Should we change our pricing? Should we invest more in our online offerings?
What charitable causes could we support that would best align with our customers' values? Does this concept we are developing make sense?
Businesses are constantly faced with these types of questions. And while you may think you know the answers, or have a good hunch about the best direction to take, the reality is that your customers may not fully agree with your conclusions.
So when these are the types of questions your company is looking to answer, then discrete choice modeling represents a wise choice as a market research methodology that can deliver solid data-driven insights.
Discrete choice modeling can provide better answers than you would likely get from direct, standalone survey questions. That’s because this model allows for a deeper, more detailed dive by looking at several factors within a decision. This forces respondents to make trade-offs among different options, resulting in having to choose what they value the most from the options provided.
What is the result?
Data that offers richer insight into which factors our audiences find most important, thus providing you with data-driven insights to make better business decisions and investments.
So, put another way, discrete choice makes sense when your organization needs to make choices—which is pretty much all the time. Whether you want greater insight on a new product or service, or want to test out a new concept, by forcing your survey respondents to choose, life becomes easier for you because the choices that you have to make are backed by solid data.
Discrete choice models helps answer burning questions
By definition, a discrete choice model specifies the probability that a person chooses a particular alternative, with the probability expressed as a function of observed variables that relate to the alternatives and the person choosing.
This differs from other surveys that typically focus on “how much” questions, such as rating questions that aim to get a sense of the degree to which a survey respondent likes—or dislikes—a particular product or service.
Mountain resort example
So if your business is a mountain resort and spa trying to get a better read on how your guests prefer to get to your facility, a discrete choice question might be “do you prefer to arrive by train, plane or automobile?” In this instance, if the train were to top the list you might look into working a deal with Amrtrak or highlighting the scenic trek that the train takes through mountains and valleys before arriving near your resort.
This provides much more insightful feedback than simply asking survey respondents of how they arrived at your destination for a particular visit. While they might have flown, perhaps they would have much rather taken the train if they were more aware that option was available.
Quickly get the insights you need with survey solutions built specifically to help you make better decisions.
If your aim is to have respondents make choices, the first step in your survey process is to identify what those choices will be. You want to be thinking about factors that have the potential to make whatever you’re asking the best, while keeping in mind any limitations such as budget or complexity that would make something unrealistic to offer.
Let’s revisit our scenic mountain resort example. Perhaps your facility is among many resorts that does the local landscape, and you want to create a spa experience that can separate you from the pack. Based on previous formal and informal feedback from customers, and your keen awareness about trends in the resort spa industry, you have a pretty good inkling of some offerings that your target audience is likely looking for.
Let’s assume you end up with a list of five possible offerings or factors at your mountain resort. These offerings are:
Now that you have your factors or choices that have the potential to create an optimal spa experience for your guests, you can focus on a discrete modeling survey design that poses these choices to your respondents.
After you receive your results you can conduct statistical analysis to provide insight and perspective on the choices respondents made. In this instance, let’s assume the hot trend of full-body immersion in ice water pools is a top choice. With this insight from your survey, you’re now prepared to take action by investing in several ice-water pools and building a marketing campaign to promote them.
Discrete choice modeling can help you address a wide range of issues facing your business. By survey respondents answering the “which one?” question, they help you determine what changes or enhancements represent the best investment of time, resources and money. Below are several examples of how discrete choice modeling can play out.
Discrete choice modeling can be particularly valuable when determining the influence that both price and product features have on how much customers’ are willing to pay for a particular product, package or service.
Let’s say a maker of stationary exercise bikes wants to determine a combination of features and price customers would value most as a means to price a new model so it is attractive to customers, competitive with other options on the market and, of course, profitable.
The resulting survey should generate data that reveals what customers value most and what they are willing to pay for that value—the premium high-tech live instruction, the convenience of easy storage or the comfort of a cushy seat. Note that the options share some attributes, but there are distinctive elements that differentiate each package of features, which forces respondents to choose which they like most and would be willing to pay for.
Let’s say you run a successful coffee shop in a small town and you learn that your top competitor is planning on opening a new location pretty close to your shop. You don’t have the budget to open a new location or expand your current one, but want to make sure you have a strategy to respond. By using a discrete choice model you could assess what might help develop customer satisfaction, and attract some new ones. For example, options might include:
Based on the responses you receive, you should get some clear direction on what addition to your business your customers would value the most—and what might keep the competition at bay.
You’re part of a marketing team at a software company that is developing a new product aimed at helping people manage at-home businesses. The software offers a range of benefits—so many that it’s hard to know which features and benefits will resonate most effectively with your target audience.
To get a clearer idea of which message has the best chance of triggering a buying decision, you ask survey respondents what they would value most in the product. Here’s an example:
For an at home business management system, what would be most important to you?
The nice thing is that the software offers all of the benefits listed as choices—and all of these messages can be woven in some way in your marketing messages and descriptions of the product. Yet by learning what respondents value the most, you can tailor your high-level marketing and advertising messaging to highlight their top priorities.
Imagine you run four fitness centers located throughout a medium-sized city and the COVID-19 crisis has prompted you to re-think your business model. To get insights from members on how you should proceed to grow your business you present two scenarios, both of which are priced at the same monthly price. You then ask survey respondents which they prefer:
Hybrid at-home/on-site training
Enhanced on-site training and convenience
Through the data and insights you collect from this survey, you can answer the critical “which one” question that discrete choice modeling poses. Your members can provide valuable direction if you should invest in establishing a hybrid model or make more enhancements at your on-site locations.
There are several distinct advantages to using discrete choice models that should be considered as you determine your market research approach.
Like any analytical method, discrete choice has its limitations as well.
Once you have captured your data from a discrete choice survey, the depth of your analysis will depend on different factors based on the complexity of the question, or questions, you aim to answer.
The common thread is that the analysis that is done aims to draw inferences from the pattern of choices that groups of people make. While it is impossible to completely understand what lies behind a single individual’s buying decision, by analyzing your data you can identify trends and generate insights that can provide a clear direction of what most of your target market values. This lays the foundation for more informed decisions and actions that are not based on gut reactions, but rather statistically significant data.
So clearly discrete choice modeling has a wide range of potential benefits and applications for just about any business. It should be the go-to tool in your market research program when you’re faced with needing to answer the “which one” question. Through discrete choice modelling, your survey respondents can help provide some good answers to the hard questions.
Interested in learning more? Move forward with custom services developed specifically for your needs. Whether you need help with discrete choice modeling or general survey training, work with SurveyMonkey’s research experts to optimize your products or services and drive higher sales.
Collect market research data by sending your survey to a representative sample
Get help with your market research project by working with our expert research team
Test creative or product concepts using an automated approach to analysis and reporting