There are several ways to categorize the various market research methods. The vast majority of techniques fit into one of six categories: (1) secondary research, (2) surveys, (3) focus groups, (4) interviews, (5) observation, or (6) experiments/field trials.
- Text messaging surveys and voting (SMS Surveys) – One good example of this is a company called “Poll Everywhere.” They allow seminar attendees to vote and respond to poll questions via SMS (text messaging).
- Smartphone designed surveys – Good mobile surveys are ones designed specifically for the smartphone form factor. There are many companies working on this, such as iOpinionSurveys and OpinionMeter. These surveys can be web-based, optimized for phones, or they can be native applications built specifically for iOS, Android, or Windows mobile operating systems.
- Location Awareness – Advanced phone market research techniques can leverage smartphone location (GPS) information to trigger questions or simply track movement over time. For example, you can imagine a survey question that only appears when the phone knows the user is at the gas station.
- Mobile Ethnography – Using information like location awareness, researchers are able to gather rich contextual data (using mobile phones) about behaviors, allowing them to really understand the habits and lifestyles of subjects.
- Research of social media — Simply researching the market of social media. For example, “X% of people use Facebook and the average age of a Google+ user is X.”
- Research using social media data — There is a lot of data that can be gleaned from social media sites. Looking at how many times a certain news story or product is shared across sites can tell researchers a lot about what works and doesn’t work in journalism, product concepts, etc. “Listening” to social media is like eavesdropping on a million conversations and can be a great place to pick up on trends.
- Research using social media as part of the methodology or delivery mechanism — Many companies have a large following on social media sites and can leverage that audience to ask questions. Often, if a customer is willing to follow/friend/subscribe/whatever to a company on a social media site, they are a big fan of that company and one of the best customers (probably a “promoter” in NPS, or net promoter score language). What a gold mine for companies to have instant access to their highly loyal and interested customers for market research purposes. A good example of using social media to conduct market research is GoPollGo, a twitter polling company.
Computer-driven 3D virtual shop research is becoming increasingly common amongst retailers and consumer product manufacturers. Done properly, the tests can deliver a highly accurate representation of behaviour when consumers are faced with products on the shelf, as they would be in a real-life scenario. For researchers, the technique offers the chance to get results more quickly and at a much lower cost than real store field tests.
What is 3D Virtual Shop Research?
3D virtual shop research, also called virtual store research, involves a virtual reality environment that mimics a real store. A research subject is taken through the virtual environment to test and measure an array of research topics, such as shelf layouts, signage variations, packaging research, brand-blocking strategies, point-of-sale materials, and price changes. The goal is to simulate a life like experience that can be used to run research experiments on subjects without the cost and logistics of a physical environment.
While it’s true that the startup cost for building a simulation (especially on the more elaborate end) as well as establishing the internal capabilities needed to work with it may be significant, virtual store simulators offer a wide variety of benefits to the market researcher and once a company has created their ‘virtual aisle,’ they can reuse it in future projects for a much lower cost. The quality of the research is generally very high; participants answer questions based on realistic shelf displays instead of fake-looking cobbled together displays or hypothetically survey questions.
Applications of 3D Virtual Shop Research
Where can 3D virtual shops be used? In one case, the researchers at Play MR recreated an entire aisle from a large format supermarket (with more than 500 products) for a client selling snack foods. They have released a case study video showing the power of this kind of simulation.
3D research need not always be as complex and involved, of course. It can even be used for online surveys. This works by creating a ‘shelf set’ which is shown to the respondent after a ‘fly in’ video that helps recreate the experience of visiting the store and going to a particular aisle. There may be sounds, other shoppers, and signage adding realism.
Once they’ve arrived at the shelf, the respondent can take products from the shelf, examine them more closely, zoom in on side and back panels, and place them in their shopping cart before heading to the checkout stand.
From the perspective of the researcher, this answers valuable questions like:
– Which products drew their attention initially, and how long did they consider them?
– Which products did they pick off the shelf?
– How long did it take them to read the label and make a purchase decision?
Using these basic data points marketers can then evaluate variables such as the ‘shelf impact’ of a given product, how shelf positioning affects market share, what the optimal price is to generate maximum profit or highest market share, or even what the estimated sales volume will be of a given package.
When to use 3D research
3D virtual shopping research can be used at nearly any point of new product development, from initial brainstorming to market evaluation, product design, prototyping, testing, and finally the launch. Especially if it is implemented early on, it can significantly improve the success of a launch.
It’s true that in-store tests remain the gold standard for gathering data on consumer behaviour, and it’s unlikely to lose that position any time soon. Nevertheless in-store tests are time consuming and expensive, and store owners and shoppers alike frequently complain they are intrusive.
With virtual shop research it’s possible to conduct tests early and often in a confidential environment where competitors have no way to learn of new designs. The technology can test multiple scenarios and ideas at the same time, allowing researchers to make changes ‘on the go’ and without the lead times of creating actual new product packaging.
Virtual simulations also give researchers far more control. There is no concern about the weather, competitive activity, out of stock situations, or the many other variables that affect real-world tests. Without the uncontrolled variables, retailer reticence, and expenses of physical tests, simulations can provide data that is nearly impossible to get otherwise.
