If there's one thing to be sure about marketing, it is the fact that it can burn a lot of money. A Gartner study reveals businesses spend roughly 12 percent of their revenue on advertising and marketing.8 That's $1.20 out of every $10 they make! Luckily, marketing doesn't always have to be wasteful. In fact, if marketers know where and how to invest their money, they can save the company tons of unwanted costs while maximizing ROI. This is the reason for experimental research. Let's dive in and learn all about this research method.
Experimental Research in Marketing
Before we learn what experimental research means, we must first understand why it is adopted. To do so, let's go back to two fundamental concepts: Mass marketing and Targeted marketing.
Mass marketing is a strategy where companies ignore the differences between market segments and try to reach as many people as possible.
Unfortunately, this is rather expensive and only affordable by prominent firms. For smaller brands, a better way to access customers is through targeted marketing:
Targeted marketing means breaking the market into smaller segments and focusing marketing efforts on them individually.
However, even when companies know what customers to target, it remains a challenge to capture their attention. How to know what the audience is interested in and craft a marketing message that will appeal to them? This is where experimental research comes in.
Experimental Research Definition
Experimental research is a test of which marketing strategy is most likely to work. But rather than guesswork, it uses real data from experiments.
Experimental research involves making hypotheses about what marketing activity is likely to appeal to customers and collecting data to see if it is true.
One key feature of experimental research is that it explores the relationship between independent and dependent variables.
- Independent variables are variables that do not depend on other variables.
- Dependent variables are variables that change as independent variables change.
In experimental research, the independent variable is the cause and the dependent variable is the effect. As a marketer, you can influence the cause to observe changes in the effect, but not the other way around. The effect can only be tested or measured.
This is why experimental research is so powerful. It not only allows you to see the immediate results of a marketing decision but also manipulates the cause to influence the result.
When you're confused about independent and dependent variables, ask these three questions:
| Types of variables | Independent | Dependent |
| Can the variable be manipulated or controlled by the researcher? | Yes | No |
| Does the variable cause an outcome? | Yes | No |
| Is the variable a result of another variable? | No | Yes |
Table 1. Independent and dependent variablesRemember earlier we mentioned experimental research examines the relationship of independent and dependent variables, can you guess what this relationship is called? (Hint: an independent variable is a cause and a dependent variable effect.)
Did you say it is a cause-and-effect relationship? Spot on! We don't have to dive deeper into this right now. Just keep in mind that in a cause-and-effect relationship, one event takes place before another, and the second event can't happen before the first. For instance, you must add or remove a menu item to see how it affects sales revenue.
Experimental Research Example
Suppose you start an ice cream stall in the summer and decide to try out different ice cream flavors to see which one performs best. In this case, the ice cream flavor is the independent variable, and the ice cream sale revenue affected by the change of flavor is a dependent variable.
Note here that the independent variable is the cause while the dependent variable is the effect. It is the change in the ice cream flavor that leads to the change in ice cream sales. While you can control which ice cream flavor to sell, you can't know which flavor will bring you the most profit without experimenting. This brings us back to a point early on: A cause can be manipulated to test or measure the effect.
Types of Experimental Research
When it comes to experimental research, there are three main types: controlled, manipulated, and random.
Controlled experimental research - Research where all outside factors are kept constant. Only the measured variable is changed. For example, a hamburger store changes its packaging while keeping everything else (ingredients, flavor, etc.) the same to observe the effect of the new packaging on sales.
Manipulated experimental research - Research where you can change the independent variable to measure the effect on the dependent variable. For instance, a bakery changes the amount of flour in bread and sees how customers respond.
Random experimental research - A combination of the above two. For example, you add a new drink to the menu of all coffee stores you own, with everything remaining the same. If you randomly check a store, you will see that the sales may go up or down depending on the location.
The experimental research method may be different, but the idea is always the same, to find out the strategy that helps the business improve its performance.
Descriptive vs Correlational vs Experimental Research
Have you heard of descriptive and correlational research before? If you have, you might also know that they are human behavior research methods just like experimental research. Don't worry if they can't tell the difference between them. This section will help clearly distinguish these three types of research methods.
