Imagine your boss asks you to develop new services to improve customer loyalty. A few ideas have come to your mind that you think will probably have a positive result. The problem is that you are not sure which one will have the most impact and you do not have enough time and resources to implement all your ideas. So, which idea should you choose? In such a situation, the use of commercial experiments will help you. By implementing this method, you can try ideas on a small scale so that you do not have to spend a lot of time and resources on large projects. In this article, we will talk about how to run business trials and see how you can learn important tips from them.
Many businesses will not grow without innovation, and innovation is created by developing and implementing new ideas. However, every innovation has its risks: the initial hypothesis may be flawed, new products may not work well enough, customers may not accept new ideas, and problems may arise.
That is why it is useful to run commercial experiments. By testing the ideas, you can see how they will work on a larger scale. You can also learn from your failures and mistakes and correct your ideas.
Commercial trials involve many steps. From basic tests (for example, changing the welcome message on your site page or testing a new process for answering phone questions) to more complex projects (such as testing a new product or service with a small number of customers).
Business and risk tests
In many cases, commercial trials can be associated with many instabilities and risks. Because most of the time there is no way to know if they are useful or not.
However, the point is here. When you do business trading, you put yourself at risk in a controlled way. While you certainly want to avoid failure, you are not really afraid of failure, because this way you can avoid more costly mistakes in the future and make the most of opportunities.
In the short term, commercial trials may seem so small and dangerous that they are not worth the effort. However, if you think about the benefits they will have in the long run, you will come to the conclusion that sometimes not doing them will be more costly than doing them. If you learn important ideas from both successful and failed ideas, you will come up with ideas that will help your organization grow and prosper.
Depending on what your role is in the organization, you may need the permission of your boss or manager to perform this test, even when the risk is very low.
If you need help with your experiment, read our article on sales skills for non-sellers and our short tutorial on selling your ideas.
Commercial trials are not appropriate in all situations. For example, it may not work when you are dealing with a very small market, or when it is very quick and easy to achieve success. Here, you have to make a decision based on your instincts and the information you have.
Before conducting a business test in a real-world environment, it may be best to test your ideas by interviewing users or customers through focus groups or polls.
How can an effective experiment be set up?
To set up an effective business experiment, you can use the approach that we will introduce below. This approach is based on a process developed by the innovation expert Thomans H. Designed, created by Thomas H. Davenport.
1. Make a hypothesis
Just like scientific experiments, the first step in this process is to formulate a hypothesis. Hypothesis is your prediction of what will happen if the experiment is successful. For example, “Sending two additional shipments per month will increase overall revenue” or “Designing a menu on the right side of the website will keep more users on the website.”
As you formulate your hypotheses, consider the following questions:
- Can you measure what will happen after the change? (You must be able to reject or accept the test based on the hypothesis)
- Is this an experiment with the overall strategy, goals, and values of your team in the organization?
- Does it add value to what you do?
If your answer to any of these questions is no, you need to prepare yourself to change the hypothesis or not to test.
Finally, in this step, specify the methods by which you measure the success of the experiment. Ask yourself, “What is success?”
What needs to be measured depends on your industry and the type of testing you do, and can include metrics such as sales or revenue, number of new visitors, number of visitors to the website keywords, or more general factors such as ease of use, or customer and member feedback. Be a team.
Here, too, it is better to think about how each of these scales affects the overall goals and strategies of your organization. For example, you may find success in “increasing website traffic.” However, if this causes customers to visit the online store less and ultimately leads to a reduction in overall revenue, it will no longer be considered a success.
2. Design your experiment
Now you have to think about how you can run your business test. Draw an outline of the items to be tested and the duration of the test.
Testing a new product or process is relatively easy, you just need to run the items for a certain amount of time and then measure the success of the test using the measurement scales mentioned in the previous step.
However, if you want to change an existing product or process, you must use “like for like” scenarios to be able to measure the success of your hypothesis. Scenarios based on a comparison of the general situation over a period of time, before and after the change. You can use “control group” and “group therapy” to do this. (This test is also known as separation or A / B.)
To get started, control as many different variables as you can. This means that you need to minimize change as much as possible.
Your control group is what you use for basic measurements. Try not to change this group as much as possible during the test. Do the same with the group therapy, ie do not change any other conditions in this group except the one you want to test. You can then compare the control group with the group therapy to test the hypothesis.
For example, suppose your hypothesis is that “Retail stores open until 10pm will increase profits.” To test this hypothesis, you have to choose two stores as a group therapy that will be open until 10 pm. Your other stores will be considered as a control group, meaning they will not change their working hours. Then, after two weeks, you compare the revenue of these stores to see if the opening of the stores until 10 pm has an effect on increasing profits.
In some cases, using the control group and the treatment group at the same time will not work. (For example, if you only have one store, you can not do the above test.)
When this happens, you can use data that you have carefully collected from the past and compare it to the information you get after making a change. In the example above, if you only have one store, you can compare this year’s sales with last year’s sales.
In general, keep your experiment as simple as you can. The more complex the test, the more costly it will be, the more risky it will be, and the more time you will have to analyze the results. However, the test should be such that you can be sure that it will give you reliable results.
When using control and group therapy, try only one factor at a time. For example, if you want to check for longer store hours and brighter ones, you should try each one separately to be sure of the results.
Use tools such as risk analysis and risk probability / impact charts to better understand the risks associated with trading trials.
If you want to try a product or service for the first time, do not underestimate the effort you need to make. If you produce a product that is below the standard, your test is likely to fail.
3. Perform the test
Once you’ve designed your business test, it’s time to get started and implement it.
Before doing anything, let everyone in the organization know who is involved in the experiment. They need to know the reasons for this test.
As soon as the test process begins, monitor it closely to make sure nothing else is affecting your findings. For example, in the scenario mentioned above, factors such as bad weather can affect sales. If this happens, repeat the experiment or remove some of the data from the final findings.
If you are going to test a new product or service with customers, you need to tell them that these are trial products. That way, if the product is not as good as it should be, your reputation will not be damaged. You also encourage customers to give you feedback.
If the test product has a negative effect on the people who used it, you must compensate them.
4. Analyze the results and perform additional tests if necessary
After your test is over, take some time and carefully review the results.
First, evaluate the actual performance with your hypothesis:
- What was going to happen and what really happened?
- Why the difference?
- Was the experiment successful? (If not successful, do not despair)
- What were the unexpected results? How did you manage to get the most out of them?
It’s time to think about how you can benefit from what you have learned. Are other experiments useful for achieving more goals than that idea?
For example, does opening a store until 9 pm instead of 10 pm increase profits more? Or will making more changes to the products be met with good feedback from customers?
In some cases, you can do other experiments using the new material you have learned. Sometimes, you will have to do additional tests to get the best results.
There are several statistical tests to analyze the results of your business test. When using them, remember that it is not easy to control the various factors that affect the outcome. This means that the degree of reliability in commercial tests is usually much lower than in-laboratory tests.