Finally, 3D virtual shop research can be a powerful tool when negotiating with retailers. Pre-testing in a virtual environment allows the company to show their retailers how the new products will succeed, giving them leverage to negotiate greater space on the shelves. Key to obtaining reliable data in this regard is a quality simulation — rather than using fake or dated looking shelves, the simulation must be realistic to the customer and the researcher must be able to show the collected data actually corresponds with in-store sales.
Alex Pejak is an economist currently working on a few projects in Australia. She is interested in topics related to market research and project management.
You’ve got your data, you’ve made some sense of it, and now it is time to communicate your results. Great! This article will provide examples of many types of charts and graphs and explain how to pick the best one for your data depending on the message you want to convey.
Choosing a type of chart depends first and foremost on what kind of data you have and what you want to express. I find that charts and graphs are typically used to convey one of the following: comparisons/relationships, distribution, trends, composition, flow/process, or location.
1. Comparison/Relationship Charts – Pretty self explanatory, right? You have data on two or more variables and you want to show them together, probably to show a correlation or pattern of some type. Examples might include MPG of three different cars, average heights according to race, etc. Bar charts and line charts, or combinations of the two, are very commonly used for the purpose of comparison.
Venn diagrams are especially useful to show relationships. These are typically qualitative in nature.
2. Distribution Charts – These types of charts aim to convey “what is the distribution?” of my data. For example, let’s say you did a survey and you asked everyone their age. A distribution chart would be usefeul to visualize the distribution of ages among respondents. Column and Line Histogram charts are probably the most common forms of distribution charts. Scatter plot charts are also great for this purpose.
Word clouds are an interesting way to visualize the frequency distribution of words with textual data. Here’s an example of a word cloud from the 2011 Academy Award acceptance speeches. While these aren’t the solution to the world’s problems, they can be useful from time to time in quickly analyzing open-ended comments from surveys.
3. Charts that show Trends – While the chart categories mentioned above can certainly show trends, I think it is deserving to identify this as a category of its own. The most common way to show trends over time is with a line chart. Nowhere is this more common than in showing stock price trends over time. The chart below is a “candlestick” chart.
Here is a good article if you want to learn more about stock market charts.
If you have a few extra minutes, here’s an incredible video of Hans Rosling showing charts in motion, demonstrating both relationships and trends all at once.
4. Composition Charts – The next category of chart types is “composition” charts, which attempt to show viewers “this is how my data is composed.” By far, the most common “composition” chart is a pie chart. A pie chart might show that 60% of my survey respondents were composed of women and 40% were men.
I like using doughnut charts as a variation on the all-too-common pie chart.
5. Flow/Process Charts – Flow charts are used to show–you guessed it–the flow of a process. These are often used to guide a decision. In fact, they are often called “decision trees.” Here’s a couple of examples:
6. Location Charts (Maps) – Geographical maps and data overlays on maps cannot be left off this list. Of course, everyone knows about basic maps, but take a look at some of these map-inspired charts and graphs:
Advanced Charts and Data Visualizations
Most of the charts and graphs shown above are pretty traditional. Data visualization is a hot field and there are many new cool forms of “charts” emerging. Calling them “charts” doesn’t do them justice. I’ll dedicate more time to this in another article, but here’s a sample of what I’m talking about.
That’s about it. I’ve leave you with some links to additional information.
Additional Links and Information
The best guide for choosing a chart that I’ve found to date is this one. They’ve narrowed down the categories of charts down to four.
“Juice Labs” has a nice tool they call the “chart chooser.” It allows you to filter based on your needs.
Finally, here is a nice infographic from online-behavior.com (snipit below) explaining a wide variety of chart types.
I hope this article was informative and that you have a better understanding of the types of charts and graphs out there in the world. If you have any additions or comments, please chime in below.
While traditional market research techniques such as surveys and focus groups are still widely used, there are many new market research methods and techniques to spice things up. As technology and socioeconomic trends change, so will our means of gaining customer insights. As you’ll notice, many of these are really just new technologies applied to traditional methods, as opposed to radically different methodologies. In any case, here are a sampling of some of the new market research trends and techniques popular now, in no particular order:
1. A shift from data collection to data analysis: Today, actual customer behavior data is collected with ease, to the point where analysis (or data mining) is much more challenging than obtaining the data. For example, Google Analytics provides webmasters with tons of information about website visitors, including languages, pages visited, screen resolutions, etc. All of this information can be used to fine tune a website to the audience. Another example of “big data” data mining of is Amazon’s predictive recommendations. By carefully monitoring the products a user purchases/views and correlating that information with purchase histories of others, Amazon is able to very effectively present product recommendations. All of this is done through data mining, without having to ask the user “what other products might you like?, which would be crazy.” Twitter is another great source of readily available data that can be mined (text analytics). Jonathan Harris performed a great TED talk that beautifully demonstrates how readily available data can be visualized.
