-The role of marketing research in making decisions -The steps in marketing research [SLIDE 1] Marketing research is the process of planning, collecting and analyzing data relevant to a marketing decision. The results are used to help management understand the wider market and the specific target markets. Market research is a key factor in understanding the market in order to create the appropriate marketing mix. Market research has three roles: -Descriptive: The gathering of factual information about the market. -Diagnostic: Explaining the data and its impact upon sales, what is working and what should be changed. -Predictive: Address “what if?” questions in order to plan for a changing market. [SLIDE 2] There are multiple purposes for marketing research. The four key uses are: -Improve the quality of decision making: Marketers can explore the desirability of multiple options based on the data. -Trace problems: Are initial assumptions correct? Has there been a market change? Is the brand being perceived as marketing planned? -Understand detailed and complicated relationships: As has been shown, both B2C and B2B marketing are complex. Clarification improves planning. -Serve both current and future customers accurately and efficiently: Optimizing market understanding improves both service and profits. The purposes are not mutually exclusive and can reinforce each other. Understanding a changing market can help a company trace sale problems and better serve the customer. One example is how Ford re-envisioned their ads for the Ford Mustang. It has almost always been positioned as an affordable muscle car, with the focus on power. However, research shows that exposure to ads at an early age builds loyalty, even with adult products. Ford’s market research showed Generation Z is more interested in social bonding and shared experience, so even before that cohort could drive, Ford began an ad campaign showing how a car could bring people together and, “just by chance”, showed a Mustang in every image of young people having fun. [SLIDE 3] The Phoenicians practiced market research as they travelled from port to port in the Mediterranean, and market research has probably been used far earlier than that. Markets keep changing and it is important to recognize and even plan for those changes. Technology is a key driver of that change. The internet continues to shift how markets make buying decisions. The Internet of Things (IoT) is something that is seen as an aspect of the Internet that will create even more changes over the next few years. IoT means connecting devices to devices, and that includes consumer devices such as watches, home smoke detectors, refrigerators and more. Multiple studies state large growth, with one predicting the IoT market to be worth $104.4 billion by 2023. The geographical differences mentioned in earlier lessons should also be remembered. A nationwide study found that consumers in the ten largest cities are more likely to self-identify as early adopters. There are also regional differences. For instance, customers in the Northeast were surveyed as being significantly more likely to buy IoT devices than in those in the Southeast. [SLIDE 4] With the importance of market research being clear, marketing must understand how to perform the process. The key marketing research problem is determining what information is needed and how that information can be obtained efficiently and effectively. The objective of market research should always be kept in mind; it is to provide insightful decision-making information. Management can then combine that information with personal experience and other information in order to make decisions. Regardless of the size of the market research program, the same general process should be followed: -Gather secondary data -Planning research design and gathering primary data -Specifying the sample procedures -Collecting the data -Analyzing the data -Preparing and presenting the report -Following up [SLIDE 5] Secondary data is that data that already exists, having been previously gathered for other purposes that can be leveraged for the current process. That can range from information in annual reports, to customer support data, to social media and other existing sources. It can be from inside the company or outside. However, quality can be a challenge with secondary data. Often the data does not come with the necessary information to understand the accuracy or the relevance of the data, and it often lacks the demographic information needed to accurately define the target market. Using secondary data is efficient, as the existence of the data saves time and money over research needed to collect additional information. Social media deserves a mention of its own. The vast amount of secondary data being created on social media is useful and also valuable. Facebook, Pinterest, Twitter, and other social media sights are working to monetize that data, charging companies to access it. To add further value, they have begun to incorporate other secondary data, such as that from Acxiom, Datalogix, and Alliance Data, alongside the data gathered on the sites. A term that many people might have heard about but not understood is that of “big data”. It is a term used to describe the exponential growth of data that has been caused by the Internet, and also is a place holder for the advanced tools necessary to analyze the data. Retailers rely on the data and tools to better identify their target markets in order to more efficiently spend marketing dollars. [SLIDE 6] Secondary data provides an initial basis for analysis. Marketers can understand what information exists and realize what is missing. That then helps define the research design, which describes research questions that need to be answered, how and when the data will be gathered, and how the data will be analyzed. Sometimes, those questions