Testing a product, service, or business model, with an example from the Democratic Republic of the Congo

A manager often wants to test a proposed product or service. For example, they may want to find out whether their customers will like a product, or they may want to know what the best material for it or how large it should be. They could just make an educated guess, which may sometimes work particularly if they have a great deal of knowledge in the area, but it would be risky. The product’s or service’s development and launch could be expensive, and the manager’s organisation could incur large losses if their guess is wrong. A test is a situation and set of activities set up by the manager to provide information about something of interest, and by using this information to make decisions a manager can reduce the risk from their subsequent actions.
 
A test may also concern a business model for a proposed product or service, which describes how the manager’s organisation will operate and make money in respect of the product or service. For instance, the test may try to find out about potential retailers of their product, or about the response of competitors to the product’s launch. Again, a test can reduce the risks associated with the innovation project.
 
This blog post will describe how the test can be broken into steps to help ensure that it delivers valuable information. An example from the Democratic Republic of the Congo (D.R.C.) will illustrate their application.
 
Step 1: State what information is required
A manager may want to evaluate something, such as the likely performance of a design or the accuracy of an assumption in a business model. Alternatively, they may want to determine something, such as the best number of employees to deliver a service. With a clear statement of this required information, a test can be better targeted to gather it and not irrelevant information that may confuse analysis.
 
Step 2: Separate the required information into small pieces
If the manager wants a large amount of information, they could run a single test to collect all of it at once. But that would often be inefficient, because coordinating a large test can be difficult and expensive, and the resulting information may be hard to interpret. Instead, they could divide the information into smaller pieces before collection, and then run a series of small tests each gathering information about one of these pieces at a lower overall cost and effort than running one large test. The manager should be able to find the pieces of information separately from the other pieces; for example, information about the performance of a planned restaurant could be divided into information about the performances of its menu, its location, its staff, and its suppliers.
 
Sometimes it is not possible to divide up the required information. For example, a manager may want to test the performance of a new product on a market before full launch, and in that case they may simultaneously collect information about production and sales of an advanced prototype of the product.
 
Step 3: Plan the test(s)
The manager should next plan the tests to collect the information at the lowest cost and in the shortest time. A test describes a situation and a series of actions, and the manager should choose only those are necessary to generate the information, and no more. For example, if the manager wants to know whether a product’s material will be strong enough when it is used, they could take a piece of material and apply weights to it until it breaks. They can then compare the breaking point with the typical loads that the product bears in use. In some cases, a manager may not need to build or set up anything physical at all; they could just use a written or computer model to see what is likely to happen to their product, service, or business, and collect information that way.
 
Often, the same situation and actions can be reused with small adjustments to collect similar pieces of information. For example, when a manager wants to know which colour is preferred by consumers, they can reuse a survey group that was previously used to collect information on which size they prefer. Iteration may sometimes be used, such as when the manager makes small changes in a product or service following consumer recommendations and then presents again to the group to see whether they would like any additional revisions.
 
The manager should also consider what exact information they will collect from the tests. The collected information should correspond closely to the information that they require. For example, if the manager wants to know how many employees will be required to provide a service, they could measure the performance of a trial service with different numbers of employees, and the satisfaction of those employees.
 
Step 4: Assess the cost of the planned test
The next step is to assess the cost of the planned test. Some simple tests may only be very low cost, for example using a paper or foam model, or presenting some consumers with a written description of proposed service. Other tests may be much more expensive, for example setting up a trial production facility to produce something that closely matches the final product. The cost of the test could include the costs of design, development, materials, labour, production, and marketing, which can be itemised in the total cost.
 
Given the cost of the test, the manager can then ask whether it could be reduced. The itemisation is useful here, since the manager can try to reduce the costs of the different components. If the manager recognises a way of reducing costs, they can modify the plan so that cost is as low as possible. The final cost should be reasonable, considering the value of the information gained and the risk of not having it.
 
Step 5: Execute the planned test
Once all the other steps are complete, the manager can execute the test as planned and collect the required information.
 
Example: A test for a new dress in Haut-Katanga
A company is proposing to introduce a dress with a new pattern and material in its shops in Likasi and Lubumbashi, in the Haut-Katanga province of the D.R.C. The company’s manager would like to test whether customers will buy the new dress, and they apply the above steps.
 
Step 1: State what information is required
The manager’s broad question is whether customers will buy the new dress, but they sharpen it to ensure that the information is more valuable in product development and decision-making. The precise information that they require is the number of purchases of the new dress at a likely price for it.
 
Step 2: Separate the required information into small pieces
The manager divides the required information into parts. They want to find out the number of purchases for the new design in their Likasi shop and in their Lubumbashi shop separately. They also want to know the number of purchases with the new pattern and with the new material separately. The information will allow them to revise their dress design and make other decisions about the whole project.
 
The initial information that the manager will collect is the number of purchases with the new pattern and a material used in earlier dresses in the Likasi shop. The information that they collect may make the manager change their future information collections – for example, if nobody buys the dress with the new pattern, then the manager may choose to stop collecting information about the current dress, and instead change the pattern before investigating further. The rest of the steps will focus in this single test.
 
Step 3: Plan the test(s)
The manager plans the test. Although the proposed dress will be mass-produced by machine if it goes into production, for the test just ten dresses with the new pattern will be made, and by hand. For such a very small volume, the cost per unit will be lower if hand-made, so the costs of the test will be lower. The new dresses will be placed next to other dresses in the Likasi shop, at the same price as similar dresses. An advert will be placed in the shop window to raise awareness of the new dress’ availability, and sales will be monitored over a two-week period. The number of sales will be counted at the end of the period, as well as the number of sales of other dresses. Although this latter information was not initially required, it is potentially useful in assessing whether the sales of the new dress would replace sales of the old ones, and it can be collected at no extra cost.
 
Step 4: Assess the cost and value of the planned test
The test’s cost is equal to the production cost for the ten dresses, plus some additional costs, minus whatever receipts there are from sales in the two weeks of display. The additional costs are small, as transport, rent, and shop-staff wages are already paid. The total cost will be 340000 Congolese francs (about 120 United States dollars on 27 July 2024), minus an amount between 0 CF (0 USD) and 850000 CF (about 300 USD) depending on sales.
 
Step 5: Execute the planned test
The manager considers the test will provide useful information at a low cost, and executes it.