2. A shift from “how do think you will behave?” (self-reporting) to “I know how you behaved” (observational research): If you wanted to know what color cereal box would sell the most cereal, would you rather base you decision on a survey or an actual experiment where colors are tested? Of course the experiment would be more valuable. I want to know what customers actually do/want, not what they think they do/want. It’s not that customers are trying to deceive researchers; it’s just that it’s difficult for users to predict their own future actions. In any case, the world of market research is shifting from self-reporting techniques (surveys, focus groups), to observational research methods whenever possible. The data is much more reliable.
3. Mobile market research methods: Smart phones and tablets are taking the world by storm. These devices are becoming a preferred platform for many applications and markets, including market research. Examples of how these devices are being used in market research include:
4. Biometric Market Research Techniques: New biometric research methods that measure a subject’s physical response to stimuli (e.g., television commercial) provide valuable data that a subject might not be able or willing to express verbally. Examples of biometric market research methods include heart rate monitoring, respiration monitoring, skin and muscle activity, brain activity (using functional MRI) and eye tracking. A good article on the subject can be found here. Campbell Soup has used such methods in their market research.
5. Prediction Markets: A prediction market is like a mini stock market, where a group of people can buy and sell “predictions” of various events. For example, one event might be “who will win the presidency?” Participants could use their “currency” (fake or real) to buy or sell whoever they think will win. Early on, the price of one candidate or the other might be $0.50, but as the election probability becomes more certain, a bid on one candidate will grow closer to $1.00. At the end of an election, one candiate will be worth $1.00 and the other $0.00. Participants can buy and sell their stake in a candidate along the way. The beauty of these prediction markets is that they tend to be good indications of reality. So what does this have to do with market research? Well, forward thinking companies are setting up these prediction markets to tap into the wisdom of their employees. For example, a company could ask employees to bid on a prediction market that has to do with competitors, industry trends, or the success of product concepts in order to get an early read on those ideas. If this is still foggy, check out intrade, a public prediction market. Consensus Point makes business to business software that has been used by companies like Best Buy.
6. Virtual Shopping: This involves the use of virtual store simulation to mimic a shopping experience for participants–a good way to test things retail issues like product placement, store layout, packaging, etc. Once again, the idea is to replicate a real situation for research subjects and observe behavior, as opposed to asking them what they think they will do.
7. Live Audience Response: In conferences or lectures, presenters often have difficulty engaging with the audience. One tool to remedy this problem is live audience response systems. These systems involve a handhold remote control for audience members to respond to questions that appear on-screen (usually in a PowerPoint slide). You can imaging the applications for this: professors doing on-the-fly quizzes to see if students understand the concepts, presenters asking demographic questions to better understand their audience, polling, etc.
7. Online Collaboration Tools: Tools like Skype (video calling), instant messaging, and shared whiteboarding allow researchers to conduct a variety of “traditional” market research techniques using new technology. These technologies are often much cheaper than physically gathering people. They also allow researchers to gather people from broader geographies much easier.
8. Social Media Market Research: Social media dominates the Web, so it is natural that market researchers are looking for ways to leverage this technology. When people say “social media market research” they might mean several different things:
9. QR Code Surveys: This overlaps with mobile phone market research. A poster could ask a simple survey question and provide two QR codes, asking people to scan their choice. Such an approach makes it very easy for someone to take a one-question survey without doing much more than pointing a phone. A webmaster would then be able to gather the response data in aggregate. Other companies are using QR codes as a simple launch point to a mobile survey. A good example of this is Tiipz.
There you have it–an overview of new market research methods and techniques. This article will continue to evolve and update over time as new research methodologies and technologies emerge. I hope this was informative. If you have other examples of new market research, or if you have anything to add, please do so in the comments below.
When it comes to experiments and data analysis, there are two main types of variables: dependent variables and independent variables. It’s easy to get these mixed up, but the difference between dependent and independent variables is simple. Here is a quick and easy definition of each one, along with some examples.
Dependent Variable: This is the output variable you are really interested in monitoring to see if it was affected or not. It can also be called the “measured variable,” the “responding variable,” the “explained variable,” etc. I think it is easy to remember this one because it is dependent on the other variables.
Independent Variables: These are the individual variables that you believe may have an effect on the dependent variable. They are sometimes called “explanatory variables,” “manipulated variables,” or “controlled variables.”
If this is not already clear, it will be after a couple of examples.
There are four measurement scales (or types of data): nominal, ordinal, interval and ratio. These are simply ways to categorize different types of variables. This topic is usually discussed in the context of academic teaching and less often in the “real world.” If you are brushing up on this concept for a statistics test, thank a psychologist researcher named Stanley Stevens for coming up with these terms. These four measurement scales (nominal, ordinal, interval, and ratio) are best understood with example, as you’ll see below.
Let’s start with the easiest one to understand. Nominal scales are used for labeling variables, without any quantitative value. “Nominal” scales could simply be called “labels.” Here are some examples, below. Notice that all of these scales are mutually exclusive (no overlap) and none of them have any numerical significance. A good way to remember all of this is that “nominal” sounds a lot like “name” and nominal scales are kind of like “names” or labels.
Note: a sub-type of nominal scale with only two categories (e.g. male/female) is called “dichotomous.” If you are a student, you can use that to impress your teacher.