Let's start with the definitions:
Descriptive research is the type of research that provides an accurate description of a situation. As a result, it is often used to discover information, make predictions, and test a hypothesis. The only drawback is that it does not explain the relationship between different variables.
Correlational research is also used to test hypotheses and make predictions. However, unlike descriptive research, it allows the researcher to observe the causal relationship between variables.
Experimental research is similar to correlational research but goes one step further by letting the researcher manipulate the variable to see different results. 1
Here's a more detailed comparison table:
| Descriptive | Correlational | Experimental |
Goal | Provide an accurate description of what is going on.2 | Establish a relationship between variables.1 | Understand the causal relationship between variables.3 |
Uses | | Test hypotheses. Make future predictions.
| |
Feature | Do not assess the relationship between variables. | Assess the relationship between variables without manipulation. | Assess the relationship between variables, with independent variables being manipulated and dependent variables being observed. |
Examples | Who is buying the company product? To which age group do they belong? How much do they earn per year? | What is the relationship between ice cream sales and temperature? | How do customers' reactions change as the restaurant tries out different sauce recipes? |
Table 2. Descriptive vs Correlational vs Experimental Research. Source: Openpress.
Difference between Correlational and Experimental Research
Correlational research shows only the association between two variables, not necessarily the cause-and-effect relationship between them. Experimental research, on the other hand, allows the marketer to manipulate the independent variable to measure its effect on the dependent variable.
Correlational research only answers the question: Is there a positive or negative correlation between two variables? Experimental research goes one step further by measuring the impact of this correlation.
Suppose a restaurant wants to observe the effect of a new sauce recipe on sales. If using correlational research, the restaurant owner can only observe the relationship between one type of sauce and sales at a time. With experimental research, the owner can try out different types of sauce for a week and see which one drives the most sales and adds it to the menu.
Quasi-Experimental Research
Marketers use quasi-experimental research to understand the change in customer or firm behavior. According to Campell, this is the type of research where the "data generating process is not intentionally experimental" but caused by an external shock. 6
In quasi-experimental research, an external shock causes a variation that the researcher will use to study its impact on a situation.
Quasi-experimental research of eBay: The company paused advertising on Bing and saw little traffic loss. This inspired a follow-up experiment where the company randomly stopped paid search advertising and discovered similar results. In this example, the Quasi-experiment was used to study the consequence of the company's actions.7
Experimental Research - Key Takeaways
- Experimental research conducts experiments to determine which marketing activity appeals to customers.
Experimental research is based on actual data and real-life situations rather than guesswork by the researcher.
Besides experimental research, there are two other approaches to studying human behavior - descriptive and correlational research.
Experimental research differs from other types of research in that it studies the causal relationship between independent and dependent variables.
Quasi-experimental research involves researching the impact of a variation caused by an external shock on a situation. Marketers use quasi-experimental research to understand the change in customer or firm behavior.
References
- Charles Stangor & Jennifer Walinga, Psychologists Use Descriptive, Correlational, and Experimental Research Designs to Understand Behaviour, n.d., https://openpress.usask.ca/introductiontopsychology/chapter/psychologists-use-descriptive-correlational-and-experimental-research-designs-to-understand-behavior/
- Know This, Descriptive Market Research, n.d., https://www.knowthis.com/planning-for-marketing-research/descriptive-market-research/
- Know This, Causal Market Research, n.d., https://www.knowthis.com/planning-for-marketing-research/causal-market-research/
- Stangor, Research methods for the behavioural sciences (4th ed.), 2011.
- Pritha Bhandari, Independent vs. Dependent Variables | Definition & Examples, 2022.
- Campbell, Roy H., A Managerial Approach to Advertising Measurement, Journal of Marketing, 1965.
- Blake, Thomas, Nosko, Chris, Tadelis, Steven, Consumer Heterogeneity and Paid Search Effectiveness: A Large-Scale Field Experiment, Econometrica, 83 (1), 155–74, 2015.
- Tanya Castaneda, How Successful Companies Spend Their Advertising Budget, 2018, https://mediamaxnetwork.com/blog/how-successful-companies-spend-their-advertising-budget/.
Explanations 
Exams 
Magazine 