can be addressed by accessing more secondary data. When that is not possible, it is time to gather primary data: that data which is collected for the first time and which is used for solving the particular problem under investigation. Primary data can be gathered directly by marketing or can be gathered by outside contractors. The key is that it is gathered for the purpose of the current market analysis. Primary data can be expensive. For instance, a nationwide, 15-minute telephone interview with 1,000 adults can cost $50,000 or more. There are a few techniques which are regularly used: -Surveys -Observation -Ethnographic research -Virtual shopping -Experiments [SLIDE 7] The most popular market research technique is that of surveys, where a researcher interacts with people to obtain facts, opinions and attitudes. There are multiple types of surveys, including: In-Home Personal interviews providing high quality information through in-depth interviewing. Due to the costs, this valuable tool is being used less frequently. Mall Intercept Mall intercepts are conducted in common areas of shopping centers. The survey must be short as people are standing. When necessary, more detailed questions can be used to follow up. One disadvantage is the difficulty in getting a representative sample of a broad population. Technology is helping in that computer-assisted personal interviewing helps as the interviewer asks questions from a screen and inputs answers immediately. That has even been extended to computer-assisted self-interviewing, where the interviewer has the respondent directly interact with the computer. Telephone Telephone interviews are less expensive than personal interviews, but costs are rising as the large volume of unsolicited calls is causing people to refuse to participate. Interviews are typically run from a central-location telephone (CLT) facility, which often uses computer-assisted interviewing. U.S. federal “Do Not Call” lists do not apply to survey research. Mail Mail surveys are low cost, but they also have a low response rate. Email surveys are even lower cost and have an even lower response rate in B2C but are a bit more effective in B2B marketing. One drawback to mail marketing is that follow-up is usually not possible; the answers stand on their own. A mail panel is a way to recruit people, usually with cash or a prize, to actively participate in mail surveys over a specified period of time. Executive interviews Executive interviews are a survey tool similar to an in-home interview but used to interview executives and other personnel at their place of work. Setting up an executive interview can be difficult and delays are a frequent occurrence. That has caused this valuable survey technique to move online in order to save time for the executive and costs for the surveying company. Focus groups These are panels where people are brought to the same place to give feedback together. They are moderated and the technique is often used for brainstorming new products or services. [SLIDE 8] In order to ensure that all interviewed people are asked the same things, every survey requires a questionnaire. Three types of questions are used: -Open-ended : Encourages an answer in the respondent’s own words. -Close-ended: Asks a respondent to select an answer from a limited set of responses. -Scaled response: Asks respondent to measure the intensity of an answer using a numbers scale. Open-ended questions can bring more nuanced answers, and they take more time. Closed-ended make assumptions about the inclusiveness of the list, so an “other” option can be useful. With scaled response, it can be too vague without researcher direction. “What is the level of efficiency on a 1-5 scale” is confusing, while “are you 1) very satisfied, 2) somewhat satisfied, 3) neutral, 4) somewhat dissatisfied, or 5) very dissatisfied” give the respondent focus and direction. [SLIDE 9] As opposed to surveys, where the researcher asks about what people do, observation is watching what people do. The researcher watches the actions while recording information that can be compared across instances. One of the classic examples of observation is researchers watching children play with toys to see which toys are chosen. The researcher can be visible, interacting with the children and asking questions, or the researcher can be on the other side of a one-way mirror or a recording device. Mystery shoppers play an important role in B2C retail sales. Researchers posing as customers record information about stores and store personnel. Behavioral targeting (BT) is a form of observation that combines a consumer’s online activity with psychographic and demographic profiles compiled in databases. BT began with the storing of “cookies” on people’s computers to track their site activity in web browsers. Technology has added more complex observations such as social media monitoring, which uses automated tools to watch for key word trends and identify buzz. Newer technology is working to link individuals over their use of multiple devices. BT helps companies build up demographic information as the anonymized data can be aggregated to spot trends in different market groups. [SLIDE 10] Ethnographic research is the study of human behavior in its natural contest. It includes observation within the physical settings of decision making and use. Cambridge SoundWorks sells high end audio equipment including speakers. Men would come to the stores, seem ready to buy, and then not buy. They hired a research firm who followed the men around. The problem? Their spouses had problems with the look of the speakers and talked the men out of buying. A new line of speakers which looked like furniture resulted and has become their fastest growing product line. [SLIDE 11] Virtual shopping is using computers to simulate a retail experience. Doing so allows the researcher to track a