. Is the decision of the jury enough to determine which idea is worth investing? Alternatively, is it possible to proceed with a test and data-driven approach and determine the best idea for investment?

Yes, it is possible!

It all started with an innocent vote

In 1999, three youngsters put two large trash cans with the words Yes and No next to their smoothie booth at a UK music festival and asked the participants: Should we leave our job to make these smoothies? At the end of two days, the bucket that said yes was filled to the brim. This is how the story of Innocent, which has become the most successful smoothie brand of England and perhaps the world today, begun.

The test which was conducted by the trio of graduates of the University of Cambridge, Richard, Adam, and Jon can be described as “guerilla” testing. The guerrilla test method is a relatively low-cost method that allows you to get quick feedback on your product or prototype. The most classic example is to grab people who you think are involved in your target audience and get feedback from them.

If you are reasonably assertive or skeptical, the test result in the Innocent sample may not be enough for you to make the final decision. Rather than blindly embarking on the adventure, checking the pulse of your potential customers can benefit you in your decision making.

First verify the problem

“A problem well stated is a problem half solved”

Charles Kettering

Most of the monologues that start with the phrase ‘I have an excellent idea,’ directly address the solution without mentioning the problem. On the other hand, solutions are created and exist to solve a problem. Therefore, before verifying your opinion, it is necessary to test the problem that your solution aims to solve. 

What if the problem you wholeheartedly believe exists isn’t received well by your target audience?

Let’s look at an example. A bank whose motto is to make life easier has a new product idea. This product brings convenience to payment processes for real estate and car sales. This is done through a system integration, which triggers the transfer of money and the transfer of deed-license simultaneously. In this way, risks such as cash transport of large amounts and the possibility of deed transfer without the payment transfer no longer exist.

That sounds good, doesn’t it? However, what if we told you that a significant number of people who sell a house choose to do it in cash and declare a lower house value to receive a tax advantage? Or that when the bank loan is used, the loan amount is transferred to the seller automatically with the transfer of the deed? 

When you think that your target audience does not have a significant problem in this regard, doesn’t the solution begin to lose its appeal?

How to verify a solution

Let’s move on to a parallel universe and assume that the research results confirm the problem. The target audience really feels the risks we are talking about and is in search of a solution.

In this case, it is time to test how much demand there is by the target audience to the bank’s solution. The most significant benefit for running this test is that it offers a chance to verify (or kill) the solution without substantial investment.

How to test a product that doesn’t exist yet? By doing “as if”

This method, which is referred to as fake doors in English, is frequently used in idea verification processes. Imagine advertising on real estate sites as if the product is ready. The users who click on this advertisement will read this message: ‘Our product will be implemented very soon. Thank you very much for your attention and patience!’ Thus the first step of the test is completed. By analyzing the ad’s clickthrough rates, you get the chance to gain insight on current demand.

Finally, let’s emphasize that the following 3 points should be taken into consideration in the problem and idea verification studies:

  1. Correct identification of target audience.
  2. The correct design and implementation of the research.
  3. The confidence interval of the research result is sufficient to make a decision.

Developing and testing a hypothesis

The concept of testing is a critical part of ‘design thinking,’ which is to consider the logic of the design, and it has gained much popularity recently. The cycle that constitutes this methodology consists of researching and defining the needs of the user, producing the solution corresponding to this need, testing the prototypes if possible, and bringing it to life after applying the changes from the feedback. Based on this cycle, we can say that the testing phase is the starting point of improvement.

The diagram was designed by Nielsen Norman Group.

In particular, brands operating in the digital environment use this methodology to improve the experience which they offer and increase the transformation of their digital assets.

Let’s say the product team of an e-commerce website that sells organic food products thinks that sharing positive customer comments on the home page will increase the rate of users viewing product category pages.

To test this hypothesis, first, determine the metric to be measured and check the current value for comparison. For the test result to be considered successful, the target percentage increase is determined. Then the test phase begins, and the alternative design with user comments starts to be shown to 50% of the traffic. 

After reaching a sufficient number of visits which can grant confidence, the results are compared, and it is decided whether to proceed with the current design or the new one. This setup is called the A / B test. It is also possible to test more than two variants at the same time, or even run multiple tests at the same time.

When the domain name Amazon.com is pronounced today, a single entity is revived in everyone’s mind. However, Amazon has simultaneously done so many tests that it is not possible to talk about a single Amazon.com.

To sum it all up

The test-driven approach allows you to make decisions based on data and reduce the risk you take instead of basing on your intuition, which is more than likely to mislead you.

Nowadays, especially with the decreasing cost of digital test methods and tools, the applicability of this approach has increased.

Internalizing the test logic and making this approach part of your thought processes, whether in business or your private life, will benefit you to the extent that you cannot imagine. Of course, do not quickly accept my hypothesis, test it first!

Further reading