consumer’s trip through the store, identifying what items interest the customer, how long is spent in each area, recording what items are purchased, and tracking the shopping experience more fully. For instance, a retail apparel company used virtual shopping to note that many men had problems putting together outfits. The company began selling items together to make the outfits more obvious and made other related changes. The result was that men’s clothing sales increased by forty percent. The use of virtual shopping has been shown to be so effective that it has begun to move past retail. Aviation, automotive, financial and other industries have begun to use this as a tool to improve customer support. [SLIDE 12] Experiments are real world tests of the marketing mix. A researcher alters one or more variables while observing the effects of those alterations on another variable. The best way to experiment is to hold all but one variable constant to focus on the effect of that variable. Of course, that is difficult for some products. Weather, economic conditions, and other factors outside corporate controls can vary and sometimes blur the impact of a change. Usually companies experiment with price or size of the product, seeing how volume of sales, revenue, and profit change based on the changes. Mars, a candy company, was seeing falling sales. Other research showed that shrinking candy bar size would lose value in the eyes of the customer. They experimented with a marketing mix that remained the same in each market except for the varying size of the candy bar. The experiment found that a larger bar would increase sales enough to increase revenue even with a smaller profit margin. [SLIDE 13] The universe is defined as the full population of the target market. As it is not possible so survey the entire population, a subset must be chosen. This is the sample. The question of whether or not the sample is representative of the universe depends on sampling technique. The two key categories are: -Probability sample: Every element in the population has a known statistical likelihood of being selected. -Nonprobability sample: Little or no attempt is made to get a representative cross section of the population One type of probability sample is the random sample. In a large enough sample, every element of the population has an equal chance of being selected as part of the sample. A typical method of nonprobability sampling is the convenience sample, interviewing friends or coworkers is an example of this type of sampling. Nonprobability samples are less expensive to run and can be useful as long as the researcher remains aware of the nonrepresentative nature of the sample. There are many ways that errors can creep into a sample, but there are four key errors marketers should know: -Measurement: When there is a difference between sample feedback and external information. For instance, people may say they like a product when they do not because they think the answer is expected. -Sampling: When the sample does not reflect the target population. -Frame error: A specific type of sampling error where the sample does not represent the target population -Random error: When the sample is an imperfect representation of the overall population. The difference between frame and random error is that a frame error misses a specific segment through lack of completeness of the model, while random errors are variations from the sample that happen through the laws of probability. [SLIDE 14] While a marketing organization can do its own market research, that work can often be more efficiently done by an outside organization which specializes in research. A field service firm is one that specializes in interviewing respondents on a contract basis. Because of the larger target populations in B2C markets, those companies are more likely to use a field service firm. For many more narrow research types, B2B firms can do their own research, only moving to field service firms when the numbers of respondents is large. [SLIDE 15] There is no purpose to collecting data unless you can turn the data into information with which to draw conclusions. There are three primary methods of analyzing research data: -One-way frequency counts: Simply total the different responses to each question -Cross-tabulations: Look at relationships between answers to different questions -Statistical analysis: Use far more complex techniques to find insights from the data One-way frequency counts provide an initial foundation for analysis. “What percentage of people like product X?” Cross-tabulation begins to link answers “What percentage of people who like product X buy product Y?” Statistical analysis provides complex insights but is beyond the scope of this course. [SLIDE 16] Once the analysis is complete, the researchers must communicate the insight to management. That means the research must be appropriately marketed. Both written and oral reports must be prepared. While the basic written report is often just the detailed information on the questions, researchers must be prepared to speak to management about the results, explaining the research methodology and the resulting information. Most people in marketing will be research consumers and not research suppliers. At the same time, those consumers must have a basic understanding of the research in order to put the research information in context of other information, and to make decisions based on insight. [SLIDE 17] Handing off the reports is not the end. If a researcher does not get feedback, then the process cannot be improved. Knowing how the decision makers used the data and why they ended up making specific decisions can improve later research, change the communications used to present the research results, and otherwise create a cycle that helps marketing have more accurate information and create better marketing mixes.